university of groningen fostering the learning potential ... · contents chapter 1 chapter 2...
TRANSCRIPT
University of Groningen
Fostering the learning potential of at-risk students in the classroomTiekstra, Marlous
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.
Document VersionPublisher's PDF, also known as Version of record
Publication date:2016
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):Tiekstra, M. (2016). Fostering the learning potential of at-risk students in the classroom: Studies into theconsequential validity of dynamic assessment. University of Groningen.
CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.
Download date: 27-03-2021
Fostering the learning potential of at-risk students in the classroomStudies into the consequential validity of dynamic assessment
Marlous Tiekstra
© 2016, Marlous Tiekstra
All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system of any nature or transmitted in any form or by any means, electronic, mechanical,
photocopying, recording or otherwise, without prior permission of the author, or when
appropriate, the publishers of the papers.
ISBN 978-90-367-8577-8
ISBN E-version 978-90-367-8576-1
NUR-code 848
Cover photo photo by Marlous Tiekstra, vignobles de Belvèze-du-Razès
Graphic design myra nijman concept & design | [email protected]
Printed by CPI - Koninklijke Wöhrmann
Fostering the learning potential of at-risk students in the classroomStudies into the consequential validity of dynamic assessment
Proefschrift
ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen
op gezag van de rector magnificus prof. dr. E. Sterken
en volgens besluit van het College voor Promoties.
De openbare verdediging zal plaatsvinden op
donderdag 11 februari 2016 om 14.30 uur
door
Marlous Tiekstra
geboren op 6 september 1985te Delft
Promotores Prof. dr. A.E.M.G. Minnaert Prof. dr. M.G.P. Hessels Beoordelingscommissie Prof. dr. A.J.J.M. Ruijssenaars Prof. dr. W.C.M. Resing Prof. dr. K.H. Wiedl
To
Gjalt Christiaan Tiekstra
a very wise and honourable man who is deeply missed
thank you for your inspiration
Contents
Chapter 1
Chapter 2
Chapter 3
Chapter 4
Chapter 5
Chapter 6
Chapter 7
References
Samenvatting [summary in Dutch]
Dankwoord [acknowledgements in Dutch]
About the author
09
25
57
81
99
119
141
155
171
181
187
General introduction
A review scrutinising the consequential validity of
dynamic assessment
Can didactic resistance be measured by using the
Response-to-Instruction model? A pilot study
At-risk students and the role of implicit theories of
intelligence in educational professionals’ actions
Voices from practice: when is the gap between diagnosis
and intervention apparent?
The next step in bridging the gap: a pilot study to
enhance teacher’s skills with at-risk children
General discussion
“On trace la route,on s’regarde pasOn change de route sans savoir pourquoi”(C. Maé)
Chapter 1General Introduction
This chapter is partly based on: Tiekstra, M., Hessels, M. G. P., & Minnaert, A. E. M. G. (2009).
Learning capacity in adolescents with mild intellectual disabilities. Psychological Reports, 105, 804-814. doi:10.2466/PR0.105.3.804-814
CHAPTER 1
10
GENERAL INTRODUCTION
11
1
Yamina, a ten year-old girl of Moroccan origin, is in third grade with 27 classmates. Lately, Yamina
struggles to keep up with her peers. In the classroom she appears easily distracted and during breaks she
is alone more often. Her school results are poor since the start of the academic year already, especially
in reading. Since five months Yamina is supported by the remedial teacher. Twice a week she receives
additional reading time outside the classroom. Due to suspicion of dyslexia she has been recently assessed
by a school psychologist. Results of this assessment revealed an IQ of 96, and low scores on both
reading and arithmetic achievement tests. Yamina was not identified as dyslectic. The remedial teacher
noticed that Yamina grasps some parts of his instruction very well, especially in arithmetic, but much
less in other domains. Meanwhile, the regular classroom teacher does not know how to support Yamina
with her reading. He and the remedial teacher decide to continue the general didactical instruction as
modelled in the class’s reading program, since it worked for other students previously. Besides, they decide
that Yamina should receive more time to finish her tasks. As a result, she often stays inside the classroom
during breaks to finish her tasks.
This is an example of daily educational practice. The questions that arise are: Is this
the right way to remediate Yamina’s problems? How to develop the student’s learning
potential? What does the general classroom teacher think about the learning potential
of this student? How does this relate to the general classroom instruction? How should
‘care’ around this student be effectuated? The present thesis is grounded in these
questions.
The rationale of this thesis is that every student can learn. Some students learn
easier than others, but every student has potential to learn. To optimise capacities
in learning every student, either a high or low achiever, needs assistance in learning
from a more experienced significant other. This is underlined in literature, as well as
in the case of Yamina: it is recognised that she needs assistance from the remedial
teacher. Vygotksy (1978) emphasised the importance of this significant other, or so-
called mediator in the terminology of Feuerstein, Rand, and Hoffman (1979), in the
learning processes of students. This significant other (mediator), for example a teacher
or a parent, both organises information by filtering or structuring, and gives meaning to
CHAPTER 1
12
the new information so that it can be integrated in the learning processes of students.
Many other researchers underlined the importance of significant others to the learning
outcomes of students. The quality of teacher-student interpersonal relationships has
been shown to substantially contribute to motivational, and consequently, learning
outcomes, especially in at-risk populations (see e.g., Den Brok, Van Tartwijk, Wubbels,
& Veldman, 2010).
Learning occurs when the mediator operates in the Zone of Proximal Development
(ZPD) of a particular student. This ZPD, the difference between the current independent
achievement of the student and the performances a student may show when tailored
help is provided (see also Lidz, 1995), is an important aspect in the optimisation of
learning potential. Consequently, Vygotsky’s conceptualisation of the ZPD has several
implications for psycho-educational practice, such as assessment, and the diagnosis of
learning disabilities, next to intervention, and general classroom instruction.
Assessment and diagnosisIn the case of ongoing difficulties in learning, as in the case of Yamina, students are often
subjected to an assessment procedure to investigate the causes of their difficulties. Such
an assessment can be effectuated in different ways and for different purposes, e.g., it can
be formative or summative, focused on general abilities or specific domains, and it can
be either for classification purposes or rather focused on the special educational needs
of the student. Moreover, for the diagnosis of a specific learning disorder an estimation
of the student’s intelligence is often requested (see e.g., the DSM-5). Assessment is,
however, subject to an ongoing debate about the test administration itself: it can be
administered either statically, or dynamically.
Static assessment. ‘Static’ refers to the fact that only scant interaction between the
examiner and the examinee during testing is allowed: the examiner asks the questions
following the standardised protocol and the child has to respond. No additional
information, explanation, or training is provided, although sometimes a question may
GENERAL INTRODUCTION
13
1
be repeated or the child may be asked to “say a bit more”. The underlying assumption
with static tests is that in order to obtain reliable test scores, the psychological trait
that is to be measured must be stable over time and, therefore, no training or teaching
can be established during testing (Hessels-Schlatter & Hessels, 2009). The results of
the assessment will be the product of previous learning, but will not reveal the child’s
capacity to learn new things. Also, no information about learning processes is provided
by such tests, although some static intelligence tests do try to focus on underlying
neuropsychological processes (Daniel, 1997). The examiner, however, still has a relatively
passive role, since there is no explicit teaching incorporated in these tests.
Dynamic assessment. As defined by Thorndike (1924) intelligence is the ability to
learn. Therefore, in dynamic assessment, examinees are offered an opportunity to learn
during testing. This opportunity to learn, referred to as learning phase, is a moment
during test administration in which the examiner provides hints and feedback that
address processes needed to solve the task in the test. After the learning phase, items
are administered to measure the extent to which the examinee has profited from the
learning phase. Outcomes, then, will give an indication of the examinee’s learning
potential. Since a learning phase is part of the administration of the test, this is the
main difference to static testing, in which only products of former learning experiences
can be measured.
In a previous study (Tiekstra, Hessels, & Minnaert, 2009), we investigated the differences
between static and dynamic assessment of intelligence in a specific population. In static
tests it is implicitly assumed that all children have had the same opportunities to learn
and have had the same learning experiences, and that differences in test scores reflect
differences in children’s capacity to profit from their learning experiences. However,
since learning experiences vary from one child to the other, many other authors argue
that intelligence tests only deliver information about products of former learning
opportunities, but do not provide any information about current learning opportunities.
With regard to specific populations, such as people with intellectual deficiencies, the
CHAPTER 1
14
use of these tests might lead to a serious underestimation of their learning capacities
(Hessels & Tiekstra, 2010). Moreover, these tests do not specify which cognitive processes
have been employed and do not give useful information for the development of tailored
cognitive interventions (Bosma & Resing, 2006).
The results of our study in adolescents with mild intellectual disabilities showed that an
IQ test predicted 8,8% of the variance in a novel school related learning task, whereas
the dynamic test of learning potential improved this prediction to 48%. This added
value was highly significant (ΔR2=.40; ΔF1,20 =15.28, p=.001), with effect sizes for IQ
η2 = .24 and for the learning potential test η2 = .43. The learning potential test, thus,
provided a more accurate prediction of future learning outcomes compared to an IQ
measure. More specifically, whereas static tests evaluated these students’ capacities in the
same range (i.e., low IQ), the learning potential test showed the ability to differentiate
substantially between students’ capacities to learn (Tiekstra, et al., 2009).
InterventionDynamic assessment is scarcely implemented in educational practice (Deutsch & Reynolds,
2000; Elliott, 2003). While being studied extensively for more than fifty years, dynamic tests,
still, come off worst compared to traditional intelligence tests with regard to their application
in practice (Grigorenko & Sternberg, 1998). IQ tests receive great attention and appreciation
due to their designated necessity for diagnosing learning disabilities, and subsequently, for
financial resources for children with special educational needs. Consequently, the outcomes
of these tests are used for setting up Individualised Educational Plans (IEP). However, IQ
test outcomes lack specific information that could be used to adapt instruction (Freeman &
Miller, 2001; Ysseldyke, 2001). In the present case, Yamina gets support from the remedial
teacher, but it might not be the support she actually needs to improve her reading skills.
The more tailored an intervention is, the more effect it will have. The outcomes of the
assessment procedure revealed that Yamina had an IQ of 96 and low scores on school
achievement tests. This information did not provide useful clues for the construction of
an intervention or for an adaptation of the present classroom instruction. Therefore, the
GENERAL INTRODUCTION
15
1
teacher and remedial teacher had difficulty in providing the appropriate help to Yamina.
This is common for educational practice: actions in the classroom are often not based on the
IEP of students. A gap remains between diagnosis and actions in- or outside the classroom
(Shapiro & Kratochwill, 2000).
In the new edition of the Diagnostic and Statistical Manual of Mental Disorders (5th ed.;
DSM-5; American Psychiatric Association, 2013), used by practitioners as a guideline for
diagnosis of disabilities, it is highlighted that ‘targeted intervention’ is the ultimate aim of an
accurate diagnosis. Since dynamic assessment has proved to be a better estimator of future
learning outcomes (e.g., Elleman, Compton, Fuchs, Fuchs, & Bouton, 2011; Swanson,
2011; Swanson & Howard, 2005), especially in populations in which learning problems are
eminent (e.g., Beckmann, 2006; Tiekstra et al., 2009), and could provide useful information
for classroom practice (e.g., Bosma & Resing, 2006; Daniel, 1997), one should wonder why
these tests are not used in educational practice. A decade ago, Elliott (2003) provided several
reasons for this, such as lengthy test administrations and the focus on rather small ranges
of age groups. Meanwhile, attempts have been made to overcome these issues, but these
have not yet led to more use of dynamic assessment in educational practice to date. Elliott
(2003) underlined the need for studies that investigate the potential of dynamic assessment
to bridge the gap between diagnosis and interventions. This can be considered a prerequisite
for applicability in educational practice.
Although information from dynamic test reports is appreciated by teachers (Bosma,
Hessels, & Resing, 2012; Freeman & Miller, 2001), the question remains to what extent
principles from these reports are translated into the classroom practice. Here, too, the
gap between assessment information and classroom practices remains present (Haywood,
2012). Therefore, from our point of view, more aspects play a role when providing students
with targeted intervention: next to adequate (dynamic) test results, skills of educational
professionals need to be developed or improved in order to fully optimise students’ learning
potential in educational practice.
CHAPTER 1
16
General classroom instructionAlready underlined in 1987 by Delclos, Burns, and Kulewicz, dynamic assessment
positively influences the view of teachers about a child’s ability due to the teaching
component (i.e., the learning phase) during testing. This is especially important
in the case of at-risk students. At-risk students are students that are susceptible for
developing learning delays and/or problems due to their risk for non-optimal learning
opportunities and experiences. Generally, this population is composed of immigrant
children, children that have been brought up in low-income families or families with
low socioeconomic status, children with severe learning difficulties, communication
disorders, and/or behavioural or emotional problems, or children with intellectual
disabilities. These children neither had equal learning opportunities, nor the same
learning experiences as average indigenous, middle class children without any learning
difficulties. At-risk students are often confronted with lower educational expectations,
and consequently, lower educational efforts of educators. Eventually this will lead to
a self-fulfilling prophecy: the student is not inclined to learn (Haywood, Tzuriel, &
Vaught, 1992; Sternberg, 2004).
General classroom instruction is oriented towards average students, as illustrated by
the case of Yamina: she received no specific adapted instructions, since the method
used worked well in other students. However, it is questionable whether this is the best
solution for the specific problems of Yamina. Actually, it relates to a bigger problem:
how to differentiate general classroom instruction in such way that each individual
student benefits from it?
To resolve this issue, the Response-to-Instruction/Intervention, or so-called RTI-model,
has been developed in the United States. This organisational model requires regular
progress monitoring of each student in the classroom, which in turn allows for regular
adaptation of teacher’s instruction. The core assumption of the RTI model is that every
child reacts differently to whole-class teaching. The extent to which each student is
sensitive to instruction can be estimated by regular progress monitoring of each student.
GENERAL INTRODUCTION
17
1
Progress monitoring is carried out regularly (daily, weekly, or fortnightly) with short
curriculum-related tests, which can be developed by the teacher. The model consists of
three levels of instruction-intensity (Fuchs & Fuchs, 2006). After approximately six weeks
of general classroom instruction (tier one), progress monitoring shows which students
did not progress (compared to the mean progress of the group). These students, then,
will be offered intensified instruction in small groups several times a week (tier two). If
extra instruction does not provoke enough progress, as demonstrated by the regular
progress monitoring, an individualised route is set up for these students (tier three).
This model is not (yet) implemented in European countries, whereas in the United
States it is even used for the diagnosis of learning disabilities. Several authors (e.g.,
Kavale, 2005; Reynolds & Shaywitz, 2009), however, criticised the model since its
effectiveness has not yet been demonstrated empirically. Interestingly, in theory, the
model acknowledges the role of the teacher in providing quality teaching. Therefore, it
could be used in the operationalisation of the concept of ‘didactic resistance’, a criterion
used for the diagnosis of learning disabilities. The concept refers to the responsiveness
of the student to teaching. Thus, the quality and amount of teaching should be carefully
documented. Due to the progress monitoring the teacher becomes aware of the effects of
his/her own teaching. Moreover, the model accounts for an opportunity to differentiate
between different needs of students. In the general classroom, students with IEPs could
receive as much quality teaching as regular students, if the model is well-implemented.
Significant othersThe theory of learning by Vygotsky, and his conceptualisation of the ZPD in particular,
underlined the importance of significant others. As mentioned before, significant others
play an important role in students’ learning processes: exploration of the student’s ZPD,
quality instruction in the ZPD, and motivational techniques combined can be expected
to lead to the student’s responsiveness to instruction. In the case of at-risk children this
implies the following:
CHAPTER 1
18
1. The assessment of capacities should be done dynamically to explore the
child’s ZPD;
2. The activities of the assessor should be highlighted during test interpretation;
3. The role of the educator should be highlighted in diagnosing disabilities (i.e.
didactic resistance);
4. The roles of both the educator and assessor are critical when aiming at
tailored instruction and targeted interventions.
Significant others, thus, play an important role in the ‘care’ processes concerning at-risk
children when optimising their learning potential. Therefore, the focus of this thesis will
be on these significant others in the educational context, involving teachers and support
professionals, such as remedial teachers and school psychologists.
As demonstrated by many (e.g., Azjen, 2005; Pajares, 1992), each of these significant
others have their own beliefs about learning, teaching, their self, and these beliefs affect
people’s actions. This means that the actions of significant others in the issues mentioned
above are influenced by specific beliefs. Especially in the case of at-risk children this is
very relevant. With respect to Yamina, the teacher might have lowered his educational
expectations about her due to the outcomes of the assessment, leading to less effort in
teaching Yamina, since the results of the IQ test confirmed his expectations about her
intelligence. Why, then, put the effort in providing her with different instruction from
the rest of her classmates? This might be in contrast with the beliefs of the remedial
teacher who might think that Yamina is able to learn new things, since he noticed
progress in Yamina’s response to his individualised instruction.
This example illustrates the importance of one’s implicit theory of intelligence. Every
one of us has an idea about the malleability of intelligence, nonetheless, we are often
unaware of this idea. As demonstrated by Dweck and colleagues (see e.g., Dweck, Chiu,
& Hong, 1995) one’s implicit theory of intelligence greatly influences one’s actions and
own learning. Especially in the case of at-risk students, implicit theories of intelligence
GENERAL INTRODUCTION
19
1
could play a substantial role. Moreover, the self-efficacy of educators, i.e., their beliefs
about the influence they have on their students’ learning (Bandura, 1997; Gibbs &
Powell, 2012; Imants & De Brabander, 1996) has an impact on teacher’s behaviour
in the classroom (Tschannen-Moran & Woolfolk Hoy, 2001). It is for these reasons,
and in line with the concluding remarks of Ghesquière (2002) in the book on learning
potential assessment by Van der Aalsvoort, Resing, and Ruijssenaars (2002), that in the
present thesis we consider the influence of beliefs of educational professionals, being
the significant others in the learning processes of at-risk students, of importance.
Consequential validityIn the case of Yamina, the static assessment did not provide any useful clues for specific
intervention. As a result, the teacher as well as the remedial teacher, did not know which
instruction or intervention to provide, and therefore, continued to follow the proposed
instruction as provided by the conventional reading method. Moreover, we could state
that the school psychologist might have provided more support to the teacher and
remedial teacher. The gap between diagnosis and intervention is very visible. The focus
of this thesis is to provide information about bridging this gap.
Actions should be based upon outcomes of assessment. Consequences of, or actions
that follow the assessment procedure fall within the concept of consequential validity
of assessment. Messick (1995) defined this type of validity as “evidence and rationales
for evaluating the intended and unintended consequences of score interpretation
and use in both the short- and long-term” (p. 7). The enhancement of consequential
validity of assessment will lead to bridging the gap between diagnosis and intervention.
Subsequently, the central research question of this thesis is: How to increase
consequential validity of (dynamic) assessment?
Research regarding consequential validity demands for a two-sided approach. Not only
the assessment procedure itself should be investigated, but also the skills of professionals
involved who interpret and translate the outcomes into educational practice, i.e.,
CHAPTER 1
20
the significant others. Questions that will be addressed are: What is the role of test
outcomes in psycho-educational practice, and how do current dynamic assessment
procedures contribute to this issue? How should educational professionals’ beliefs and
skills be enhanced in the case of at-risk students? The present thesis brings the results
of several explorative studies that address these questions in an attempt to enhance the
consequential validity of (dynamic) assessment procedures.
Overview of the thesisDue to the focus on significant others in the optimisation of learning potential of at-risk
students, and herewith the consequential validity of dynamic assessment, a strong focus on
contextual aspects is inevitable. Therefore, to warrant ecological validity, a mixed method
approach is required for the thesis. The combination of qualitative and quantitative studies
allows examining the current educational practices in depth.
In the next chapter, chapter 2, first the results of a literature review on dynamic assessment
procedures will be presented. The introduction of this article deals with the different formats
of dynamic assessment and provides evidence for the use of dynamic assessment in practice.
However, the consequential validity of existing dynamic procedures needs to be investigated.
Do outcomes indeed provide more accurate clues to bridge the gap between assessment and
intervention than static procedures? This chapter will provide an overview of the current
dynamic assessment procedures, the differences in their administrations and outcomes,
and its consequences for psycho-educational practice. A qualitative analysis of the learning
phase of each instrument revealed an overview of actions undertaken by examiners. This
will provide useful information to the above-mentioned bridging gap issue.
The issue of diagnosis of learning difficulties, and differentiated instruction in the
general classroom is addressed in the third chapter. More specifically, the issue of ‘didactic
resistance’ is central to this chapter. From a Vygotskian perspective didactic resistance also
implies an assessment of teachers’ activities instead of a focus upon students’ outcomes
solely. As described above, the Response to Instruction (RTI) model is an instrument
GENERAL INTRODUCTION
21
1
to assess both student’s progress and teachers’ instruction, in theory. To examine its
usefulness in practice, a study was carried out in which the model was implemented at
11 schools in The Netherlands. A qualitative multi-method study provided a thorough
analysis of the implementation process. The chapter gives information about the RTI
model and its application in the Dutch context with respect to its value for assessment
of didactic resistance.
The outcomes of the third chapter revealed a difference between beliefs and behaviour
of teachers. The fourth chapter elaborates on this issue, since the behaviour of teachers
has shown to have great influences on students’ learning outcomes. A quantitative
study amongst 44 teachers and 57 support professionals has been carried out to study
the influence of beliefs, in particular implicit theories of intelligence, to the actions
of educational professionals. Hierarchical regression analyses and Structural Equation
Modelling revealed the underlying links and moderations. This chapter shows results
on the interplay between implicit theories of intelligence and behaviours of educational
professionals in the case of at-risk students.
The above-mentioned chapters provide detailed information, but all these studies have
been carried out in a specific context. To warrant ecological validity, and to examine
the possibility for implementation of dynamic assessment in the Dutch educational
practice, a careful examination of the Dutch educational context should be carried
out. Therefore, chapter 5 describes the results of two qualitative studies. The first study
is a qualitative analysis of in-vivo questionnaires among 36 Dutch educational advisors,
21 remedial teachers and 44 teachers. Moreover, a case study has been conducted to
investigate outcomes of the first study more in depth. The ‘care’ processes around
an at-risk student in first grade has been followed during a four-month period. The
combination of both studies revealed a detailed overview of the current practices in
case of at-risk students. The chapter has a strong focus on the outcomes of assessment
procedures in Dutch educational practice.
CHAPTER 1
22
Since an underlying aim of the thesis was to optimise learning potential in the general
classroom, an intervention was developed to enhance teachers’ skillfulness in handling
at-risk students in the general classroom. Chapter 6 describes a pilot study in which a short
intervention of four sessions was implemented at a Dutch regular primary school (N=21
teachers). By making use of multilevel growth curve analyses, the effects of the training
have been analysed quantitatively. The results are presented in the sixth chapter.
In sum, with respect to the two-sided approach of the issue around consequential validity
these chapters can be grouped. Chapters 2 and 5 explore consequential validity from
an assessment-perspective, whereas chapter 3, 4, and 6 focus on the significant others-
perspective. The results of the above-mentioned studies gave thought for reflection
on theoretical assumptions. Consequential and ecological validity will be mirrored,
next to the methods used in the thesis. The value of dynamic assessment for (psycho-)
educational practice will be stressed from a classroom perspective. Hence, theoretical
and practical implications of the studies will be listed in the final chapter.
GENERAL INTRODUCTION
23
1
Chapter 2A review scrutinising the consequential validity of dynamic assessment
This chapter is published as: Tiekstra, M., Minnaert, A., & Hessels, M. G. P. (2016).
A review scrutinising the consequential validity of dynamic assessment. Educational Psychology, 36, 112-137. doi:10.1080/01443410.2014.915930
CHAPTER 2
26
Abstract
This literature review explored whether dynamic assessment procedures in psycho-
educational practice might bridge the well-known gap between diagnosis and
intervention. Due to a learning phase included in the testing procedure, qualitative
information about the child’s learning needs can be revealed by means of dynamic
assessment. The question is, however, what the consequential validity, i.e. the extent
to which assessment influences instructional and learning processes, of dynamic
assessment procedures really is. The review of 31 articles that met the inclusion criteria
showed that proximal consequential validity of dynamic assessment is warranted, but
distal consequential validity is warranted to a lesser extent (e.g., some guidelines for
practice). Furthermore, it can be noticed that motivational aspects never played an
explicit role during learning phases. In order to design student-tailored interventions
following dynamic assessment, there is a need for more explicitness of learning phases
and types of feedback in the development of these instruments.
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
27
2
Introduction
Nowadays, in cohesion with current inclusive education policies, it is a challenge for
educational practitioners to include children having severe learning difficulties into
regular education (UNESCO, 1994). To be able to adopt these children, a proper
assessment of the child’s capacities is requested. On top of this, in case of special needs
education, individualized education plans (IEP) are obligatory in order to construct
adequate interventions. This, in turn, requires an indication of the child’s ability to
learn. Not only is the necessity for appropriate assessment procedures underlined here,
but also the need for appropriate intervention guidelines. Nevertheless, in practice, the
gap between diagnosing and intervention is far too often neglected or even disregarded
(Minnaert & Vermunt, 2006; Resing, 2006; Ruijssenaars, 2001; Shapiro, 1987; Shapiro
& Kratochwill, 2000; Wilson, 1998).
A way to resolve this issue is to highlight the importance of consequential validity
during assessment procedures. Additionally, already in 1989, Messick pointed at the
lack of focus on consequences of test use in validity research of tests. He stated that
validity refers to the “degree to which empirical evidence and theoretical rationales
support the adequacy and appropriateness of inferences and actions based on test scores”
(Messick, 1989, p.5). Accordingly, Gielen, Dochy and Dierick (2003) investigated the
influence of assessment on learning of students, which could be defined as one form
of consequential validity of assessment. Besides, consequential validity refers to the
influence of assessment on the instructional process (Messick, 1995). Furthermore, a
division can be made in the extent to which or the moment that consequential validity
occurs: proximal consequential validity refers to the consequences that influence
learning and teaching during the testing procedure, while distal consequential validity
denotes consequences of testing beyond the testing procedure, i.e., later on in the
instructional processes in the classroom.
CHAPTER 2
28
In static testing both types of consequential validity can profoundly be questioned.
Amongst others, Shapiro and Kratochwill (2000) pointed at the great demands
from educational practice for cues following the outcomes of an IQ test. A dynamic
assessment procedure is a more promising instrument to bridge the gap between
assessment and instruction. During dynamic assessment a learning phase is included in
which the examiner has an active role. During testing the examinee is being taught the
prerequisites of the test procedure and interaction plays a major role. In general, the aim
of dynamic assessment is twofold, namely to be able to make a more accurate evaluation
of a child’s capacities and to provide useful information for clinical purposes. One could
state that during dynamic testing the proximal consequential validity is warranted, since
an explicit opportunity to learn is provided during testing and the examiner should
act in such way that an optimization of the child’s learning is revealed. Moreover, it is
measured to what extent the examinee profits from explanations and feedback, which
could give clues for future actions in the classroom. Therefore, dynamic testing could be
a very promising tool to bridge the gap between diagnosis and intervention.
Dynamic assessmentThe theoretical framework for dynamic assessment procedures is based on the theory
of Vygotsky and his conceptualization of the Zone of Proximal Development (Vygotsky,
1978). This Zone of Proximal Development (ZPD) is the difference between the actual
independent achievement of the child and the performance level of the child when
tailored help is provided (see also Lidz, 1995). Teaching in the Zone of Proximal
Development helps fulfilling the child’s learning potential. Hence, the aim of the
learning phase in dynamic assessment procedures is to optimize learning processes
by means of teaching learning strategies or provide scaffolding and prompts during
learning processes.
Due to teaching and consequently the active role of the examiner, the testing
situation is adaptive to the examinee. In other words, during dynamic assessment the
responsiveness of the child to adaptive teaching will be evaluated which provides an
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
29
2
individualized diagnosis of the child’s ability to learn and the learning processes that
have been addressed. This is of value while diagnosing children, especially for children
having learning difficulties. Resing (2006) underlined the fact that a dynamic approach
provides qualitative information and much more resembles classroom learning than
static procedures. Meijer (2001) demonstrated that dynamic assessment procedures are
less biased than static measures according to test anxiety and the lack of self-confidence.
Several authors showed the added value of dynamic procedures in the assessment of
at-risk children (Beckmann, 2006; Carlson & Wiedl, 1992; Hessels, 1993, 2000; Resing,
1990; Schlatter & Büchel, 2000).
Different approaches to dynamic assessment exist, e.g., domain-specific versus
general capacity evaluation or focus on intervention versus focus on evaluation. The
main distinction in approaches, however, lies in diverging goals and, thus, manifests
in two different procedures (distinction in role of examiner). In this, there is the
clinical approach of Feuerstein (Feuerstein, Rand & Hoffman, 1979) on the one side.
Supported by amongst others Tzuriel (Shamir & Tzuriel, 2004), this approach aims
at the evaluation of the cognitive modifiability of the child during mediated learning
experiences. On the other side, there are standardised approaches: dynamic tests or so-
called learning potential tests, aiming at the evaluation of learning abilities of children.
These tests try to measure to what extent a child is able to apply new information
that has been taught during the learning phase. In a (pre-test –) training – post-test
format the learning abilities of a child are assessed by measuring the extent to which
the child has profited from the learning phase. In another format, the train-within-test
format, the examinees are offered graduated prompting or feedback. Both formats are
standardised and require an active role of the examiner in teaching strategies required
for resolving test items.
Standardisation of testing procedures makes it possible to replicate testing and diminish
the influence of personal characteristics of the examiner, which is a condition to meet
scientific criteria for tests (objectivity, reliability, and validity). Moreover, criteria for the
CHAPTER 2
30
identification of disabilities applied by the World Health Organization or as indicated
in the Diagnostic Statistic Manual (DSM) are based on assessments showing satisfactory
psychometric characteristics.
During dynamic assessment examiners play an important role, since it is their duty to
optimize performances of the examinee. In dynamic tests this role is more standardised,
whereas it is not in the clinical approaches. Dynamic testing is regarded as a promising
tool to estimate the real capacities of at-risk children and to make better distinguishes
between children with learning disorders and children having learning delays (as
a consequence of their learning environments and former learning experiences).
Dynamic tests fit into these requirements of psychometric characteristics. Therefore,
the focus of this article is rather on dynamic testing as on the more clinical assessment
approaches. More precisely, only dynamic tests with short term learning phases are used
for this review, in order to make a clear distinction between tests that try to measure
learning capacity as opposed to tests that try to change capacity. Consequently, in this
article, the term dynamic assessment refers to a dynamic procedure with a standardised
approach in which a short term learning phase is included in testing.
Aim of the present reviewCurrently, articles about dynamic assessment merely focus on comparisons to
traditional (static) measures or on psychometric characteristics. Since one of the aims
of dynamic assessment is to optimize learning, the purpose of this review is to focus at
the dynamic part in dynamic assessment and to examine what has been written about
the consequential validity of dynamic tests. It is of added value to investigate what the
consequences of dynamic assessment are for practice. Over time, quite a few dynamic
assessment procedures have been developed. In order to clarify what makes dynamic
assessment ‘dynamic’, the learning phases of dynamic assessment procedures will be
scrutinized. The learning phases in these procedures differ latently and manifestly. The
differences will be revealed due to a screening of the presence of cognitive, metacognitive
or motivational strategies (Dina & Efklides, 2009; Wigfield, Byrnes, & Eccles, 2006)
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
31
2
as a goal of the examiner’s activities during the learning phase. In other words, the
focus is on the activities the examiner carries out and whether these are more cognitive,
regulative or affective oriented. Cognitive strategies are strategies that the student
applies to process information, e.g., relating, structuralizing, analysing, memorizing
and applying. Metacognitive strategies are strategies to regulate learning processes,
like planning, evaluating, adjusting, diagnosing, reflecting. Whereas motivational
strategies are more affective oriented, like adjustment of attributions, values, emotions,
expectations and concentration. A further elaboration on these strategies can be found
in the classification schemes of Vermunt (1996).
Moreover, the outcomes of different dynamic instruments will be investigated to
unravel the value of dynamic assessment for construction of interventions that might
follow the assessment. In this way the consequential validity of each instrument will be
emphasised.
Over years, several dynamic assessment procedures have been developed for higher
education (e.g., Antón, 2009; Meijer & Elshout, 2001; Nirmalakhandan, 2009). In
these assessment procedures the focus is more on career or professional perspectives
than on the optimization of development of young individuals. This review, however,
aims at more insight around the construction of adequate classroom interventions for
children with some kind of special needs, since these will contribute to the optimization
of development of children. Moreover, in order to optimize a child’s development (or in
other words, to adapt the curriculum as best as possible to the child’s learning needs), an
early recognition of a child’s learning difficulties is necessary. An early evaluation of a
child’s learning weaknesses and strengths is a prerequisite to provide early intervention,
which, in turn, paves the way for optimal development. Therefore, this review will only
include dynamic tests estimating the ability, capacity or potential to learn of children
in regular primary (inclusive) educational settings and the transition to secondary or
middle schools. Hence, studies carried out in grades one to eight in regular education
are the review’s objectives.
CHAPTER 2
32
Method
The literature review consisted of an extensive computer search. For this purpose the
database of Web of Science was used. This database includes the Science Citation
Index Expanded (SCIE), the Arts and Humanities Citation Index (A&HCI) and Social
Sciences Citation Index (SSCI). The use of this database implies that the references
found are from articles that are published in international peer reviewed scientific
journals. References from 1995 until 2011 have been included.
In order to capture as many relevant references as possible, a ‘search model’ had been
developed. The model tries to combine several keywords into three categories which
were regarded as most valuable to find literature about dynamic assessment. The three
categories were the following: potential, dynamic, ability/capacity. The category potential
enclosed the keywords ‘learning potential’ and ‘students’ potential’. 245 Articles had
been found with these keywords. The second category dynamic comprised the keywords
‘dynamic assessment’; ‘dynamic test’; ‘dynamic tests’; ‘dynamic testing’. This resulted
in 2828 articles. The category ability/capacity enclosed the keywords ‘cognitive abilities’;
‘ability testing’; ‘ability test(s)’; ‘ability to learn’; ‘capacity testing’. This third category
had been used only in combination with one of the two other categories to narrow
down the number of hits in the database. For example, the keyword ‘cognitive abilities’
was combined with ‘dynamic’ in a first search, and in a second search combined with
‘potential’. A total of 574 articles were found using these combinations.
The search strategy resulted in a total of 3420 articles after having deleted overlapping
articles between (and within) the different categories. The list of 3420 references was
screened on relevance to the field of learning and development. Subsequently, the total
number of articles was reduced since they did not meet the inclusion criteria for the
review study. In line with the previous section, the following inclusion criteria were
formulated:
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
33
2
1. The article fully describes an empirical study focusing at a dynamic testing
procedure to assess cognitive abilities.
2. The dynamic testing procedure described in the article is standardised and
includes a short term learning phase.
3. The study focuses on children aged 6 to 14 years (in the case of special
education a division has been made based on the average curriculum taught
that fits these ages).
4. The study has been carried out in regular primary or special needs education.
The greatest exclusion of articles was caused by the first criterion, since a large part of
the articles found did not describe a procedure to assess cognitive abilities. Moreover,
some references were excluded since they were not full articles, such as book reviews,
comments, or abstracts for presentation at an international conference (45) or were
not written in English (18). Finally, two references were untraceable: the articles could
neither be retrieved from the internet, nor available at any of the university libraries in
The Netherlands, nor after contact with the main author. The number of articles that
has been used for the literature review is 31.
Results
Tables 1 and 2 present an overview of the results following the literature review. Table
1 summarizes the empirical studies according to the instruments being used and their
psychometric properties (if described). Table 2 zooms in on the dynamic tests that have
been used in the studies. The aim of table 2 is to reveal differences and similarities
between the dynamic tests with respect to their learning phases.
It is emphasized, though, that psychometric properties described below should be
considered critically. According to the authors, some methodological errors exist.
Reliability measures, for example, vary between Cronbach’s alpha or interrater
CHAPTER 2
34
agreements, and some studies use test-retest reliability. The latter form does, however, not refer
to reliability in this context, but rather to a measure for construct validity of the dynamic test.
After all, test-retest reliability can be seen as a measure for stability, which is not in accordance
with the basis of dynamic testing.
As can be seen in table 1, slightly more than half (18 of 31) of the studies enclosed in the review
have been carried out in North-America (USA/Canada), the remaining in Europe, Israel or
Africa. In line with our inclusion criteria, dynamic assessment procedures described were mostly
developed for children in grade 1-8, but the age ranged 4,5 years to 78,6. Some articles report
studies that have been carried out in special education. In these studies the mean chronological
age goes beyond the age of our interest, however, the average curriculum taught or tested fits our
criterion age range. Roughly half of the articles found (18 of 31) describe studies with a sample
size larger than or equal to 100.
Test procedures differed from train-within-test format to (pre-test –) training – post-test format.
Overall, reliability, as mentioned in the method section as well as in results section in the articles,
of the instruments was satisfactory, but a huge variation in reliability coefficients was encountered.
The internal consistencies ranged from α .37 to .98, test-retest or pre-test – post-test reliabilities
from .54 to .72 and interrater agreements from .64 – 1.00. Validity measures include different
values and calculations, as is shown in table 1. Predictive validity (if indicated in the articles) was
of particular interest, since this form of validity matches one of the reasons to do an assessment
in educational settings; namely to provide an estimation or prediction of students’ future
learning outcomes. Therefore, a division was made of predictive validity based on comparisons
of dynamic assessment scores with either school achievement test or static intelligence test
(IQ) scores. The correlations with school achievement test scores varied from .11 to .71, and
described explained variances from 1% to 54%. One study showed an explained variance
of 40% in reading growth. Three studies demonstrate 4%, 25% and 40% explained variance
on top of IQ. The range of correlations with IQ tests varies from -.18 to .74. However, using an
interpretation of the correlation between IQ and DA measure solely is too simplistic. A criterion
measure is needed to be able to estimate the predictive validity.
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
35
2
Tabl
e 1
| O
verv
iew
of
arti
cles
sel
ecte
d fo
r th
is r
evie
w
Auth
ors
(yea
r of
pu
blic
atio
n)C
ount
ryN
Subj
ects
Dyn
amic
test
Test
pr
oced
ure
Rel
iabi
lity
(Pre
dict
ive)
va
lidit
yC
onse
quen
tial
va
lidit
y
Com
pton
, Fuc
hs,
Fuch
s, Bo
uton
, G
ilber
t, Ba
rque
ro,
Cho
, & C
rouc
h (2
010)
USA
355
Gra
de 1
DA
ps
eudo
wor
dsTr
ain
with
in te
st Te
st-re
test
(4 w
eeks
): .7
2R
ange
: .1
1 to
.65
Scor
es in
dica
te
instr
uctio
nal
scaff
oldi
ng le
vel n
eede
d to
cor
rect
ly so
lve
item
s (m
in. 3
; max
. 15)
. The
lo
wer
scor
es, t
he le
ss
scaff
oldi
ng
De
Beer
(201
0a)
Sout
h A
frica
79 (b
oys)
72 (g
irls)
mea
n ag
e bo
ys: 1
2.44
mea
n ag
e gi
rls:
11.1
8
Lear
ning
Po
tent
ial
Com
pute
rised
A
dapt
ive
Test
(LPC
AT)
Test
– tr
aini
ng
– te
st
α .9
3 to
.98
Scho
ol
achi
evem
ent:
boys
: R² =
.33
girls
: R² =
.32
Post-
test
scor
e,
diffe
renc
e sc
ore
and
com
posit
e sc
ore
Dol
ores
Cal
ero,
Be
len,
&
Aux
iliad
ora
Rob
les
(201
1)
Spai
n12
77
– 11
yA
dapt
ed
subt
ests
of th
e LP
AD
:
Posit
ions
test
The
org
aniz
er
test
Sten
cil d
esig
n te
st
Trai
n w
ithin
test
Test
– tr
aini
ng
– te
st
Test
– tr
aini
ng
– te
st
α .7
0 to
.95
Indi
rect
ly ta
ken
into
acc
ount
by
mak
ing
use
of
Disc
rimin
ant
Ana
lysis
Gai
n sc
ore,
bas
ed o
n as
sista
nce
need
ed
Raw
scor
es o
n pr
etes
t, po
sttes
t, an
d ga
in sc
ore
Gai
n sc
ore,
bas
ed o
n as
sista
nce
need
ed
CHAPTER 2
36
Elle
man
, Com
pton
, Fu
chs,
Fuch
s, &
Bo
uton
(201
1)
USA
100
Gra
de 2
m
ean
age
8y3m
DA
infe
renc
e ge
nera
tion
Test
– tr
aini
ng
– te
st –
tran
sfer
α .7
6
Inte
rrat
er
agre
emen
t 97
,8%
Cog
nitiv
e te
st:
4% u
niqu
e va
rianc
e ex
plai
ned
by D
A
3 sc
ores
: num
ber
of p
rom
pts n
eede
d,
tran
sfer s
core
and
tota
l sc
ore
Fabi
o (2
005)
Italy
287
6 –
11
(mea
n ag
e:
8 y
2 m
)
Dyn
amic
Tes
t of
Pot
entia
l In
telli
genc
e
Trai
n w
ithin
test
α .8
6IQ
: .48
(Rav
en
CPM
)Sc
hool
ac
hiev
emen
t: .2
7 to
.36
Gen
eral
Mod
ifiab
ility
In
dex
(= sc
ore
base
d on
nee
d fo
r hel
p)
Fuch
s, C
ompt
on,
Fuch
s, Bo
uton
, &
Caff
rey
(201
1)
USA
318
Gra
de 1
DA
ps
eudo
wor
dsTr
ain
with
in te
stTe
st –
rete
st (4
w
eeks
): .7
2 2.
3% u
niqu
e ex
plai
ned
varia
nce
in w
ord
iden
tifica
tion
1.0%
uni
que
expl
aine
d va
rianc
e in
read
ing
com
preh
ensio
n
Scor
es in
dica
te
instr
uctio
nal
scaff
oldi
ng le
vel n
eede
d to
cor
rect
ly so
lve
item
s (m
in. 3
; max
. 15)
. The
lo
wer
scor
es, t
he le
ss
scaff
oldi
ng
Fuch
s, C
ompt
on,
Fuch
s, H
olle
nbec
k,
Cra
ddoc
k, &
H
amle
tt (2
008)
USA
122
Gra
de 3
DA
alg
ebra
sk
ills
Trai
n w
ithin
test
Indi
rect
ly ta
ken
into
acc
ount
by
mak
ing
use
of S
truc
tura
l Eq
uatio
n M
odel
ing
(SEM
)
Mat
h (sc
hool
ac
hiev
emen
t)
.41
to .6
0
Scor
es in
dica
te
instr
uctio
nal
scaff
oldi
ng le
vel n
eede
d to
cor
rect
ly so
lve
item
s (m
in. 0
; max
. 21)
. The
m
ore
scaff
oldi
ng, t
he
less
poi
nts
Fuch
s, C
ompt
on,
Fuch
s, H
olle
nbec
k,
Ham
lett,
&
Seet
hale
r (20
11)
USA
122
Gra
de 3
DA
alg
ebra
sk
ills
Trai
n w
ithin
test
The
aut
hors
re
fer t
o th
e ab
ove
men
tione
d ar
ticle
of
2008
Impr
ovem
ent
of c
lass
ifica
tions
du
e to
DA
: 48%
to
73%
Scor
es in
dica
te
instr
uctio
nal
scaff
oldi
ng le
vel n
eede
d to
cor
rect
ly so
lve
item
s (m
in. 0
; max
. 21)
. The
m
ore
scaff
oldi
ng, t
he
less
poi
nts
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
37
2
Gill
am, F
argo
, Fo
ley,
& O
lszew
ski
(201
1)
USA
646
– 8,
5N
onve
rbal
D
A p
hone
me
dele
tion
Trai
n w
ithin
test
α .8
8C
onstr
uct v
al.:
r. 84
Wor
d le
vel
read
ing:
r. 5
4
Scor
es re
flect
ing
num
ber o
f pr
ompt
s ne
eded
(min
. 21;
max
10
5)
Gum
mer
sall
&
Stro
ng (1
999)
USA
30 (e
xp. 1
) +
12
stude
nts
with
lang
uage
pr
oble
ms
(exp
. 2)
8 –
9 (g
rade
3)
Nar
rativ
e sk
ill
asse
ssm
ent
Test
– tr
aini
ng
– te
st
Inte
rrat
er
agre
emen
t 99
.2%
to 1
00%
.68
to .9
8
?St
atic
scor
e (s
tory
-le
ngth
and
synt
ax)
+ q
ualit
ativ
e in
terp
reta
tion
of e
rror
s (a
udio
tape
d)
Ham
ers,
Hes
sels,
&
Penn
ings
(199
6)T
he
Net
herla
nds
500
5 y
4 m
– 7
y
9 m
T
he L
earn
ing
Pote
ntia
l Tes
t fo
r Eth
nic
Min
oriti
es
(LEM
)
Trai
n w
ithin
test
Ran
ge
α .8
8 to
.92
IQ:
.45
to .6
8Sc
hool
ac
hiev
emen
t:.3
2 to
.69
stand
ardi
zed
LP sc
ores
(n
orm
refe
renc
ed)
Jelto
va, B
irney
, Fr
edin
e, Ja
rvin
, St
ernb
erg,
&
Grig
oren
ko (2
011)
USA
1332
Gra
de 4
Gro
up-
adm
inist
ered
dy
nam
ic
asse
ssm
ent
(GD
A)
mat
hem
atic
s
Test
– tr
aini
ng
– te
st
IRT
relia
bilit
y .7
9 to
.84
for
subt
ests
Take
n in
to
acco
unt b
y m
akin
g us
e of
H
LM
Raw
scor
es o
n pr
e- a
nd
postt
est
Jiten
dra
&
Kam
eenu
i (20
01)
USA
46G
rade
3
(8 y
9 m
–
9 y
10 m
)
Mat
hem
atic
al
wor
d pr
oble
ms
Trai
n w
ithin
test
Inte
rrat
er
agre
emen
t 92
to 1
00%
?N
umbe
r of
prom
pts
and
time
elap
sed
Kan
evsk
y &
Gea
ke
(200
4)C
anad
a25
8 y
10 m
–
10 y
1 m
Mat
hem
atic
al
patte
rn-
reco
gniti
on
task
s
Trai
n w
ithin
test
Inte
rrat
er
agre
emen
t 92
to 9
6%
Thi
s stu
dy:
IQ:
r = .3
4
Nee
d of
supp
ort
(tota
l num
ber o
f hi
nts +
qua
litat
ive
obse
rvat
ions
)
CHAPTER 2
38
Kat
z, G
olsta
nd,
Bar-
Ilan,
& P
arus
h (2
007)
Isra
el38
16-
12 y
The
Dyn
amic
O
ccup
atio
nal
The
rapy
C
ogni
tive
Ass
essm
ent
for c
hild
ren
(DO
TC
A-C
h)
Test
– tr
aini
ng
– te
st
ICC
(int
erra
ter)
.87
to .9
9In
t. co
nsist
ency
α.
61 to
.77
Ran
ge .3
6 to
.71
Leve
l of
med
iatio
n ne
eded
: cut
off
scor
es
+ d
ynam
ic p
erce
ntag
e sc
ores
Kirk
woo
d, W
eile
r, Be
rnste
in, F
orbe
s, &
W
aber
(200
1)
USA
202
(het
erog
eneo
us
lear
ning
pr
oble
ms)
7 –
11R
ey-O
sterr
ieth
C
ompl
ex
Figu
re (R
OC
F)
– dy
nam
ic
adap
tatio
n
Test
– tr
aini
ng
– te
st
??
Raw
scor
es b
ased
on
Dev
elop
men
tal S
corin
g Sy
stem
are
div
ided
in
leve
ls; d
ivisi
on in
two
subg
roup
s (be
nefit
/no
bene
fit)
Lars
en &
Nip
pold
(2
007)
USA
50G
rade
6M
ean
age
12,2
(age
ra
nge:
10
y 9
m –
12
y 10
m)
Dyn
amic
A
sses
smen
t Ta
sk o
f M
orph
olog
ical
A
naly
sis
(DAT
MA
)
Trai
n w
ithin
test
Inte
rsco
rer
agre
emen
t 95
to 1
00 %
Inte
rnal
co
nsist
ency
is
stron
g *
DAT
MA
&
PPV
T-ΙΙ
Ι : r.
36Sc
hool
ac
hiev
emen
t :
DAT
MA
&
OSA
: r.5
0
Scor
es b
ased
on
need
fo
r pro
mpt
s: ra
w
scor
es, m
ax. 7
5
Lauc
hlan
& E
lliot
t (2
001)
UK
30 (s
peci
al
educ
atio
n)M
ean
age
9;
mod
erat
e/se
vere
le
arni
ng
diffi
culti
es
The
Chi
ldre
n’s
Ana
logi
cal
Thi
nkin
g M
odifi
abili
ty
test
(CAT
M)
Prel
imin
ary
phas
e –
test
– tr
aini
ng
– te
st
Rel
. pro
ven
*Va
l. pr
oven
*R
aw sc
ores
(min
. 0;
max
. 18)
Lidz
& M
acrin
e (2
001)
USA
81
Gra
de
1 –
5 N
aglie
ri N
onve
rbal
A
bilit
y Te
st –
mod
ified
fo
r dyn
amic
ad
min
istra
tion
(NN
AT/D
A)
Test
– tr
aini
ng
– te
st
?C
oncu
rren
t va
lidity
K-A
BC
.64
to .7
4
Stan
dard
scor
es, b
ased
on
NN
AT n
orm
s
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
39
2
Pelte
nbur
g, V
an d
en
Heu
vel-P
anhu
izen
, &
Doi
g (2
009)
The
N
ethe
rland
s37
(spe
cial
ed
ucat
ion)
8,9
– 12
,11
(mea
n ag
e 10
,5 y
ears
)Av
erag
e cu
rric
ulum
7-
8 ye
ars
CIT
O
Mon
itorin
g Te
st fo
r M
athe
mat
ics,
rede
signe
d fo
r IC
T a
nd D
A
Trai
n w
ithin
test
??
Raw
scor
es +
qu
alita
tive
obse
rvat
ions
on
stra
tegy
use
Pelte
nbur
g, V
an d
en
Heu
vel-P
anhu
izen
, &
Rob
itzsc
h (2
010)
The
N
ethe
rland
s37
+ 4
3 (sp
ecia
l ed
ucat
ion)
8 –
12
Aver
age
curr
icul
um
leve
l: en
d gr
ade
2
CIT
O
Mon
itorin
g Te
st fo
r M
athe
mat
ics,
rede
signe
d fo
r IC
T a
nd D
A
Trai
n w
ithin
test
Indi
rect
ly ta
ken
into
acc
ount
by
mak
ing
use
of
diffe
rent
ial i
tem
fu
nctio
ning
(D
IF)
?R
aw sc
ores
+
qual
itativ
e ob
serv
atio
ns
on st
rate
gy u
se
Res
ing,
Tun
tele
r, D
e Jo
ng, &
Bos
ma
(200
9)
The
N
ethe
rland
s10
9 (5
5 et
hnic
- m
inor
ity
child
ren)
7y –
9y
Seria
-Thi
nk
– ad
apte
d ve
rsio
n
Test
– tr
aini
ng
– te
st
Pre-
postt
est
corr
elat
ion:
.5
7; .6
1
?N
umbe
r of
inse
rtio
ns
and
mea
sure
men
ts +
str
ateg
y-le
vel c
ateg
orie
s
Ster
nber
g,
Grig
oren
ko,
Ngo
rosh
o,
Tant
ufuy
e, M
bise
, N
okes
, Juk
es, &
Bu
ndy
(200
2)
Tanz
ania
358
(+ 1
00
child
ren
in
cont
rol g
roup
)
11y
– 13
ySy
llogi
sms
Sort
ing
Twen
ty
Que
stion
s
Test
– tr
aini
ng
– te
st
Test
– tr
aini
ng
– te
st
Test
– tr
aini
ng
– te
st
Test
– re
test
relia
bilit
y: .5
4
Test
– re
test
relia
bilit
y: .5
8 /
.64
Test
– re
test
relia
bilit
y: .6
6
Cog
nitiv
e te
sts:
.21
to .3
1Sc
hool
ac
hiev
emen
t:.2
9 to
.33
Cog
nitiv
e te
sts:
.29
to .3
5 /
.32
to .3
8Sc
hool
ac
hiev
emen
t:.3
5 to
.42
/.3
4 to
.50
Cog
nitiv
e te
sts:
.12
to .3
2Sc
hool
ac
hiev
emen
t:.1
5 to
.28
Num
ber o
f co
rrec
t an
swer
s in
pret
est,
trai
ning
and
pos
ttest
Num
ber o
f pe
rsev
erat
ive
erro
rs
and
cate
gorie
s pr
oduc
ed +
num
ber o
f hi
nts (
trai
ning
pha
se)
Rat
io o
f co
nstr
aint
-se
ekin
g qu
estio
ns
to to
tal n
umbe
r of
ques
tions
CHAPTER 2
40
Swan
son
(199
5a)
USA
Exp.
1: 6
1Ex
p. 2
: 191
Mea
n ag
e 10
.63
Mea
n ag
e 10
.66
S-C
PT
(Sw
anso
n-C
ogni
tive
Proc
essin
g Te
st)
Trai
n w
ithin
test
α' s:
.80
to .9
5
Ove
rall
relia
bilit
y: α
. 9
8
14%
to 2
6%
expl
aine
d va
rianc
e on
scho
ol
achi
evem
ent
In a
dditi
on to
initi
al
stand
ard
scor
e: g
ain
scor
e, p
robe
scor
e an
d m
aint
enan
ce sc
ore
Swan
son
(199
5b)
USA
/
Can
ada
1611
(10%
sp
ecia
l edu
c.)
Age
rang
e:
4,5
– 78
,6
S-C
PT
Trai
n w
ithin
test
α‘s:
.82
to .9
5 32
% e
xpla
ined
va
rianc
e in
read
ing
reco
gniti
on
7 co
mpo
site
stand
ard
scor
es (M
=10
0;
SD=
15)
Swan
son
(201
1)U
SA78
M
ean
age
11.6
24
subt
ests
of S
-CPT
: rh
ymin
g,
digi
t-sen
tenc
e,
visu
al m
atrix
, m
appi
ng a
nd
dire
ctio
ns
Trai
n w
ithin
test
Subt
ests
corr
elat
ion
rang
e .4
0 to
.80
40%
exp
lain
ed
varia
nce
in
read
ing
grow
th
Com
posit
e sta
ndar
d sc
ores
. Ini
tial s
core
, ga
in sc
ore,
pro
be sc
ore
and
mai
nten
ance
scor
e
Swan
son
& H
owar
d (2
005)
USA
70 st
uden
ts w
ith re
adin
g (a
nd m
ath)
pr
oble
ms
Mea
n ag
e 11
.93
2 su
btes
ts of
S-C
PT:
rhym
ing
and
digi
t-sen
tenc
e
Trai
n w
ithin
test
α’s .
76 to
.85
25%
ext
ra
(uni
que)
va
rianc
e on
top
of IQ
Initi
al sc
ore,
gai
n sc
ore,
pro
be sc
ore
and
mai
nten
ance
scor
e
Tie
kstr
a, H
esse
ls, &
M
inna
ert (
2009
)Sw
itzer
land
46 (s
peci
al
educ
atio
n)15
y 6
m
– 19
y 4
m
(mea
n ag
e 17
yea
rs)
Aver
age
curr
icul
um
8 –
10
year
s
Hes
sels
Ana
logi
cal
Rea
soni
ng T
est
(HA
RT)
Trai
ning
–
postt
est
α’s .
69 to
.92
Thi
s stu
dy: .
65
IQ:
-.18
Scho
ol
achi
evem
ent
(CLT
):.5
7
40%
pre
dict
ion
in C
LT o
n to
p of
IQ
Raw
scor
es (m
in. 0
; m
ax. 2
0)
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
41
2
Tzur
iel (
2000
)Is
rael
48G
rade
2
(mea
n ag
e =
7,5
y)
Seria
-Thi
nk
Instr
umen
tPr
elim
inar
y ph
ase
– te
st –
trai
ning
–
test
α ' s
.37
to .8
5Sc
hool
ac
hiev
emen
t (m
ath)
:R
² = .1
9 R
² = .5
4
Raw
scor
es, d
ivid
ed
in p
erfo
rman
ce le
vel
(max
. 20)
; num
ber o
f in
sert
ions
; num
ber o
f m
easu
rem
ents
Tzur
iel &
Kau
fman
(1
999)
Isra
el52
Gra
de 1
(m
ean
age
= 7
,16
and
Ethi
opia
n gr
oup:
6 y
–
7 y
6 m
)
The
Chi
ldre
n’s
Ana
logi
cal
Thi
nkin
g M
odifi
abili
ty
test
(CAT
M)
The
Chi
ldre
n’s
Infe
rent
ial
Thi
nkin
g M
odifi
abili
ty
test
(CIT
M)
Prel
imin
ary
phas
e –
test
– tr
aini
ng
– te
st
Test
– tr
aini
ng
– te
st –
tran
sfer
phas
e
Rel
. ok
*
α' s
.82
to .9
0
Val.
ok *
Val.
ok *
Raw
scor
es: “
all-o
r-no
ne”
and
“par
tial
cred
it”
Raw
scor
es
Vilj
oen,
Odg
ers,
Gris
so &
Till
broo
k (2
007)
USA
190
(subg
roup
)11
y –
13y
Mac
Art
hur
Com
pete
nce
Ass
essm
ent
Tool
-Crim
inal
A
djud
icat
ion
(Mac
CAT
-CA
)
Trai
n w
ithin
test
Intr
acla
ss
corr
elat
ion
.64
to .9
1
?R
aw sc
ores
Note.
* =
onl
y in
form
atio
n in
dica
ted
in th
e ar
ticle
; ? =
lack
ing
info
rmat
ion
in th
e ar
ticle.
CHAPTER 2
42
Consequential validityWith regard to consequential validity of dynamic assessment procedures included in the
review, it is of importance to investigate the possible consequences of these testing procedures
in practice. During analysis of articles this was reflected by the extent to which the outcomes
of dynamic assessment procedures were made explicit for practice. Indeed, the procedures
differed in their extent of explicit information for practice. There were three categories in
which the dynamic assessment procedures appeared to be divided. It is highlighted to what
extent proximal consequential validity (raw or self-evident scores) and distal consequential
validity (explicit information for practice) occur in assessment procedures in each of these
categories.
Raw or standardised scores. Dynamic assessment procedures that result in a raw or
standardised score have a minimal form of consequential validity. Although the LPCAT
(De Beer, 2010a) provides an indication of the education level of the examinee compared
to national education levels (De Beer, 2010b), the test does not result in explicit clues for the
development of an intervention. Even so, in the studies with the LEM (Hamers et al., 1996),
the NNAT/DA (Lidz & Marcine, 2001), and the study of Dolores-Calero et al. (2011), the test
scores are norm referenced, but what these scores are made of is not indicated in this article
and thus, it is unclear whether the scores provide indications for practice. Other dynamic
assessment instruments (Jeltova et al., 2011; Tiekstra et al., 2009; Tzuriel & Kaufman, 1999;
Tzuriel, 2000; Viljoen et al., 2007) result in raw scores. Lauchlan and Elliott (2001) describe
a study that investigated the extent to which scores of the CATM could predict the learning
progress of children following a cognitive intervention program. Although this indicates
some potential for distal consequential validity, the CATM scores, however, were only raw
scores similar to the version applied in study of Tzuriel and Kaufman (1999) mentioned
before.
Since a learning opportunity is provided during the abovementioned procedures, the
proximal consequential validity is warranted. The outcomes are represented by a number,
and this number should be appropriate for predicting future learning outcomes. However,
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
43
2
no explicit information about the learning processes is reflected in this number. The gap
between diagnosis and intervention will be apparent after administration of these tests. To
conclude, the distal consequential validity is not guaranteed according to these test outcomes.
Raw scores + indication of level of help needed. Several dynamic assessment
procedures resulted in an indication of the level of support needed in addition to raw
scores. Some of the procedures still result in an isolated number, which needs to be
interpreted in a specific manner. Compton et al. (2010) and Fuchs, Compton, Fuchs,
Bouton et al. (2011), for instance, demonstrated that a dynamic assessment procedure
on pseudo words could forecast a child’s responsiveness to intervention: the procedure
provides a score which indicates the instructional scaffolding level needed which
could link assessment to educational practices, nonetheless implicitly. In the studies of
Fuchs, Compton, Fuchs, Hollenbeck et al. (2008, 2011) , and Larsen & Nippold (2007)
administration of scores is done in a similar way. Minimum and maximum scores are
composed of points for items solved correctly, reduced (or increased) by the number
of hints needed. The tests of Sternberg et al. (2002) and Elleman et al. (2011) result in
raw scores combined with the number of errors made or the number of hints required
during training. This can be seen as a slightly more advanced form of proximal
consequential validity than the previous category, but still, distal consequential validity
is not warranted. No explicit clues for practice could be detected in these procedures.
This also holds for the instruments that result in multiple scores, like the S-CPT
(Swanson, 1995a, 1995b, 2011; Swanson & Howard, 2005), the dynamic administration
of the Rey-Osterrieth Complex Figure Test (Kirkwood et al., 2001), and the adapted
version of the Seria-Think Instrument (Resing et al., 2009). In the latter, though,
an estimation was made of the level of strategies used (least sophisticated strategy to
advanced strategy) in addition to the raw scores. The link to instruction or intervention
after the test is not made explicitly, and the estimation of the child’s competences is
central to these instruments.
CHAPTER 2
44
Other procedures slightly touched distal consequential validity, since these added
qualitative information to the raw scores. Gummersall and Strong (1999) tested a
dynamic procedure for narrative skill assessment. After completion of this assessment
not only static scores are presented, but the qualitative interpretation of errors (audio
taped) provides useful individualized clues for intervention. Similarly, next to raw scores,
Peltenburg et al. (2009, 2010) add qualitative information about strategy use based
on video observations which were made during the assessment procedure. How this
qualitative information is given back to the educational psychologists in order to set up
an IEP remains, however, unclear. Therefore, these instruments only slightly address
distal consequential validity.
Level of mediation needed / need of support. Some dynamic assessment
procedures did not result in raw scores at all. These procedures have been grouped
in this third category. Initially, one would think that this category provides the most
advanced form of consequential validity since scores only reflect the level of help
needed. It appeared, however, that these procedures also never explicitly mentioned
clues for practice. For example, the instruments of Fabio (2005), Gillam et al. (2011),
Jitendra and Kameenui (1996), and Katz et al. (2007) resulted in scores composed of
number of prompts needed during testing. The scores as such do not provide information
for designing interventions. A qualitative analysis of the probes used by each child,
however, can, since the standardised probes depended on the errors made. In this way
the scores provide more information for interventions than the outcomes of dynamic
assessment procedures mentioned previously, although information remains implicit.
Scores that provide more explicit information for practice can be found in the study of
Kanevsky and Geake (2004). They investigated the interactions during testing: concrete,
cognitive or metacognitive instruction by examiner and content-related or spontaneous
(prompting) responses of student. The outcomes of the dynamic procedure indicate the
need of support, composed of the total number of hints, time needed to complete the
tasks, and added by qualitative observations.
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
45
2
To unravel the possibility of distal consequential validity of dynamic assessment
procedures more into detail, learning phases should be scrutinized. Since the learning
phase enhances a teaching component, which could provide valuable information for
future classroom or intervention actions, the next paragraph describes the learning
phases of each instrument (29 in total).
Learning phaseTable 2 shows differences and similarities between the dynamic tests with respect to their
learning phases. Dynamic assessment procedures cover general or specific domains.
Half of the tests try to assess the general ability of the child, the other half abilities in
specific domains as reading or mathematics. The HART (Tiekstra et al., 2009) and the
DA inference generation (Jeltova et al., 2011) are the only instruments that have been
group administered, including a group wise training. Although, the LPCAT (De Beer,
2010a) and the test of Peltenburg et al. (2009, 2010) can be administered in groups too,
since these are computerized tests.
Furthermore, it is of importance to scrutinize the learning phases of each dynamic
assessment procedure in order to investigate the potential consequential validity
for practice. Table 2 zooms in at the different learning phases with regard to their
manifestations and contents. Overall, one could state that a reasonable amount of
instruments use the ‘graduated prompts’ approach. According to Resing et al. (2009)
this implies a standardised training procedure in which prompts are prescribed and
hierarchically ordered. The latter makes the testing procedure adaptive, since the
examiner does not provide more hints or prompts when the child has found the right
answer. Other learning phases are composed of extra practice, extra instruction on
failed items or on strategies and skills required to solve the tasks. During some of the
learning phases general feedback is given.
In order to scrutinize the learning phases more carefully, their content, i.e., the aims of
examiner’s activities, has been examined. Therefore, a division in cognitive, metacognitive
CHAPTER 2
46
Tabl
e 2
| O
verv
iew
of
inst
rum
ents
use
d in
the
stud
ies
sele
cted
for
this
rev
iew
Test
Dom
ain
(spe
c./
gene
ral)
Adm
inis
trat
ion
(ind
./gr
oup)
Test
dur
atio
nM
anif
esta
tion
of
lear
ning
ph
ase
Con
tent
of
lear
ning
ph
ase
CM
CM
DA
pse
udow
ords
Sp
ecifi
c: re
adin
gIn
divi
dual
ly20
– 3
0 m
in.
Gra
duat
ed p
rom
ptin
g (in
struc
tiona
l sca
ffold
ing)
x-
-
LPC
ATG
ener
alIn
divi
dual
ly &
gro
up
wise
(com
pute
rized
)?
Extr
a pr
actic
e; g
ener
al st
rate
gies
&
hin
ts; fe
edba
ckx
*-
Posit
ions
Tes
t
The
Org
aniz
er T
est
Sten
cil D
esig
n Te
st
Gen
eral
(visu
o-sp
atia
l m
emor
y)
Gen
eral
Gen
eral
(per
cept
ual
struc
tura
tion)
Indi
vidu
ally
Indi
vidu
ally
Indi
vidu
ally
3 se
ssio
ns o
f 50
m
inut
es in
tota
lG
radu
ated
instr
uctio
n on
faile
d ite
ms
Instr
uctio
n on
stra
tegi
es,
mem
ory
and
orga
niza
tion
Trai
ning
on
visu
al tr
ansp
ort a
nd
inte
rnal
tran
sform
atio
n of
the
stim
ulus
x - x
- x -
- - -
DA
infe
renc
e ge
nera
tion
Spec
ific:
read
ing
Indi
vidu
ally
25 m
in. –
1 h
our
Gra
duat
ed p
rom
ptin
g +
in
struc
tion
x-
-
Dyn
amic
Tes
t of
Pote
ntia
l In
telli
genc
eG
ener
al
Indi
vidu
ally
14 it
ems
Gra
duat
ed p
rom
ptin
gx
--
DA
alg
ebra
skill
sSp
ecifi
c: m
athe
mat
ics
Indi
vidu
ally
30-4
5 m
in.
Gra
duat
ed p
rom
ptin
g (in
struc
tiona
l sca
ffold
ing)
x-
-
Non
verb
al D
A p
hone
me
dele
tion
Spec
ific:
lang
uage
Indi
vidu
ally
21 it
ems
Syste
mat
ic p
rom
ptin
g +
co
rrec
tive
feed
back
x*
-
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
47
2
Nar
rativ
e sk
ill a
sses
smen
tSp
ecifi
c: la
ngua
geIn
divi
dual
ly?
Extr
a in
struc
tion
(sent
ence
-im
itatio
n ta
sk +
pic
ture
cue
s);
gene
ral f
eedb
ack
x*
-
LEM
Gen
eral
In
divi
dual
ly
?R
epea
ting
item
s; no
nver
bal
feed
back
; dem
onstr
atio
n (g
radu
ated
pro
mpt
ing)
x
*-
GD
A m
athe
mat
ics
Spec
ific:
mat
hem
atic
sG
roup
wise
50
– 7
0 m
in.
Gro
up w
ise d
iscus
sion
and
instr
uctio
n on
pro
blem
-solv
ing
strat
egie
s and
thin
king
skill
sx
x-
Mat
hem
atic
al w
ord
prob
lem
s Sp
ecifi
c: m
athe
mat
ics
Indi
vidu
ally
?G
radu
ated
pro
mpt
ing,
feed
back
xx
-
Mat
hem
atic
al p
atte
rn-
reco
gniti
on ta
sks
Spec
ific:
mat
hem
atic
sIn
divi
dual
ly14
-56
min
. G
radu
ated
supp
ort
xx
-
DO
TC
A-C
hG
ener
al
Indi
vidu
ally
Up
to 1
,5 h
r. Pr
ompt
s; fe
edba
ck d
urin
g te
st-
--
ROC
F - d
ynam
ic a
dapt
atio
nG
ener
al
Indi
vidu
ally
?
Prom
ptin
g: st
ruct
ural
izat
ion
of
figur
e *
x-
DAT
MA
Spec
ific:
lang
uage
Indi
vidu
ally
15 it
ems
Gra
duat
ed p
rom
ptin
gx
--
NN
AT/D
AG
ener
alIn
divi
dual
ly?
Gra
duat
ed p
rom
ptin
gx
--
CIT
O M
onito
ring
Test
for
Mat
hem
atic
s – IC
T/D
ASp
ecifi
c: m
athe
mat
ics
Indi
vidu
ally
15 –
20
min
utes
Visu
aliz
atio
n an
d op
port
unity
to
use
exte
rnal
mem
ory
x*
-
Seria
-Thi
nk –
ada
pted
ver
sion
Gen
eral
: ind
uctiv
e re
ason
ing
Spec
ific:
mat
hem
atic
s
Indi
vidu
ally
3 se
ssio
ns o
f 15
-20
min
utes
(of
whi
ch o
ne se
ssio
n is
trai
ning
)
Gra
duat
ed p
rom
ptin
gx
x-
CHAPTER 2
48
Syllo
gism
sG
ener
alIn
divi
dual
ly
Tota
l adm
inist
ratio
n tim
e ra
nge:
42
– 10
1 m
in. (
mea
n tim
e: 7
1 m
in.)
Inte
rven
tion
time
rang
e: 1
1-28
Feed
back
afte
r pre
test,
than
te
achi
ng th
e sk
ills t
o so
lve
item
s. x
--
Sort
ing
Gen
eral
(exe
cutiv
e fu
nctio
ning
)In
divi
dual
lym
in.
(mea
n tim
e: 1
7 m
in.)
Teac
hing
skill
s to
solv
e th
e ta
sk
(visu
aliz
atio
n)
x-
-
Twen
ty Q
uesti
ons
Gen
eral
(exe
cutiv
e pr
oble
m-so
lvin
g sk
ills)
Indi
vidu
ally
Teac
hing
skill
s to
solv
e th
e ta
sk
x-
-
S-C
TP
Gen
eral
(wor
king
m
emor
y ab
ility
)In
divi
dual
ly±
150
min
. / ±
50
min
. (ab
brev
iate
d ve
rsio
n: 5
/6 su
btes
ts)
Gra
duat
ed p
rom
ptin
gx
--
HA
RTG
ener
al (a
nalo
gica
l re
ason
ing)
Gro
up w
ise
20 m
in. t
rain
ing
+ 2
0 ite
ms p
ostte
st Tr
aini
ng o
n sk
ills n
eede
d fo
r so
lvin
g ite
ms (
prom
ptin
g, e
xtra
in
struc
tion,
ver
baliz
atio
n an
d vi
sual
izat
ion)
xx
-
Seria
-Thi
nk In
strum
ent
Spec
ific:
mat
hem
atic
sIn
divi
dual
lyTe
achi
ng p
hase
: 30
-40
min
.In
struc
tion
on st
rate
gies
+
feed
back
xx
x
CAT
MG
ener
al (a
nalo
gica
l th
inki
ng)
Indi
vidu
ally
Teac
hing
pha
se =
30-
40 m
in.
Instr
uctio
n on
pro
blem
-solv
ing
strat
egie
s x
*x
CIT
MG
ener
al (i
nfer
entia
l th
inki
ng)
Indi
vidu
ally
12 it
ems e
ach
phas
e an
d 10
item
s tra
nsfe
r ph
ase
Extr
a in
struc
tion
+ in
struc
tion
on st
rate
gies
x
*-
Mac
CAT
-CA
Spec
ific:
adj
udic
ativ
e co
mpe
tenc
es (l
egal
un
ders
tand
ing)
Indi
vidu
ally
Onl
y 6
out o
f 8
item
s of
one
subs
cale
w
ere
dyna
mic
ally
ad
min
ister
ed
Extr
a in
struc
tion
on fa
iled
item
sx
--
Note.
Whe
re x
mea
ns e
xplic
itly
men
tione
d in
the
artic
le, a
nd *
mea
ns im
plic
itly
men
tione
d in
the
artic
le, a
nd –
mea
ns n
ot m
entio
ned
at a
ll in
the
artic
le.
C =
cog
nitiv
e le
arni
ng st
rate
gies
, MC
= m
etac
ogni
tive
lear
ning
stra
tegi
es a
nd M
= m
otiv
atio
nal l
earn
ing
strat
egie
s.? =
lack
ing
info
rmat
ion
in th
e ar
ticle.
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
49
2
and motivational content has been made. The majority of tests contains a learning
phase in which the examiner’s activities are oriented towards cognitive strategies only.
In a few instruments, metacognitive strategies are applied next to cognitive strategies.
These are, for example, activities explicitly aimed at planning behaviour or evaluating
one’s own activities and not primarily focused on the solution of task.
In the instruments CATM and Seria-Think (Lauchlan & Elliott, 2001; Tzuriel &
Kaufman, 1999; Tzuriel, 2000) a preliminary phase is included to familiarize the child
with the testing procedure and to observe potential difficulties (Tzuriel, 2000). Lauchlan
and Elliott (2001) report that the examiner addresses cognitive and affective factors
during the learning phase of the CATM. In the article of Tzuriel (2000) about the Seria-
Think Instrument it is mentioned explicitly that during mediation (learning phase) the
examiner gives feedback in order to assure the child’s feeling about competence, even
if the answer is not correct. This can be interpreted as an affective, or motivational,
strategy of the examiner. The remaining instruments do not describe motivational
factors other than some encouragements (e.g., Jeltova et al., 2011) which are not
equivalent to motivational factors as described in the introduction.
It must be mentioned, though, that these results are based on rather vague and implicit
information, in such way that most of the studies do not describe the metacognitive
or motivational aspects explicitly. Mostly, only cognitive aspects were mentioned
unambiguously.
Discussion
In an attempt to bridge the gap between diagnosis and intervention, which is nowadays
a reasonable quest in educational practice, this study examined the contribution of
dynamic assessment to this issue. Dynamic assessment is meant to be a promising tool
for estimating capacities of at-risk children and for distinguishing appropriately between
CHAPTER 2
50
children with learning disorders and children having learning delays (as a consequence of
their learning environments and former learning experiences). To date this is still one of the
challenges of educational practice. In spite of the advantages of dynamic testing procedures
they are, still, too rarely employed in practice due to the robustness (in research as well as in
practice) of traditional intelligence tests. However, a lot of dynamic assessment procedures
have been developed, demonstrating valuable psychometric properties and proving to be of
added value in the assessment of capacities of at-risk children compared to static intelligence
measures. Overall, they show a high predictive validity, and four studies (Elleman et al.,
2011; Swanson, 2011; Swanson & Howard, 2005; Tiekstra et al., 2009) even demonstrate
significant added predictive validity on top of IQ.
Over the last two years the contribution of articles about dynamic assessment to scientific
journals has been increasing: more than one-third of the articles found for this review was
published in 2010 and 2011. This trend also sustains in 2013, but was beyond the scope of
this article. The increased interest in dynamic assessment makes the necessity for review
studies on this subject evident.
Since dynamic assessment implies a teaching component during testing, the question
arose whether these dynamic assessment procedures provide valuable information for the
construction of interventions following the test, and could, thus, bridge the gap between
diagnosis and intervention. The outcomes of 29 dynamic instruments were examined with
regard to their consequential validity and presented at a continuous range, diverging from
proximal consequential validity to distal consequential validity. This means that outcomes
vary from only raw scores to more qualitative information. Although one could state that
qualitative observations during testing do provide a lot of information for intervention, it
appeared that none of the instruments resulted in explicit clues for practice. Next to raw
or standardised scores, there are tests that provide an indication of the level of help the
child needs. This can be seen as a more advanced type of consequential validity, since it
provides more information about the capacities of the child. Nevertheless, explicit clues
for practice are still unknown.
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
51
2
This outcome of the review study underlines the fact that, although dynamic
assessment procedures have potential, these procedures do not contribute to bridging
the gap between diagnosis and intervention in their current formats. The remaining
quest for bridging effects of dynamic assessment was also highlighted by Haywood
(2012). Dynamic assessment procedures that result in raw scores only do contribute
to proximal consequential validity, since an opportunity to learn is provided during
testing. However, in these cases, the tests will particularly be used to better estimate
learning capacities as opposed to traditional intelligence tests, and not as indicative
for the design of interventions or adaptations in the instructional process afterwards.
Hence, distal consequential validity is not warranted in current dynamic assessment
procedures. It should be mentioned, though, that it is likely that some of the articles
describe instruments that have been developed for research purposes only, and not (yet)
for psycho-educational practices. And, thus, do not describe consequences for practice
explicitly. This aspect could have caused some biased results.
Despite the present absence of distal consequential validity in procedures, attempts to
bridging effects are being made sporadically resulting in some qualitative information
for practice. However, the move to fully distal consequential valid procedures would not
be too far, and is actually near. The teaching component during testing provides the
opportunity to carefully administering the actions that have been taken during learning
phases. The analysis of actions of both examiner and examinee could be divided in, for
example, different categories (i.e., cognitive, metacognitive, and motivational). By explicitly
mentioning these actions during the test administration, outcomes will provide information
for practice. Outcomes of the testing procedures, then, not only indicate a score of learning
potential (or raw score), but also enhance systematic cues for setting up IEPs.
Therefore, in addition to outcomes, learning phases of each dynamic assessment procedure
have been scrutinized in this review. The content, i.e., the aims of examiner’s activities,
has been divided in cognitive, metacognitive and motivational strategies. Unfortunately, it
turned out that this information was barely explicitly mentioned in the articles. Nonetheless,
CHAPTER 2
52
a striking conclusion can be made: most of the tests apply cognitive and/or metacognitive
strategies during the learning phase, but motivational factors never played a major role
explicitly. As has been noted before by Borkowski, Carothers, Howard, Schatz, and Farris
(2007), self-regulation strategies should be integrated in dynamic assessment. This review
shows that progress still has to be made in this area. Only recently, Teo and Roodenburg
(2013) demonstrated the additive effects of including self-regulation strategies during an
assessment procedure. Moreover, in some dynamic tests general feedback is given (De Beer,
2010a; Gillam et al., 2011; Gummersall & Strong, 1999; Hamers et al., 1996; Jeltova et
al., 2011; Sternberg et al., 2002; Tzuriel, 2000), but the kind and type of feedback is not
adequately mentioned or registered. The extent to which examinees benefit from feedback
is an important factor in dynamic assessment. Woide, Beckmann, Elliott and Guthke (2005)
demonstrated that motivational, affective and non-intellective factors influence the effects
of feedback and that feedback increases the capacity to evaluate oneself. Although one of
the promising aspects of dynamic assessment approaches is the attention for non-intellective
factors (Lidz & Macrine, 2001), this study demonstrated that there is no explicit focus on
these factors during learning phases. This is an interesting finding, since it is generally known
that motivational factors interact with the learning process (e.g., Ryba, 1998) and test results
(Meijer, 2001). Moreover, one of the groundings of dynamic assessment is that during
testing social interaction takes place in order to provide a more motivating testing
situation, which is less biased to e.g. test anxiety.
These aspects should, thus, play a role during learning phases. The type of feedback
that has been given should be made explicit in the test result which will provide useful
information for the instructional design in the classroom or for the design of individualized
interventions. Motivating strategies of the examiner, like effort stimulation or explicit
feedback on the performance of the task, have been demonstrated to result in different
impacts on aspects of learning, such as task interest and self-efficacy respectively (Perry,
Turner, & Meyer, 2006; Prince, Minnaert, & Opdenakker, 2014)
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
53
2
These suggestions also underline the interdependency of cognitive, metacognitive
and motivational strategies during learning processes. These aspects are indispensably
related to each other, and thus, each of these aspects should play a prominent role
during test administration, and, consequentially in the test result. Only in this way,
an ecologically valid contribution to the consequential validity of dynamic assessment
procedures can be made.
Of course, this relates to questions about the test duration, domain specificity and
age group of the tests. Each of these variables has its implications for the question
of consequential validity. However, the overall lack of motivational components and
explicitness of learning phases, underlines the importance of qualitative reports about
the interplay between examiner and examinee during test administration, in addition to
raw scores if applicable. This implies that examiners should be reflective by nature on
both their own activities and the activities of the examinee.
There are, however, some important limitations while interpreting the outcomes of this
review. Firstly, the literature search has been carried out in a database (Web of Science)
which only includes peer reviewed journal articles. Consequently, articles which have
been published in non SSCI journals are not warranted, as well as book chapters and
dissertations. This implies a certain restriction to the scope of articles included in this
review.
Secondly, there is no consensus about the definition of dynamic assessment in the
field. Moreover, different approaches exist (standardised vs. clinical), which may cause
confusion in research as well as in practice. Next to the robustness of traditional tests,
this could contribute to the suppression of dynamic procedures in practice. Therefore
it is suggested to use clear distinctions between dynamic assessment and intervention
(see also Hessels-Schlatter & Hessels, 2009). In addition to these authors, we would like
to maintain these distinctions to promote a clearer concept of dynamic assessment.
Subsequently, in this article, the term dynamic assessment has been defined as a
CHAPTER 2
54
dynamic procedure with a standardised approach in which a learning phase is included
in testing. This form of dynamic assessment is comparable to static testing. To our
opinion, dynamic assessment does not contain an intervention phase expanded to
multiple sessions during several weeks. In the latter, one is testing the effectiveness of an
intervention, instead of making an evaluation of a child’s capacities.
Conclusion
To conclude, dynamic assessment remains a promising tool to bridge the gap between
diagnosis and intervention, but there is still room for improvement. Due to a learning phase
included during the testing procedure, qualitative information can be revealed. This, in turn,
provides an estimation and evaluation of the types of information that are essential to the
child in order to complete a task. Eventually, this could contribute to the development of an
adaptive intervention, which will provoke better learning results of the child.
Implications for practice. In order to establish this, however, the focus on consequences of
assessment for educational practice should hold a more prominent place in the development
of dynamic assessment procedures in future. There is a need for more explicitness of learning
phases and types of feedback, in order to construct student-tailored interventions. Moreover,
motivational factors should be highlighted during learning phases. Students encountering
learning difficulties generally have experienced a lot of negative emotions and failure, and
dynamic assessment procedures are the more promising if they could anticipate on these
factors pre-eminently. If the learning phase is administered and documented carefully, the
consequential validity will be more warranted, resulting in explicit clues for practice.
Additionally, the authors would like to make a plea for a more prominent place of
consequential validity in test evaluation reports. When evaluating the quality of tests, it is
important to focus on consequences for practice, next to internal validity criteria.
A REVIEW SCRUTINISING THE CONSEQUENTIAL VALIDITY OF DYNAMIC ASSESSMENT
55
2
From a researcher’s point of view, future research should focus on the bridging effect of
dynamic assessment procedures to intervention practices. This could be done, e.g., by
analysing qualitative (video) observations and focusing on motivational components during
assessment procedures and consequences for educational practice. While investigating the
distal consequential validity of dynamic assessment procedures, research contributes to
practice, which is an important aim of science.
If the gap between diagnosis and intervention can be bridged, the quality of psycho-
educational practices will be elevated. Since a careful administration of the learning
phase provides useful information for interventions or instructional processes in the
classroom, professional actions will be more grounded on empirical results. Educational
psychologists play a prominent role in this process, being the core communicator
between test result and interventions. If tests provide helpful clues, next to raw scores,
the quality of interventions and instructions will increase. Only in this way the learning
outcomes of students will be optimized.
“You can never solve a problem with the same kind of thinking that created the problem in the first place” (A. Einstein)
Chapter 3Can didactic resistance be measured by using
the Response-to-Instruction model?
A pilot study
This chapter is based on:Tiekstra, M., Prince, A., & Minnaert, A. (in prep.).
Can didactic resistance be measured by using the Response-to-Instruction model? A pilot study.
CHAPTER 3
58
Abstract
Didactic resistance is viewed as an important criterion in the definition of learning
disabilities. This study explored whether the Response-to-Instruction (RTI) model can be
an instrument to measure didactic resistance. Using Curriculum Based Measurements
(CBMs) the interaction between teacher and students can be monitored. This qualitative
multi-method study (N = 11 schools) demonstrated that in order to be an instrument to
assess didactic resistance, several aspects need to be taken into account before, after and
during the process of implementing the RTI model. Data triangulation, illustrated by
quotations from the field, revealed the interrelatedness of process and content on the one
hand, and change of beliefs and behaviours of all educational staff members on the other
hand. The results showed that there is an urgent need for a dynamic view on learning
and instruction when regarding the RTI model as an instrument to measure didactic
resistance. Besides, didactic resistance should be measured at both student and teacher
level. Theoretical and practical implications will be discussed.
DIDACTIC RESISTANCE AND THE RTI MODEL
59
3
Introduction
The diagnosis of learning disabilities has been a major issue in debates over the years
(Hallahan & Mock, 2003). Not only the way to define, but mainly how to diagnose and how
to respond to the diagnosis of a learning disability in practice is still surrounded by vagueness
(Büttner & Hasselhorn, 2011). Traditionally, the discrepancy between IQ and actual (school)
achievement was the decisive factor. In this, generally, a child is identified as learning disabled
when scores on achievement tests at school show a large discrepancy with a child’s IQ (as
a result of a standardised intelligence test). A child which does not perform according to its
potential (operationalised as IQ score) should be identified as learning disabled (Hallahan,
Kauffman, & Pullen, 2012, p.140). This approach of learning disabilities has been criticised
for inaccurate judgments (Francis et al., 2005; Stuebing, Barth, Molfese, Weiss, & Fletcher,
2009) and can be seen as a “wait-to-fail” model (Hallahan et al, 2012). In an attempt to
overcome these difficulties with identification of learning disabilities and to provide more
early identification, the Response to Instruction model (RTI) has been developed in the
United States. According to Fuchs, Fuchs, and Vaughn (2008) the discrepancy between
estimated response to instruction and actual response to instruction is examined in RTI.
Progress and results of each student are monitored systematically, which provides an insight
to specific difficulties of the individual child. In this, teachers are able to estimate whether
falling performances of a student are due to either a deficit in the student or inappropriate
(i.e., not tailored to student’s capacities) instruction of the teacher. Although criticised by
several authors (e.g., Kavale, 2005; Reynolds & Shaywitz, 2009) because of its lack of
demonstrated empirical effectiveness, the RTI model can be seen as an alternative model
to identify learning disabilities. But, in line with Kavale (2005) together with Spaulding
and Beam (Kavale et al., 2009, p.39) the model does not provide a definition of learning
disabilities. However, the RTI model can officially be used to identify children as learning
disabled in the United States since 2004.
Consequently, the definition of learning disabilities needs to be adapted, as is the case
in the new edition of the Diagnostic and Statistical Manual of Mental Disorders (5th
CHAPTER 3
60
ed.; DSM-5; American Psychiatric Association, 2013). More attention will be drawn to
clinical based data instead of the discrepancy criterion. It remains, however, unclear
what the definition is exactly (Scanlon, 2013) and how it will be applied in practice.
Generally, a somewhat broader definition of the IQ-achievement discrepancy approach
is used to diagnose learning disabilities. This definition encompasses three criteria for
identifying learning disabilities (see also Tannock, 2013). Firstly, student’s performances
are below what is expected. In this, a measure of IQ is included, but in contrast to what
was mentioned before, background variables are taken into account when interpreting
IQ scores. Secondly, a criterion of didactic resistance is included. This means that
students who do not profit from (tailored) instruction show a certain persistence of
difficulties. And thirdly, the criterion of exclusivity, problems are not caused by other
impairments like hearing or visual impairments. This broader definition, which
has been adopted by the DSM-5, thus, reflects an extension to the endorsement of
individual differences in one classroom instead of general, one-pattern-fits-all, group
wise education. It remains, however, unclear how the above-mentioned three criteria
should be measured (Cavendish, 2013).
Didactic resistanceAs mentioned, the construct didactic resistance is one of the three main elements in
the determination of learning disabilities. This is an interesting construct, however,
little research has been published about the interpretation and measurement of this
construct as was revealed by a recent search in the EBSCO host database. It seems that
a grounding for a thorough understanding of this construct is lacking, as is the empirical
grounding for the psychometric characteristics: how should we interpret and measure
didactic resistance? In practice, students are regarded as didactically resistant if they are
unresponsive to quality instruction, i.e., the problems persist despite tailored teaching.
As such, the problem is stated at the student level, as are the measured outcomes. Since
an indication of the quality of the teaching provided remains unmentioned, this seems
a rather limited interpretation of the concept of didactic resistance. To illustrate, one
could state that didactic resistance not only implies a deficit in the student, but also
DIDACTIC RESISTANCE AND THE RTI MODEL
61
3
a deficit in the teacher (or the teaching method), since the teacher is not capable to
adapt instruction to a level that fits the level of the student. In practice, regrettably,
instruction is not as dynamically interpreted or implemented as assumed in the criterion
on didactic resistance. Instruction, after all, involves multiple partners, as is emphasised
by the theory of Vygotsky (1978), and should be regarded from a dynamic perspective in
which teacher and student play reciprocal roles (Steenbeek, Jansen, & Van Geert, 2012).
Timperly and Parr (2009) underlined the importance of a rather dynamic view on
instruction, as was demonstrated by their research on the influence of different teaching
strategies on students’ learning. Additionally, from a more general point of view, Chak
(2001) emphasised the importance of reflection upon adults’ sensitivity to children’s
learning in optimising one’s zone of proximal development. This applies for teacher-
student interactions as well. Hence, didactic resistance should not only be measured at
student’s level, but also at teacher’s level.
According to the above-mentioned, the criterion of didactic resistance could be
measured during a RTI approach. Due to regular adaptation of their instruction to
the specific needs of their students, teachers are able to monitor progress and highlight
problems of each student. Moreover, the careful progress monitoring provides an
insight in needs and, consequently, tailored instruction can be developed. In this way,
teachers should be more aware of their own teaching practices on students’ learning.
In contrast to intervention programs aiming at individual outcomes, for example a
certain instruction method to deal with learning difficulties, the RTI model can be seen
as an organisational model. It includes an adaptation of administering and teaching,
which, in turn, demands an adaptation of the whole organisation (in this: school). The
core assumption of the RTI model is that every child reacts differently to whole-class
teaching. Due to progress monitoring of each student, the extent to which each student
is sensitive to instruction can be estimated. When a child does not show enough progress
(compared to the mean progress of the group) this child is offered a kind of intervention.
The model consists of three levels of instruction (Fuchs & Fuchs, 2006). In the first tier,
students are offered general evidence-based classroom wide instruction. Students that
CHAPTER 3
62
do not show enough progress compared to their classmates receive extra instruction in
small groups several times a week; this is tier two instruction. If extra instruction does
not provide enough progress, an individualised route is set up for these students (tier
three). Other RTI models exist, though, with variations in the number of instruction-
levels. These models, however, are still based on the core model of Fuchs and Fuchs
(Kavale & Spaulding, 2008). The progress-monitoring is carried out with curriculum-
based measurements (CBMs; Espin, Wallace, Lembke, Campbell, & Long, 2010).
CBMs are closely related to the curriculum taught, and, therefore, provide insight into a
student’s response to current instruction in a specific domain. Due to several aspects of
the model, such as progress monitoring by CBMs and the differentiation of instruction
in tiers, the model delivers valuable information for planning instruction.
In practice, also when working with the RTI model, didactic resistance is often
interpreted as a static feature of the student. However, it should rather be regarded as
a dynamic concept of the interaction between student and teacher. In the RTI model,
theoretically, the interaction between teacher and student is monitored by the CBMs.
Since the CBMs are closely related to the current instruction, effects of the instruction
on student’s outcomes can be measured. Nevertheless, the instruction component in
the interpretation of the CBMs is often ignored and an emphasis is placed upon the
outcomes of the student solely. This results in, again, a rather static interpretation, and
resembles the IQ-discrepancy approach in identifying learning disabilities. Therefore,
contrary to the more standardised approach as favoured by many researchers, CBM
outcomes should not only be interpreted quantitatively, but also qualitatively. The
errors made in the CBM should not only be counted, but also analysed by the teacher
to gain insight in the mistakes that were made. Only in this way quality teaching can be
warranted, because instruction is tailored to a student’s needs.
Process of implementing the RTI modelIn theory, the RTI model seems a very promising instrument to measure didactic
resistance. However, when implementing a model to practice multiple factors come into
DIDACTIC RESISTANCE AND THE RTI MODEL
63
3
play (for an overview see Prince, Tiekstra, & Minnaert, 2014) that influence the outcomes
of the implementation process. A lot of (US) literature describes the core principles,
ideas, features, promises of, and experiences with the RTI model. It is still, however,
difficult to find literature about the effectiveness of implementation of the model (Fuchs
& Vaughn, 2012). This results in different models and different operationalisation of
each model in each context, which contributes to confusion surrounding RTI (Fuchs,
Mock, Morgan, & Young, 2003, p.159). Research often focuses only at one aspect of the
RTI model and the enormous amount of literature is difficult to overview in order to
get guidelines for practitioners with respect to its implementation (Jones & Ball, 2012).
Information for teachers on how to effectuate differentiated instruction in tiers remains
generalised and ambiguous (Fuchs & Vaughn, 2012). Moreover, specific information
about training or coaching during implementation of the RTI model is hard to find.
Ikeda (2012) stressed the fact that the focus is rather on decision making processes in the
model than on the content and effectuation of support for teachers.
The way information is implemented plays an determining role, next to the content
of what is implemented (see also Prince et al., 2014). In their qualitative case study,
Datnow and Castellano (2000) described the process of implementation of a school
wide program and illustrated the different outcomes between teachers. They touched
upon this important distinction between process of implementation and the content of the
model that is implemented. Generally it is underlined that engagement of teachers in
developing the program favours the implementation process. Although this was not the
case in Datnow and Castellano’s research, teachers still implemented the program since
they appreciated its content.
Furthermore, a main aim of implementation lies in the change of behaviour and
belief of participants. The Theory of Planned Behavior (Ajzen, 2005) describes the
interrelatedness of changing belief and behaviour. Beliefs have their influences on
intentions that precede behaviour. More particular to the classroom practice, Bandura
(1993) highlighted the importance of teachers’ beliefs and the influence of these beliefs
CHAPTER 3
64
on their behaviour in the classroom and quality of their teaching: “teachers . . . beset
by self-doubts construct classroom environments that are likely to undermine students’
sense of self-efficacy and cognitive development” (p. 140). Teachers’ beliefs, in turn, are
affected by experiences. This illustrates the interrelatedness of belief and behaviour
when implementing new models into the classroom. Moreover, as underlined by Van
de Ven, Polley, Garud, and Venkaraman (1999), implementation is a cyclic process. The
above-mentioned factors have an impact on the methodology used for this study. In the
Method section, therefore, the used methodology will be carefully described.
Aim of the present studyWhilst theoretically grounded the RTI model seems to be a promising instrument to
measure didactic resistance, this article deals with the issue whether the RTI model
can function as an instrument to measure didactic resistance in practice. Thence, the
implementation of the RTI model will be studied carefully, as the way of implementation
will have a major influence on the outcomes of the study. As was mentioned before, the
RTI model can be regarded as an organisational model. This has several consequences
for the implementation of the model. One could state that it can only be implemented
school wide and on all levels, from teacher to educational support staff. Since the goal
of this study is to provide detailed qualitative information about the possibility to assess
didactic resistance, rather than information about the implementation, the model has
been implemented at one academic grade only.
Method
Participants and settingIn 2011 a pilot study has been conducted which attempted to implement the RTI
model partially at a cluster of schools in the Netherlands. Care for children having
learning problems is organised in a stratified manner in the Netherlands, which implies
(theoretically) clear roles and responsibilities for all involved. In practice, decisions
DIDACTIC RESISTANCE AND THE RTI MODEL
65
3
are usually made democratically. In this particular cluster of schools the care process
involved one school psychologist who led 11 senior coaches, of whom each worked
at multiple schools. These senior coaches supported remedial teachers (one remedial
teacher per school). Remedial teachers supported the teacher and supplied information
and advice when the teacher asked for it. It was decided that one school per senior coach
participated in the study, resulting in a total of 11 schools that participated initially.
In the region where this cluster was situated, people generally speak a strong dialect,
which might influence students’ outcomes. The cluster of schools, however, can be
regarded as an ordinary cluster of public schools (comprising religious and non-religious
schools), with a non-deviant proportion of minority students. Participation to the pilot
study was voluntary. A notable fact was that educational staff (all participants) defined
themselves as being progressive teachers, since they attended professional courses
frequently. Attending professional courses is one of the requirements of inspectorate
and government. Next to this, quality of schools is assessed by students’ outcomes on
nationwide tests and visits in the classrooms by inspectorates.
InstrumentsThis study had a multi-method qualitative design. Questionnaires, learning reports,
interviews, and observations were qualitatively interpreted to provide data and are
described in detail below.
Questionnaires were constructed for senior coaches, remedial teachers and teachers.
The questionnaire for senior coaches and remedial teachers (SQ) was based on the
questionnaires that Meijer and Tijhaar (1995) developed in their studies around teacher
consultation and professionalisation. It described six phases in the implementation
process of the RTI model; introduction, problem identification, analysis of problems,
solution seeking, implementing solutions, evaluation. Participants had to rate on a
5-point scale to what extent they felt competent about each phase. The questionnaire
for teachers (TQ) differed, since items were more related to the classroom level, such as
CHAPTER 3
66
“at the beginning of the school year, I do a screening to assess each student’s reading
level.” Ratings were also on a 5-point scale. The aim of the questionnaire was to get an
idea of the current situation (with regard to RTI) at the schools and whether participants
knew how to translate knowledge.
The learning report (LR) consisted of five open questions. The learning report was
administered to senior coaches and remedial teachers during one training session to
detect possible bottlenecks in the implementation process. Moreover, we wanted some
insight in the opinions and experiences with the model until that moment.
Interviews took place with some of the senior coaches (SI), remedial teachers (SI) and
teachers (TI). Interviews were unstructured, but focused at gathering personalised
information about the implementation process and the strengths and weaknesses of the
RTI model. Some took place during training sessions, others in the schools, others at the
moment that a school dropped out of the project.
The first author (an educational psychologist) observed every training session and
additional meeting (OT) in order to monitor the implementation process closely.
She also visited three schools to observe how the RTI model was implemented. The
observations in the classroom (OC) took approximately 30 minutes in which the teacher
had planned reading instruction. In this way, some insight could be provided in how the
different tiers were organised in one class and how the RTI model was implemented
on a classroom level. The observations, thus, were only completed at the schools that
finished the project.
ProcedureThe planning was to provide participants with a training of two consecutive days at
the start of the academic year. Furthermore, several additional meetings were planned
during the project. For this, a training of several days was developed in order to provide
senior coaches and remedial teachers with knowledge about the RTI model and tools
DIDACTIC RESISTANCE AND THE RTI MODEL
67
3
to translate this knowledge to their teachers. During the six months project, every four
weeks a meeting was organized to coach, assist and encourage the senior coaches and
remedial teachers. In this way, the implementation process could be monitored carefully.
The questionnaire (SQ) was administered prior to the first (and after the last) training
session in order to measure the participants’ knowledge of the RTI model. Teachers’
questionnaires were distributed via remedial teachers and were collected between
second and third additional meeting. Next to the questionnaires, the learning report
was administered to remedial teachers in order to detect potential struggles during the
course of the project.
Data collection took place at certain fixed points in time (T1 to T6) that were planned
beforehand, but also in between (not planned). An overview is presented in Table 1.
Table 1 | Sequence of data collection during the pilot study
Instrument T1 T2 T3 T4 T5 T6
SQ x x
TQ x x
LR x
SI x x
TI x
OT x x x x x x
OC x x
Note. T1: Start training (2 days); T2: additional meeting 1; T3: additional meeting 2; T4: additional meeting 3; T5: additional meeting 4; T6: additional meeting 5
CHAPTER 3
68
Information input: trainingTraining was provided for senior coaches and remedial teachers, due to pragmatic reasons:
since it already took a great effort to assemble them, it would have been even more difficult
to assemble all teachers. It also resembles the way in which care is organised in schools in the
Netherlands: these people support teachers and provide them with new insights during the
school year. The goal of the training was to provide knowledge about the RTI model and
to provide tools and skills to facilitate the implementation process. The participants needed
to translate their knowledge to their teachers, which, again, resembles the normal process
in the Netherlands. Training took two days prior to the whole RTI project, and had five
additional meetings.
The RTI-training was developed by commercial training experts in collaboration with an
educational psychologist (the first author) to guard scientific input and to watch over the
implementation process as a whole. The RTI model is an abstract model and one of the
difficulties in developing the training was translating the information to such a practical
level that participants would understand the core ideas of the RTI model. This implied
the selection of one grade in which the model would be implemented instead of a school
wide implementation (multiple grades), since constructing CBMs for all grades would take
too much time. Grade 2 was selected to implement the RTI model, since severe learning
difficulties can be detected at these ages.
CBMs were constructed in advance by the trainers, to relieve teachers of another task. Once
a week, students were administered individually a CBM by their teacher or support staff
(i.e., remedial teacher or classroom assistant). The CBM consisted of a one-minute reading
out loud test and each week another part of the same story (with reading level appropriate
for end grade 2) was used. The number of correct words could then be calculated and
the errors made could be identified. A quantitative as well as qualitative analysis of the
errors was emphasised. An Excel-tool was developed to register progress and calculate dual
discrepancy.
DIDACTIC RESISTANCE AND THE RTI MODEL
69
3
Data analysisA qualitative data analysis was used, since the method should be sensitive to contextual and
situational characteristics in order to determine implementation effectiveness. In line with
the principles of Interpretative Phenomenological Analysis, data analysis took place in two
stages (Bazeley, 2013). First, data have been categorised with help of the ATLAS.ti program
(version 6.2) in order to establish thematic analysis. Second, from a more phenomenological
point of view, the researcher (first author) interpreted the data subjectively and in line with
participants’ experiences.
For the trustworthiness of data (Lincoln & Guba, 1985) the data has been triangulated
from the multiple data sources mentioned above. In each step of the data analysis peer
debriefing was established by the second and third author to obtain objectivity and reliability.
Subjectivity of first author, however, remains important, and has been highlighted in the
Results section when needed (Altheide & Johnson, 2011).
Results
In this section the researcher’s perception of the data and data of the researcher’s perception
will be listed with an emphasis on data and perceptions that are needed to answer the
research question Is the RTI model an instrument to measure didactic resistance? The results have
been structured according to the chronological sequence of events during the project.
Therefore, a division has been made between initial training, the five additional meetings
as follow up sessions, and eventually, observations at the schools that completed the project.
Each description of these phases comprises a short description of core events, highlighted
by quotations from the field, and the authors’ reflections based upon these. Generally, these
reflections relate to the division between implementation process and content of the RTI model
and the division between change in belief (point of view) and behaviour which appeared to be
core elements as revealed by the thematic analysis with ATLAS.ti.
CHAPTER 3
70
Initial trainingIt can be stated that the content of the RTI model and the process of implementation are
two separate factors in this pilot study. As pictured below, the training needed to be adapted
to the specific needs and situation of the targeted schools. This is a rather common situation
while implementing interventions. However, after the concessions to participants’ needs
(principally dictated by logistical situations), the implementation process of the RTI model
still appeared to be problematic during the follow ups.
Observations during the initial training revealed that participants did not agree with a new
screening at the beginning of the school year as several of them remarked: “this takes too
much time, and we have already done this at the end of previous grade”. Due to this refusal
of participants, the training reached at an impasse. Although the questionnaires showed
that there was little knowledge about the RTI model beforehand (scores 1 and 2, out of
5), there was no need for knowledge about the RTI model during the training session. In
order to continue the project, the implementation of the RTI model had to be adapted to
some requirements of the participants. The first condition of the participants was “no extra
screening at the beginning of the school year”. Secondly, “differentiation of instruction right
at the start of the school year instead of first 6 weeks of same instruction to all students”.
The researcher states that, as a consequence, the idea of differentiating between responding
and non-responding students to the current teacher’s instruction could not be determined.
However, she hoped that several weeks of implementation and working with the model
would change participants’ view and, thus, understanding of this idea.
After their first refusal and the concessions towards their requirements, some participants
started to see the additive value of the RTI model in such way that poor readers would
be detected earlier in the instruction process. These participants highlighted the need for
support of school administrators when implementing the model. The researcher thinks that
this illustrates rather a focus on implementation than on content of the training, since this
aspect never played an explicit role during the rest of the project. Although, 8 schools agreed
to continue the pilot study after this impasse, the researcher still had some doubts whether
DIDACTIC RESISTANCE AND THE RTI MODEL
71
3
the participants shared the view about the RTI model as was aimed for at the start of the
training.
Follow upsIn addition to the two days of initial training 5 meetings were planned in which, originally,
key moments during the process could be discussed in order to monitor the implementation
process. It appeared, however, that these meetings needed to be used for more explanations
and information input to the participants. Apparently, the core principles of the RTI model
were, still, not understood.
Process of implementation: unchanged behaviour and unchanged belief.
During the follow ups, serious discussions emerged about the general instruction and
screening of all students, and even caused some drop outs of the project. The researcher
thinks this is a remarkable fact, since decisions and agreements had been made about these
points during initial training. By the third additional meeting, already three schools had
refrained from participating in the project, of which only one school informed the trainers
officially. One of the reasons this school mentioned while quitting the project was: “we did
not want to stop our own instruction and intervention method”, and it appeared that after
four weeks of tier 1 instruction (and CBMs) they had already stopped the whole project.
Several weeks later, after additional meeting 4, another two schools dropped out, of which
one mentioned their problems with the general classroom instruction: “we’ll never going to
abstain children from extra instruction again”. According to the researcher, this reflected the
severe resistance to change and a certain fear to adapt their current instructions to outcomes
of the CBMs. It is likely that participants stick to their own (safe) intervention method. These
participants dropped out of the implementation process since they did not (want to) change
their behaviour. The quotations illustrate their unchanged beliefs about the role of the teacher
in didactic resistance. Since they continue with their current differentiation in reading instruction
and extra interventions, based on outcomes of the previous grade with instruction from another
teacher, it can be stated that they had a rather static view of didactic resistance.
CHAPTER 3
72
Process of implementation: changed behaviour but unchanged belief. First things
to implement were the CBMs, i.e. progress monitoring. Due to the precise instruction of one of
the trainers, the researcher expected that administration of CBMs could be effectuated easily.
However, a returning point in discussions with participants during the additional meetings and
interviews was the planning of the administration and the time it took. These issues relate to
the implementation process, rather than discussion regarding the content of the RTI model.
For instance, one participant remarked that “administering CBMs takes too much time, what
is the added value of this? We already knew which students needed extra instruction.” The
administration of the CBMs, appeared, thus, a rather negative assignment than an information
tool for teachers to adapt their instruction to the specific needs of their students. Later on in
the project participants became more positive about the CBMs, as the graph provided by the
(CBM-) tool was rated as “useful” and “motivating for students as well as teachers” according
to the learning reports and interviews. For the researcher it remained questionable whether the
discussions can be attributed to the real negative experiences in practice with the CBMs or rather
to a resistance to change or refusal of the participants. These participants implemented the
CBMs in their classroom practice, and changed their behaviour. The hesitations of participants
and quotations, however, reflect a lack of dynamic view on didactic resistance, since, again, the
decisions based on the previous grade remain very important. The changes in behaviour are more
related to the process of implementation, rather than to the content of the RTI model. These
behavioural changes will be highlighted in the next section, whereas belief remains unchanged.
RTI content: changed behaviour but unchanged belief. Aspects of the RTI model
that are crucial in determining whether it can be used for measuring didactic resistance
needed to be examined in detail. With regard to the interpretation of CBM outcomes, none
of the participants had problems, as was demonstrated in the learning reports and in the
observations during training. The researcher was surprised by this fact, since she estimated
that the selection moments would be crucial moments, and maybe might cause problems
as she experienced during the development of the training. As demonstrated in interviews
and observations, the interpretation of the CBM outcomes by the participants, however,
focused mainly on students. Besides, although never explicitly mentioned, the connection
DIDACTIC RESISTANCE AND THE RTI MODEL
73
3
to teacher’s instruction appeared to be very difficult. Observations, learning reports and
interviews revealed that it was unclear which interventions one should use, when to effectuate
these interventions next to the current (tier 1) instruction, and how to link these interventions
to the errors made in the CBMs. Participants were, for example, unsure about the content
of tier 2. They expected the RTI model to provide interventions, but the model rather aims
at activating the teacher to search for appropriate interventions that suit the outcomes of
the CBMs. The mentioned examples reflect a change in behaviour, but unchanged view on
didactic resistance. This is best illustrated by the focus on students’ outcomes solely instead
of the link to quality of instruction. The researcher thinks that participants (senior coaches
and remedial teachers), perhaps, tried to avoid any accountability on behalf of their teachers
or themselves, as a kind of self-protection. The change in belief, still, stays behind. Despite
these negative experiences, there were still some schools that continued the project and,
even, finished it in a very promising way.
Observations in practice Finally, three schools completed the whole project. These three schools have been paid
special attention to, manifesting in observations in the classrooms, and extra interviews with
participants in the field, i.e., teachers and remedial teachers. One school implemented the
RTI model, but still under some resistance. In the classroom it was observed that the teacher
had difficulties in combining the current reading method with different interventions, and,
thus differentiating between students. Afterwards the remedial teacher mentioned that she
did not see the value of the RTI model: “Our current reading method functions well, why
should we replace something that is going well? Why do we need to do this pilot study?” As
mentioned before, maybe this resistance has to do with the different implementation levels:
if the remedial teacher does not understand the core message of the RTI model, then, it will
not be communicated to the teacher and it will not be executed in the field properly. Bearing
the research question in mind, the RTI model does not function as an instrument to measure
didactic resistance in this school.
CHAPTER 3
74
Another school implemented the RTI model properly, but according to the observations
and interviews it can be noticed that there was only a change in behaviour and not yet
a change in belief. The quotation “we are very positive about the RTI model, since
students are more engaged during instruction”, indicates that outcomes are focused at
a student level, rather than at student and teacher level. The CBMs were administered
and interpreted by a classroom assistant. The remedial teacher made decisions, while
the teacher looked for interventions that match these outcomes herself. But, to what
extent are these outcomes representative for the instruction the teacher gives? In order
to provide real tailor-made teaching, the communication between these partners should
be good and regular. In this situation, according to the research question, the RTI model
can be a tool for didactic resistance, but communication should be open and honest,
in order to achieve congruence between CBMs interpretation and tailored instruction.
The third school was an excellent example of a well-implemented RTI model. The
teacher interpreted the CBMs herself, and was supported by the remedial teacher in
searching for appropriate interventions. Although, this had caused some uncertainty in
the teacher at first, she mentioned afterwards that she felt more capable in providing
instruction that fits her students. The teacher reported that she was more aware of what
students learn and how she effectuates her instruction. Furthermore she remarked “I
leave the instruction method in order to search for instruction that the specific student
fits.” Next to these outcomes on teacher level, the teacher observed “a large progress in
students and a greater engagement during instruction.” The teacher concluded “I am
going to use the graphical representations of student’s progress in parental consultations
and I will continue working with the model in future. I am going to develop CBMs by
myself.” This demonstrates a change in behaviour and belief, since the teacher related
the effects of her instruction to the progress of students. In this example, the RTI model
is used as an instrument to measure didactic resistance.
DIDACTIC RESISTANCE AND THE RTI MODEL
75
3
Discussion
This study investigated whether the RTI model could be an instrument to assess
a student’s didactic resistance. It appeared that it can be a valuable tool, if several
conditions are met. These conditions relate to the implementation of the RTI model,
and focus especially at the dynamic view which is needed in order to estimate a student’s
didactic resistance. As demonstrated by this pilot study, the implementation and the
content of the RTI model are two separate factors, but are intricately related. For
example, the schools that dropped out of the project did not support or understand
the ideas of the RTI model, but also showed a high resistance to change which occurs
frequently during implementation of any kind of new model or method. Next to these
factors, the change of behaviour and change of belief are important aspects to take
into account when implementing the RTI model. Belief and behavioural change can
be seen as separate factors, but at the same time are interrelated. This can best be
illustrated by great differences in the group of schools that finished the project. The
schools that implemented the RTI model only superficially did change their behaviour,
but did not change their belief in line with the RTI model. These participants still had
a strong focus on students’ outcomes and experienced great problems in linking the
CBM outcomes to their instruction. The change of behaviour and belief is of utmost
importance when implementing the RTI model, as demonstrated by the third school,
in which the teacher recognised her role during the teaching process and tried to reflect
on her teaching. In this case only, the RTI model can be used as an instrument to
assess didactic resistance. The interrelatedness between change of belief and change
of behaviour demands, however, further investigation. Longitudinal studies should
focus on the development and nature of the relationship between behaviour and belief
change. This is important when translating abstract models to practical skills and should
be taken into account when developing (RTI) training procedures.
The pilot study showed the necessity of a dynamic view on instruction and learning
when implementing the RTI model. Teachers can only adapt their instruction if they
CHAPTER 3
76
recognise their influences on learning of students. It is important not to rely only on
outcomes of standardised tests when identifying learning disabilities. The amount and
quality of teaching a student receives and has received in the past is a more determining
factor. After all, from a Vygotskian perspective, learning occurs when teaching takes
place in the zone of proximal development (ZPD). If teachers do not teach in a student’s
ZPD, one could state that instruction is not tailored to the student’s needs and learning
problems occur. Hence, the resistance criterion should not only be measured at the
student level, but also the received teaching should be assessed carefully. Consequently,
with respect to the interpretation and measurement of the concept didactic resistance, a
dynamic view on instruction is needed. Teachers influence the results of their students,
and students influence the way of teaching (Steenbeek et al., 2012). With Curriculum
Based Measurements these interactions can be monitored, on the condition that these
will be interpreted quantitatively and qualitatively.
The RTI model is often described as a theoretical, organisational model, since specific
intervention or implementation strategies remain unclear. As a consequence, different
versions of the model have been implemented. These versions differ in number of tiers
of instruction, but also in the way progress is monitored. Models exist in which progress
is monitored on a daily base, once a week or even once a month. Moreover, the way
CBMs are constructed can also be different: based on more standardised nationwide
tests, or constructed by the teacher. Bearing the research question in mind, if didactic
resistance needs to be measured, the CBMs should be administered at least once a
week, and the CBMs should be developed by the teacher in order to be able to assess
the effects of their current teaching. This is not in line with what Vaughn and Denton
(2008) proposed. They stated that “Teacher-developed measures are useful to inform
instruction, but they may lack the necessary precision to accurately identify students’
RTI” (p.66). Although this is completely true from a researchers’ point of view, it also
reflects a standardised view of didactic resistance. A qualitative interpretation of CBMs
constructed by the teacher, supported by a reading specialist or school psychologist,
should provide more reliable information about a student’s didactic resistance than
DIDACTIC RESISTANCE AND THE RTI MODEL
77
3
standardised measures interpreted only quantitatively. Besides, as they mention
“students’ RTI”, teacher-developed CBMs are important pre-eminently, since only in
this case an estimation of a student’s didactic resistance can be made.
Moreover, if teachers construct the CBMs themselves, their teaching becomes visible
and empowers them in their reflections upon their own teaching. In order to achieve
sensitivity to a student’s zone of proximal development (which is necessary to achieve
better learning outcomes due to a better fit to the student’s capacities) teachers should be
open for advice and reflective upon their teaching practices (as also slightly emphasized
by Chak, 2001). As a consequence, in teacher training and education teachers should
acquire skills to reflect upon their own teaching, instead of only focusing at the teaching
of a particular instruction method. Also, communication between partners in a team is
really important. To reflect upon one’s teaching peer consultation could be a valuable
tool. This is a premise for better implementation of the RTI model in future. The need
for good communication between partners with a strong focus on evaluation was also
demonstrated by the pilot study, in particular during the classroom observations.
One could state that the outcomes of this study are biased to subjectivity and are not
generalisable at all due to the small sample size. However, by triangulation of data,
extensive peer debriefing and openness to subjectivity, the outcomes became trustworthy
(Altheide & Johnson, 2011; Lincoln & Guba, 1985). Subjectivism was clearly indicated
when interpreting data. Triangulation allowed for reflection upon data from different
perspectives. The methodology used can, therefore, be regarded as a reliable method
to effectuate qualitative multi-method studies (Bazeley, 2013). Due to the fact that only
a small sample of schools participated in the pilot study, generalisability has never been
the main objective. Some remarks can be made, though, that do reflect general aspects.
Although the RTI model was only implemented partially (grade 2), several factors
emerged that can be regarded as general factors. For example, it appeared that teachers’
beliefs played an important role. However, background variables of teachers were not
CHAPTER 3
78
assessed beforehand. A causal relationship between background variables, belief and
behaviour could, thus, not be effectuated, as was also the case in the study of Datnow
and Castello (2000). Since this outcome played a role in all participants from different
schools (n = 8), this still can be regarded as a more general outcome of this study. If these
general factors already occur during the partial implementation of the RTI model, it
is estimated that a school-wide implementation will even show greater problems and
drop out. Moreover, the implications for the measurement of didactic resistance on
a micro level can be expected to be even more problematic. This underlines the need
for an implementation of the RTI model in stages. Maybe the greatest effects of the
RTI model will be effectuated while implemented bottom-up instead of the top-down
implementation as is the case in the United States. If teachers understand the core
principles of the model, and at the same time share the dynamic vision on instruction,
only then the RTI model can be an instrument to measure didactic resistance. And only
then, the RTI model is favourable in identifying learning disabilities.
With respect to the definition of learning disabilities, the concept of didactic resistance
should be explored further. This study underlined the need for a thorough understanding
of the concept and in-depth research into how it should be operationalised. When
interpreted from a Vygotskian perspective, both teacher and student should be taken into
account. As a consequence, the definition of learning disabilities should not only focus
on students’ characteristics, but also on external variables, such as quality of teaching
(or teaching method). After all, learning generally takes place in reciprocal situations.
This study is paving the way for the exploration of the concept and operationalisation
of didactic resistance by making use of the RTI model, and underlines the need for
more in-depth research into this concept.
DIDACTIC RESISTANCE AND THE RTI MODEL
79
3
This chapter is based on:Tiekstra, M., & Minnaert A. (under review).
At-risk students and the role of implicit theories of intelligence in educational professionals’ actions
Chapter 4At-risk students and the role of implicit theories of intelligence in
educational professionals’ actions
CHAPTER 4
82
Abstract
Implicit theories of intelligence play a role in teacher’s actions. Adaptive instruction
in and out of the classroom is important to optimise learning processes, especially
in the case of at-risk students. This study explored to what extent implicit theories
of intelligence play a role in the actions of educational professionals around at-risk
students. 44 teachers and 57 support professionals participated in this research. Data
were analysed separately for teachers and support staff. 34% of the actions of the
teacher can be explained by implicit theories. However, in denominational schools
this is 61%. SEM showed mediation effects of multiple belief factors in the actions
of support staff. Implicit theory of intelligence predicts the belief in IQ testing, which
precedes the belief in consequential validity of tests (i.e., link to actions according to test
outcomes). These results indicate a strong influence of implicit theories of intelligence
in educational practice.
AT-RISK STUDENTS AND THE ROLE OF IMPLICIT THEORIES OF INTELLIGENCE
83
4
Introduction
The behaviour of teachers in the classroom has an impact on students’ learning
outcomes (Pakarinen et al., 2010), students’ behavioural outcomes (Skinner & Belmont,
1993) and students’ social wellbeing (Jennings & Greenberg, 2009). Moreover, the impact
sustains into later grades (Konstantopoulos & Chung, 2011) and, thus, the ultimate
development of students. Especially in the case of at-risk students, teachers’ behaviour
might have an enormous impact on learning outcomes. Students classified as at-risk in
their learning progress often show learning delays as a consequence of environmental
causes (such as immigrant or low SES families) or caused by learning difficulties (such
as reading or arithmetic problems). Consequently, at-risk students need more coaching
than the rest of their peers, resulting in more one-to-one contacts in which the teacher
plays a crucial role. In practice, this often results in challenging situations where the
teacher needs to guard the whole classroom and to coach at-risk students at the same
time. Furthermore, in order to achieve as best learning results as possible, teachers
should adapt instruction to the level of their students. According to Vygotsky (1978)
the teacher should teach in the Zone of Proximal Development of each student. This
Zone of Proximal Development (ZPD) is the difference between the actual independent
achievement of the student and the performance level of the student when tailored
help is provided (Lidz, 1995). Teaching in the ZPD helps fulfilling the student’s learning
potential, which contributes to the development of the student.
The above-mentioned aspects can be seen as the core principles of teachers’ actions
in the classroom. However, teacher’s actions in the classroom are presumed to be
influenced by implicit theories of intelligence. Dweck and colleagues demonstrated in
their extensive research that implicit theories affect people’s judgments and reactions
(Dweck, Chiu, & Hong, 1995). The influences of teachers’ beliefs has been studied for
longtime (Wiedl, 1980) and has been demonstrated to influence behaviour (Azjen, 2005).
More specifically, Bandura (1993) described the influence of teachers’ beliefs to quality
of teaching and their behaviour in the classroom, which in turn has its impact on the
CHAPTER 4
84
school career of students (Gibbs & Powell, 2012). Beliefs that have been demonstrated
to play a role during teaching processes refer to a broad terminology (Pajares, 1992), for
example teachers’ self-efficacy beliefs (e.g., Tschannen-Moran & Woolfolk Hoy, 2001),
teachers’ perceptions (e.g., Pelletier, Seguin-Levesque, & Legault, 2002), teachers’
attributions (Gibbs & Gardiner, 2008), or teachers’ feeling of responsibility (e.g.,
Lauermann & Karabenick, 2011). Next to these examples, the influence of people’s
implicit theories on their behaviour has been investigated. Implicit theories can be seen
as beliefs that have been constituted subconsciously within an individual, thus, people
are often unaware of the theories that they support.
People’s implicit theories can be general, but are mostly domain-specific. In this article
we focus on the implicit theories on intelligence, referring to the extent to which an
individual considers intelligence as malleable or not (Dweck, et al., 1995). Studies in
the domain of implicit theories of intelligence mainly focused on the impact of an
individual’s implicit theories on the achievement goals of this individual (e.g., Dinger
& Dickhäuser, 2013), as was originally initiated by studies of Dweck (see also Dweck
& Leggett, 1988). In turn, achievement goals determine an individual’s behaviour.
According to Minnaert (2013) achievement goals are considered to contribute to school
success of students substantially, but the way in which achievement goals influence
one’s learning outcomes remains complex. He underlined the need for more research
in this area. According to Leroy, Bressoux, Sarrazin, and Trouilloud (2007) teachers’
implicit theories influence the atmosphere in the classroom. Ames (1992) pointed at
the influence of classroom structure to students’ goals. This gives an indication for the
role that implicit theories play during learning processes of students. However, studies
that explicitly address the link between teachers’ implicit theories and its influences on
students’ learning outcomes are executed to a much lesser extent (Wolters & Daugherty,
2007). A first step in this issue is to investigate the assumed relationship between implicit
theories and behaviour of teachers. Following this assumption, classroom practices of
teachers are supposed to affect students’ learning outcomes.
AT-RISK STUDENTS AND THE ROLE OF IMPLICIT THEORIES OF INTELLIGENCE
85
4
People adhere to either an entity view or an incremental view on intelligence (Dweck
et al., 1995). The entity theorists hold the opinion that intelligence is a stable trait and
something that can hardly be changed. Individuals that support the incremental view,
consider intelligence rather as malleable and subjective to change. These different views
imply differences in achievement goals, since achievement situations are experienced
contrastingly (Dweck & Leggett, 1988). An entity theorist assumes an achievement
situation as a situation to demonstrate skills oriented towards evaluation, resulting
in performance goals. Whereas incremental theorists are oriented towards learning
goals, since achievement situations are regarded as learning opportunities (Dinger
& Dickhäuser, 2013). Consequently, these views have their influence on classroom
practices. For instance, teachers with an entity view of intelligence are more likely to be
focused on performance and direct instruction practices. Whereas incremental theorists
envision mastery-oriented learning and focus on the learning process instead of the
learning results solely.
Teaching in the ZPD is, thus, influenced by these implicit theories of intelligence. This,
in turn, has its impact on educational opportunities of students. For instance, in the case
of low performing students, an entity teacher would only adapt instruction in such way
that these students get supplementary practice or extra assignments, or opt in favor of
exclusion, since low scores on achievement tests indicate a low ability to learn. Entity
teachers, do not expect these students to learn, resulting in more stigmatisation and
self-fulfilling prophecies. On the contrary, an incremental teacher would provide instruction
that is adapted to the level of the student, since low scores on achievement tests indicate
that these students needs more (or other) opportunities to learn. Exploring a student’s
ZPD is, likewise, more common among incremental theorists.
Due to deviating results from the rest of their classmates, at-risk students are frequently
eligible to assessment procedures. These assessment procedures often include IQ tests.
Accordingly, outcomes of intelligence testing are interpreted differently by entity or
incremental theorists. Moreover, entity theorists will attach great importance to the
CHAPTER 4
86
outcomes of an IQ test, whereas incremental theorists will approach these outcomes as
a starting point for adaptive support.
The influences of teacher’s implicit theories on particular groups of at-risk students have
been investigated previously. Literature can be found about the group of gifted students
(Baudson & Preckel, 2013; García-Cepero & McCoach, 2009). However, information
about SEN students seems to be scarce. Jordan and Stanovich (2001) underlined in their
research that teachers’ beliefs affect the instructional processes, and had even greater
influences in the group of at-risk students. Teachers who had an incremental view, and
therefore felt responsible for their students’ learning processes, were more frequently
involved with at-risk students than teachers with an entity view. The need for more in-
depth examination of the influence of teachers’ beliefs on their educational actions is
underlined by several researchers (e.g., García-Cepero & McCoach, 2009).
Aim of the present studySince the content and manifestation of instruction is extremely important for the
group of at-risk students, a closer look is needed in the process among educational
professionals involved with teaching these children. The current study aims at the
exploration of a relationship between implicit theories of intelligence and actions of
educational staff with regard to the care process of at-risk students. At-risk students
often undergo assessment procedures eligible for individualised education plans (IEP),
resulting in interventions and adapted instruction. Therefore, next to teachers, support
professionals were taken into account. These professionals administer the assessments
and assist teachers in their instructional processes. But in what way play implicit theories
part in the educational professionals’ actions in practice? As mentioned above, entity
and incremental theorists approach at-risk children differently. However, the extent to
which these implicit theories play their role in daily practice remains unclear. Therefore,
this study addresses this issue from a practical point of view. The role of implicit theories
will be investigated with respect to actions in practice.
AT-RISK STUDENTS AND THE ROLE OF IMPLICIT THEORIES OF INTELLIGENCE
87
4
There is plenty of evidence that background variables such as years of teaching
experience, and level of grade play a role in teachers’ beliefs (Bol, Stephenson, O’connell,
& Nunnery, 1998; Leroy et al., 2007; Wolters & Daugherty, 2007). Therefore, these
variables have been taken into account. If actions can indeed be explained by implicit
theories, then, do amount of experience, and the grade in which the educational
professionals work, interact in this relationship? Moreover, the influence of school ethos
on teachers’ beliefs has been highlighted by Gibbs and Powell (2012). They underlined
the importance of collective beliefs: Teachers are often influenced by collective beliefs
that are prominent in a school. As illustrated by Opdenakker and Van Damme (2006),
the type of school and denomination contribute to collective beliefs inside schools,
which affects students’ outcomes. Therefore, it was also investigated whether the type
of school (defined as denominational or public school1) influenced the relationship
between implicit theory and behaviour.
Method
SampleThe study was carried out in the western part of The Netherlands. This specific area
of the Netherlands is confronted with a large population of ethnic minorities. In
comparison to other parts of The Netherlands, teachers in this area are confronted
with more at-risk students in their classrooms having more serious problematic home
situations. The issue of adapting instruction to the level of students is very common
in schools in this area. Due to the high incidence of immigrant children, schools often
receive financial aids from the government, which can be used for extra assistance inside
or outside the classroom.
1 In this, denominational schools refer to schools that hold a Christian point of view in their education, whereas public schools are not explicitly grounded in an ideology. Students in public schools are not restricted to a certain religion. The division does not account for differences in financial aid.
CHAPTER 4
88
In The Netherlands elementary school begins with Kindergarten (at age 4) and continues
until grade 6 (at age 12), after which children attend secondary education. The ‘care
process’ at regular education is organised in different levels where school psychologists,
special services coordinators, remedial teachers, and teachers are involved (Imants, Van
der Aalsvoort, De Brabander, & Ruijssenaars, 2001). Generally, when teachers remark
problems with a student, the school’s special services coordinator assists the teacher in
these problems. In the case of persistency of problems, the special services coordinator
asks help from a school psychologist working at an educational advisory agency.
For this study, school psychologists working for an educational advisory agency were
asked to participate. Each school psychologist would then distribute the questionnaire
to one special services coordinator with whom they worked and three teachers with
whom the special services coordinator worked. One teacher represented the first two
years of elementary school (Kindergarten), another teacher grade 1 to 3, and the third
teacher represented grade 4 to 6 of elementary school to provide a representative
sample of teachers. It was aimed for to obtain a response of 50 school psychologist, 50
special services coordinators, and 150 teachers. A total of 44 teachers and 57 support
professionals (21 special services coordinators and 36 school psychologists) participated
in this research, which reflects a response rate of 40,4% in total.
Measures Implicit theories of intelligence (ITI) were assessed by three items that indicated the view
on intelligence, either as an entity or incremental theory. The original items (Dweck et
al., 1995) were converted to teacher items and translated to Dutch. Next to these items,
the behaviour of teachers and support staff in case of an at-risk student was assessed
by a content-driven questionnaire. Since the study was explorative by nature, and the
items needed to address behaviour that was adapted to the level of each professional,
a different version for teachers and support staff was used. Depending on the version,
the questionnaire consisted of 17 or 18 statements that had to be answered on a 1 to 5
Likert scale.
AT-RISK STUDENTS AND THE ROLE OF IMPLICIT THEORIES OF INTELLIGENCE
89
4
Teacher questionnaire. The ITI- items proved to be reliable in our sample, as
indicated by a Cronbach’s alpha of .77 in this version. A principal component analyses
(PCA), graphical representations and other ad-hoc descriptive data analyses revealed
that in the teachers’ questionnaire one factor emerged (α.73 ; 9 items) explaining 33%
of the total variance. Examples of items are I observe problems in learning progress of a student
when the student obtains low scores on achievement tests that I administered, and I evaluate instruction
I provide to the at-risk student with the special services coordinator. This factor was called “actions
in case of an at-risk student”. A higher score on this factor reflects more consistency in
actions undertaken by the teacher in case of an at-risk student.
Support professional questionnaire. The ITI- items had an Cronbach’s alpha
of .75 for support staff. The PCA’s in the support professionals’ questionnaire showed
two factors that could be distinguished. The first factor was composed of three items (α.
58) and reflected the belief that IQ testing is a correct estimator of learning capacities,
called “belief in IQ testing”. One of the items was e.g., The administration of an IQ test is
enough for the estimation of a child’s learning capacity. The higher the factor score, the more
faith to IQ testing (i.e., the score reflects learning capacity of students perfectly).
The second factor was related to three items (α .55) describing the link from test score to
actions in practice, shortly the “belief in consequential validity of tests”. For example,
Test outcomes have a binding effect upon the advice I give to teachers. A high score was indicative of
more value attributed to results of testing, implying a more performance oriented view
on capacities. The authors note that several items were not taken into account, since
these did not load on any of the factors, though the other items had very high loadings
on one of the two factors.
Both factors could, thus, clearly be separated according to the data. Moreover, both
factors explained 55% of the total variance.
CHAPTER 4
90
Analyses Hierarchical regression analyses were carried out to investigate the possible influences
of background variables on the relationship between the ITI- items and the other factor
for teachers’ questionnaire. Careful interpretation of the two factors that emerged from
the PCA’s in the support professionals’ questionnaire showed that one factor reflected
upon belief in IQ testing and the other upon belief in consequential validity of tests.
Therefore, a strong grounding for testing mediational effects appeared. By making use of
the Lisrel program, Structural Equation Modeling (SEM) was used to test the mediation
effects between different components of the support professionals’ questionnaire.
Results
TeachersThe correlation between the implicit theory of intelligence factor and the other factor
was .58 (p ≤ .01; N = 38). Ergo, 34% of the actions of a specific teacher in this sample
can be explained by the implicit theory of intelligence this teacher has. Hierarchical
regression analyses investigated the moderating effects of background variables in the
relationship between implicit theories of intelligence on the one side and “acting in case
of at-risk student” on the other side. Experience of teacher, and the group in which
the teacher works, did not significantly moderate this relationship. More interestingly,
it appeared that the variable “type of school (public vs. denominational)” did play a
moderating role in this relationship. As can be seen in Table 1, the analysis resulted
in a significant additive 9% in the variance explained by adding the interaction effect
between type of school and implicit theory in the relationship between implicit theory
and actions of the teacher (∆R2 = 0.92, ∆F = 5.45, p = .03).
AT-RISK STUDENTS AND THE ROLE OF IMPLICIT THEORIES OF INTELLIGENCE
91
4
Table 1 | Hierarchical regression analyses predicting ‘acting of teacher’ from implicit theories of intelligence, and type of school, and interaction between both predictors
Model R ∆R2 ∆F Sign. ∆F
Implicit theory .58 .34 18.13 .00
Implicit theory; Type of school .58 .00 .02 .88
Implicit theory; Type of school; Implicit theory * type of school
.65 .09 5.45 .03
Note. N = 38, distributed over 10 public and 10 denominational schools
Actually, as showed in Figure 1, this interaction effect between ITI and actions in case
of an at-risk student appears to be much stronger in denominational schools than in
public schools. In denominational school implicit theories of intelligence explain 61%
(∆R2 = 0.61, ∆F = 20,6, p = .001, N =15) of actions of the teacher, compared to only
20% (∆R2 = 0.20, ∆F = 5.2, p = .03, N = 23) of actions in public schools. This effect was
not subjective to outliers, as was tested by the Leverage and Cook criteria.
CHAPTER 4
92
Teac
hter
act
ions
5,0 7,5 10,0
Implicit theory of intelligence
12,5 15,015,0
20,0
25,0
30,0
35,0
40,0
publicdenominationalpublicdenominational
Figure 1 | The difference in influence of implicit theory of intelligence on actions of the teacher between teachers at public schools and denominational schools.
Support professionalsHierarchical regression analysis revealed no moderating effects of background variables
in this group on the relationship between implicit theories and both factors. Structural
equation modeling showed that there was a full mediational model between these
variables. The model tested the effect of implicit theories of intelligence, via belief in IQ
tests, on the belief in consequential validity of tests, as dependent variable. This model
had a very good fit, since the margin of error was small (90% CI RMSEA [0.00, 0.00],
p = .998) and the sources for unexplained variance in the model were trivial (SRMR=
.004). There were no direct effects of implicit theories on the belief in consequential
validity. As can be seen in Figure 2, implicit theory of intelligence predicts the belief in
IQ testing, which, in turn predicts the consequential validity of tests.
AT-RISK STUDENTS AND THE ROLE OF IMPLICIT THEORIES OF INTELLIGENCE
93
4
Figure 2 | Mediational model of belief factors in support staff; * significance at a 5% level. N = 56.
Discussion
Since the influence of implicit theories of intelligence (either entity or incremental)
in practice needed to be investigated, this study aimed at an in-depth exploration of
actions of educational professionals in the case of at-risk students. These students are
often confronted with the consequences of stigmatisation and self-fulfilling prophecy,
need more coaching from teachers and support professionals, and are often eligible
for assessment procedures. This study explored the underlying principles for the
educational professionals’ actions to give an insight in quality of care processes around
at-risk students. Support staff (i.e., school psychologists and remedial teachers) and
teachers were asked to fill in a questionnaire. The ad hoc constructed questionnaire was
composed of items that reflected upon practice and items that reflected one’s implicit
theory of intelligence (from Dweck et al., 1995). Data analysis showed clear distinctions
between the level of the teacher and the level of support staff. Therefore, results will be
listed separately for both groups of participants.
In respect of teachers, results show that implicit theories play a prominent role in their
actions in the classroom in the case of an at-risk student. In this sample, 34% of the
variance on the factor “actions of a teacher” can be explained by either an entity view or
Implicit theory ofintelligence Belief in IQ testing Consequential
validity of testing
.25* .33*
.08
CHAPTER 4
94
an incremental view on intelligence. Moreover, this study showed that this relationship
is moderated by type of school. Analyses showed that the prediction of teachers’ actions
by their implicit theories is more robust at denominational (Christian) schools than at
public schools. Differences between both types of schools have been found previously in
educational studies. However, as Opdenakker and Van Damme (2006) underline, these
differences are often related to other variables.
The positive correlation coefficients for both type of schools indicate that the more
a teacher supports the entity view of intelligence, the more consistency is showed in
acting in the case of an at-risk student. Thus, teachers know better what they need to
do in case of an at-risk student if they support the entity view. However, their actions
are in line with the entity view, reflecting a more performance oriented approach as
explained in the introduction. Consequently, stigmatisation and self-fulfilling prophecies
could occur in these situations. On the other hand, as was shown in Figure 1, there
is a lack of teachers that do support the incremental view on intelligence showing a
high consistency in their actions. The more incremental view on intelligence, the less
consistency in actions of the teacher. Hence, some teachers do support a more dynamic
(i.e., incremental) view on intelligence, but they do not know how to act accordingly.
This indicates a lack of declarative knowledge or a lack of procedural knowledge in the
actions of teachers. These effects are more robust at denominational schools.
With regard to support professionals, i.e. special services coordinators and school
psychologists, implicit theories do also play a role, though differently. Next to the
factor describing implicit theory of intelligence, two other factors emerged from the
questionnaires. One factor reflecting the belief in IQ testing and another factor reflecting
the belief in consequential validity of testing. Background variables appeared to be of
less importance in this group, since these did not moderate on relationships between the
factors. The SEM modeling showed that there was a full mediational model between
the factors, implying that implicit theories predict the belief one has about IQ tests,
and, in turn, explains the way people act according to these tests. The consequential
AT-RISK STUDENTS AND THE ROLE OF IMPLICIT THEORIES OF INTELLIGENCE
95
4
validity of tests is, thus, caused by implicit theories people have. This means that the
way people think about intelligence affects their opinion about the quality of these tests,
and thus, their interpretations. If one supports the more static view on intelligence (an
entity theorist), this individual will hold more account to IQ testing results, which will
result in more static decision making processes. One might regard this model rather as
logical, but when applied to the educational practice, it is rather disastrous. Due to a
stronger focus on IQ testing results, learning problems will be regarded as a problem of
the students, instead of a reciprocal effect between teacher and students. Less attention
will be drawn upon the role of the teacher in adapting to the level of the student.
Consequently, the teacher will not teach in one Zone of Proximal Development.
As demonstrated by this study, implicit theories of intelligence play a substantive
role in the educational care process around at-risk students. In this group of students
the identification of learning disabilities, or the admission to remedial activities, is a
frequently occurring issue. In each of these decision making events, implicit theories play
a role. In turn, the impact on educational opportunities of at-risk students is immense.
These findings suggest that educational professionals should develop awareness of
the influence of these implicit theories on all their actions. Educational professionals
should strive towards a model of constant reflecting upon one’s own actions and of
those professionals around them. Teachers should be supported in their instructional
practices. School directors should create opportunities for their teams in order to consult
each other’s work. An incremental view on intelligence should be supported in order to
create learning opportunities and exploration of a students’ ZPD rather than a focus on
performances (as is often the case with entity theorists).
There are, however, some limitations to the results of this study. The relatively low
response rate (40,4%) may have caused biased results. Moreover, the sample size
was rather small (44 teachers and 57 support professionals). Although the statistical
grounding of some results might be regarded as thin, the authors would like to emphasise
this small sample size which probably caused a problem of power. The alpha’s of the
CHAPTER 4
96
factors would have been higher if the test would have been lengthened. According to
Carmines & Zeller (1979) and Cortina (1993) interpretation of alphas should be done
with caution, as it is subjective to the number of items used. The reliability of the
scores on the separate factors is, therefore, above threshold, even in the case of a low
average inter-item correlation, since only three items have been used in some factors.
This implies that evidence might have been even stronger when the study would have
been carried out with more items, and in a greater sample distributed nationally instead
of a specific area in the western part of the Netherlands. Moreover, due to the high
incidence of at-risk students in this specific area, it is estimated that people hold a more
flexible approach to these students than in the rest of The Netherlands. A nation-wide
sample would probably have proved an even more persisting role of implicit theories
of intelligence in educational practice. Furthermore, results are based on self-reported
questionnaires which are always sensitive to socially acceptable answers.
To conclude, these results show that we should be very attentive to implicit theories on
intelligence of educational professionals. They play a prominent role in processes in
the classroom, and outside the classroom, in case of at-risk students. These results have
clear implications for the implementation of educational interventions. If the aim of
a specific educational intervention is to change behaviour of educational professionals,
this study indicated that more processes come into play when interpreting the effects
of the intervention. Behaviour might not be changed due to persistent implicit theories
people have. Accordingly, this study paved the way for interventions that intervene
on beliefs and behaviour at the same time. Since both aspects cannot be seen apart
from each other, one should strive toward a change in both aspects in order to achieve
sustainable effects in behaviour.
With regard to the consequential validity of testing procedures, the study also
had implications. The study demonstrated that implicit theories play a role in the
interpretation of, and actions according to test outcomes. This means that not so
much the outcome itself, but rather the interpretation play a prominent role in psycho-
AT-RISK STUDENTS AND THE ROLE OF IMPLICIT THEORIES OF INTELLIGENCE
97
4
educational practice. More focus upon the interpretation of test outcomes and the
translation to practical guidelines, in other words the consequential validity of tests, is
requested for. After all, the future of individual at-risk students often depends on few
assessment outcomes.
“The unexamined life is not worth living.”(Socrates)
Chapter 5Voices from practice:
when is the gap between diagnosis and intervention apparent?
Acknowledgement:I thank Lotte Bergwerff for her contribution to the studies in this chapter.
CHAPTER 5
100
Abstract
In this chapter an overview is provided of the current state of psycho-educational
practice in the Netherlands, in particular the role of test outcomes considered. In order
to detect clues for bridging the gap between diagnosis and intervention, one should
investigate the ecology in which this gap is apparent. Two in vivo studies have been
carried out. In a first study, a questionnaire has been administered to a total of 36 school
psychologists, 21 special care coordinators, and 44 teachers. A qualitative analysis of the
answers, by means of a classification scheme (κ .82), revealed questions about the specific
role of intelligence tests and its consequences to educational practices. Therefore, a
case study has been carried out in a second study. The case study provided in-depth
information about the targeted care process around a grade 1 student. Results indicated
a gap between diagnosis and intervention that followed after the administration of
the intelligence test. Suggestions are proposed for improvement, and the need for
interventions at the level of educational professionals is highlighted.
VOICES FROM PRACTICE: WHEN IS THE GAP BETWEEN DIAGNOSIS AND INTERVENTION APPARENT?
101
5
Introduction
The results of a recent literature review (see chapter 2 of this dissertation) showed
that more attention should be drawn to consequences of test administrations and test
results in the educational field. In other words, the consequential validity (Messick,
1995) of assessment procedures is insufficiently warranted at the moment (Tiekstra,
Minnaert, & Hessels, 2016). As a consequence, the well-known gap between diagnosis
and intervention remains present (Haywood, 2012; Shapiro & Kratochwill, 2000). If
we want the attempts in bridging this gap to be successful, one should not only focus on
the outcomes of the assessment procedures, but also focus upon educational practice.
As demonstrated in chapter four of this dissertation it appears that the way people
think about intelligence influences the way people act in practice. Implicit theories of
intelligence, thus, play a prominent role in educational practice (Dweck, Chiu, & Hong,
1995; Tiekstra, & Minnaert, submitted). Accordingly, test scores influence teachers’
ideas and beliefs about students, and are often input for decisions about a student’s
academic career (i.e., special education eligibility, targeted care). Therefore, the need
for valid outcomes of (intelligence) tests is urgent.
One of the core elements in this dissertation is the exploration of ways to enlarge
consequential validity of test procedures. A next step in this exploration is to investigate
the ecology in which the studies in this dissertation are situated. This chapter reports on
the results of an in vivo study where the educational field in case of at-risk students was
subject to a profound examination. Only in this way ecologically valid conclusions can
be drawn about the state of consequential valid test outcomes that might bridge the gap
between diagnosis and intervention.
Role of test scoresWhen diagnosing children with severe learning disabilities, test scores play a prominent
role. Generally, school achievement tests are used to detect learning delays, and
standardised intelligence tests often come into play when students show persisting
CHAPTER 5
102
learning delays. As highlighted in chapter three, in the newest definition proposed
by the DSM-5, an IQ score is still regarded as an important parameter to diagnose
learning disability. Therefore, an IQ measure is a prerequisite to get subsidisation from
the government. As a consequence, if a child displays difficulties in learning reference is
often made to an IQ test. To summarise, IQ tests play a prominent role in daily psycho-
educational practice.
The use of IQ tests in individual assessment procedures is, however, questionable.
Growing evidence shows that IQ tests do not provide reliable and valid measures in at-
risk children (for more information see e.g. Tiekstra et al., 2014). Besides, intelligence tests
have been criticised for their lack of information for practice (Gresham & Witt, 1997;
Lebeer et al., 2011), often referred to as the gap between diagnosis and intervention
(Shapiro & Kratochwill, 2000). It is worth mentioning though, that, originally,
standardised tests had been developed for classification issues instead of educational
intervention planning. Theoretically, this means that IQ tests are not purposefully
developed to provide information for educational practice. Subsequently, the question
arises whether and why we should administer IQ tests as frequently as is done in current
educational practice? Lebeer and colleagues (2011) demonstrated in their large-scale
study among European countries that there is an urgent need for practical information
according to test outcomes. Moreover, if we want to prevent teachers from planning
their instruction intuitively, a focus upon accurate translation of test outcomes to daily
practice is required.
Alternative formats of assessment should be used to bridge the gap between diagnostic
assessment and classroom practice. Bosma and Resing (2008) proposed to use dynamic
assessment. They investigated the opinions of teachers about diagnostic reports in
which a measure based on either a traditional or a dynamic test was indicated. Teachers
appreciated the dynamic reports in which clues for classroom practice were provided.
Moreover, classroom observations revealed that some of the teachers even adapted their
instruction. Research shows, however, that dynamic assessment is not yet implemented
VOICES FROM PRACTICE: WHEN IS THE GAP BETWEEN DIAGNOSIS AND INTERVENTION APPARENT?
103
5
in educational practice (e.g., Elliott, 2003). Currently, the gap remains present in Dutch
educational practice in two ways: the used assessment does not provide guidelines for
the process after the assessment, furthermore dynamic assessment (which provides such
guidelines) is not yet implemented at schools.
Professionals involvedIn terms of reference, generally, individual test administrations belong to school
psychologists. Thurlow and Ysseldyke (1982) underlined the difference between school
psychologists and teachers with regard to their opinions about information that is
needed for educational intervention planning. They indicated that school psychologists
rely on the use of standardised test outcomes solely, while teachers also supported the
use of behavioural observations in instructional planning. Rouse and Agbenu (1998)
underlined the problems teachers have with all kinds of assessment and suggest that
“many of the problems results from confusion about the nature and purpose of
assessment, the teachers’ role in this process and how to assess pupils with learning
difficulties.” In short, more insight is needed into the educational practice with regard
to processes around students showing problems in learning and information processing.
In the Netherlands another professional comes into the picture when a child has
learning difficulties: next to the remedial teacher, a special care coordinator operates
within the care system of the school. In Dutch regular education, the teacher generally
detects learning problems of students (as caused by learning delay, learning disability or
behavioural problem) in the classroom. The special care coordinator’s task is to support
the teacher in resolving these learning delays. If the learning problems exist, the school
psychologist will provide specialised assistances. In effect, usually, the care process is
organised by different professionals involved. Theoretically, this implies clear roles, but
in the educational practice these roles are less clear. As demonstrated by Imants, Van
der Aalsvoort, De Brabander, and Ruijssemaars (2001) special care coordinators have
different functions in different schools, and rarely provide coaching to teachers.
CHAPTER 5
104
Aim of the current research If the consequential validity of assessment procedures needs to be improved, then
we should know how the care processes in schools are effectuated and what role test
outcomes play in these care processes. The aim of this research is to thoroughly explore
the processes in regular education in case of an at-risk student. What role do the
professionals involved play, and what type and kind of information are decisions based
upon? Therefore, the goal of the study was to provide situational information about this
issue by interpreting information from multiple voices from practice. This demanded a two-
step approach; a questionnaire was distributed first. Second, a case-study was needed to
provide answers to questions resulting these questionnaires.
General Method
The in vivo study aimed to describe experiences and needs of educational professionals
in the Netherlands. Several studies were carried out. Both quantitative and qualitative
data was gathered. For the purpose of clarity, this chapter focuses on the results of
the qualitative part solely. The results of the quantitative part are included within the
previous chapter (chapter 4) of this dissertation, only some descriptive quantitative
(frequency) analysis will be displayed in this chapter when needed to illustrate the results
of this study.
A multiple qualitative approach was used in order to compare and explore the needs
and experiences in greater depth. A questionnaire comprising statements, vignettes,
and open-ended questions was administered to explore the current state in providing
care to at-risk students. In addition, a case study was carried out which allowed for in-
depth information. A case study describes a case within a context, resulting in detailed
information that is closely related to practice (Flyvbjerg, 2006; Swanborn, 2010).
VOICES FROM PRACTICE: WHEN IS THE GAP BETWEEN DIAGNOSIS AND INTERVENTION APPARENT?
105
5
ParticipantsThe study was set up in collaboration with an educational advisory agency in the
Western part of the Netherlands. In comparison to other parts of The Netherlands,
teachers in this area are confronted with more at-risk students in their classrooms
having more serious problematic home situations. Due to the high incidence of
immigrant children, schools often receive financial aid from the government, which
can be used for extra student oriented assistance inside or outside the classroom.
ProcedureTwo separate studies have been carried out. A questionnaire has been distributed in
the second semester of the 2012 school year. Next, a case study has been conducted
in the same semester during a 4 month period. The school selected for the case
study did not participate in the questionnaire study, nor had any knowledge of it.
Study 1: In Vivo Questionnaire
MethodA questionnaire was developed which addressed several aspects of the care process
around children showing any kind of learning problems in regular education. These
aspects were the identification of problems (related to intelligence and intelligence
tests), educational plans, communication, and the effectuation of care processes.
Examples of items can be found in the appendix.
Sample. It was aimed for to obtain a response of 50 school psychologists, 50
special care coordinators and 150 teachers. First, school psychologists attached
to the educational advisory agency completed the questionnaire. Second, school
psychologists distributed the questionnaire to one special care coordinator with
whom they worked regularly and three teachers with whom the special care
coordinator worked. A total of 44 teachers, 21 special care coordinators, and 36
CHAPTER 5
106
school psychologists participated in this research, which reflects a response rate of
40,4% in total.
Instruments. The questionnaire comprised of three divisions: statements, a vignette, and
open-ended questions. Logically, these questions needed to be closely related to the daily
practice of participants, so three different versions were developed.
The statements had to be answered on a Likert-scale, varying from 1 (completely disagree)
to 5 (completely agree). Participants needed to respond to 17 or 18 statements, depending
on the version. An example of an item in the teachers’ version is The information I get from the
special care coordinator is sufficient to adapt my instruction to the specific needs of student. And an example
which was used in all versions is Administering an IQ test is essential for setting up educational plans.
The vignette was composed of a situational description in which a student with severe
learning problems needed to be identified. The participants were asked about their role
in the care process in two multiple choice questions (special care coordinator, and school
psychologist) or one multiple choice and one open question (teacher), formulated as What
would you do if you would observe problems of student X, as in the given case?
The section with open-ended questions also focused on the care process of students showing
learning problems. In the version of school psychologist 3 questions were asked. Special
care coordinators had to answer 5 questions. The teachers’ version comprised another
multiple choice question (where the last option was: something else, namely…) and 4 open-
ended questions.
Procedure. The questionnaire was distributed to school psychologists in the second
semester of the 2012 school year. Then, school psychologists were asked to distribute the
questionnaire to special care coordinators and teachers with whom they worked. These
completed questionnaires were sent to the researcher.
Analysis. A descriptive analysis took place of the statements (frequency analysis)
VOICES FROM PRACTICE: WHEN IS THE GAP BETWEEN DIAGNOSIS AND INTERVENTION APPARENT?
107
5
and the multiple choice questions. Answers to the open-ended questions have been
analysed by means of a classification scheme. This classification scheme was developed
in two phases. First, 10% of the questionnaires was randomly selected to formulate
classifications. The selected questionnaires did not provide sufficient variation in
answers, therefore classifications based on researchers’ experiences were added to the
scheme. Several examples of items and the categories in the classification scheme can
be found in the appendix. It must be mentioned that multiple classifications could be
scored per item. The classification scheme was tested for its reliability, resulting in an
overall interrater agreement of κ.82. The Cohen’s kappa for the version of teachers was
κ.74, for special care coordinators κ.90, and κ.87 for school psychologists. Because of
these high Kappa values, the classification scheme can be considered reliable, and not
subject to chance.
Results and interpretationIdentification of problems. The vignette study showed that teachers generally
observe learning problems based on the results of the previous school year (45%), or
based on observations in the classroom (46%).
The majority of teachers thinks that administering IQ tests is essential for advice with
regard to the educational practice, whereas a minority of school psychologists and special
care coordinators supports this statement. In any case, teachers are informed about the
results of the individual test administration by the school psychologist. Less experienced
special care coordinators (0-3 years of experience) indicate that their advice to teachers
is often to very often based on these test results, whereas more experienced special care
coordinators ( ≥ 3 years) indicate sometimes this is the case. Special care coordinators do
not administer IQ tests themselves. 44% of school psychologists think that the current
intelligence tests are appropriate enough to base their advice on. Whether this advice
is rather needs-based or classification focused remains unclear. Moreover, special care
coordinators and school psychologists value the outcomes of intelligence tests to a wide
range (min. 1 max. 4), implying large individual differences.
CHAPTER 5
108
The value of information that intelligence tests provide for educational practice is
ambiguously evaluated by school psychologists. On the one side, school psychologists
often provide advice for the classroom practice based on test outcomes. On the other
side, they indicate that test outcomes do not provide such information exactly to the
specific context of the assessed student. Here, too, the type of advice remains unclear.
Half of the teachers indicated to detect learning problems when a student shows low
scores on curriculum related achievement tests. There is no consensus in the answers
about the use of these tests in identification of learning problems. 40% of teachers
consult their more experienced colleagues in case of at-risk students.
And finally it should be noted that clear differences were observed with regard to the
knowledge of dynamic assessment: only 10% of teachers knew what dynamic assessment
is, compared to 30% of special care coordinators, and 80% school psychologists. It
must be mentioned, though, that these percentages are elevated compared to average
Dutch schools. The questionnaires were administered to a group of school psychologists
who worked at the same educational advisory agency. This agency provided their staff
with training in dynamic assessment which is rather exceptional in the Netherlands.
Therefore, these percentages are not representative to the Dutch population. However,
the amount of teachers and special care coordinators is, still, very small. Moreover, only
80% of school psychologists indicated to have knowledge of dynamic assessment.
Individual Educational Plan (IEP). 70% of the special care coordinators thinks that
the psychological reports provide enough information for the instruction in the classroom.
It appeared that teachers set up IEP, sometimes in consultation with the special care
coordinator. 60% of special care coordinators and 86% of school psychologists indicate
setting up IEP for teachers only rarely. The content of the IEP is often discussed between
special care coordinators and teachers (80%). Strikingly, no special care coordinator reported
that the educational plans are discussed with school psychologist.
VOICES FROM PRACTICE: WHEN IS THE GAP BETWEEN DIAGNOSIS AND INTERVENTION APPARENT?
109
5
With regard to the psychological reports, 33 of 36 school psychologists need test results to
write these reports. Besides, 25 school psychologists would like to receive information from
the teacher, and 19 indicate information about home situation of the student.
Communication. In most cases, teachers report immediately to special care coordinators
when observing learning problems (41%). However, teachers did not answer clearly to the
question “What happens after you consulted the special care coordinator?” The majority of teachers
indicates that the information they get from special care coordinators is sufficient for adapting
their instruction. Special care coordinators and teachers evaluate the adapted instruction
with regard to at-risk student regularly. However, school psychologists are not informed of
the adaptations in the teaching process.
According to the answer to the open-ended question for teacher “What is the school psychologist’s
role?” the role of school psychologist is not clear to teachers. According to some school
psychologists, on the other hand, information they provide about the students’ results and
advice is not acted upon by the teacher. Evaluation of the care process only takes place
between teacher and special care coordinator. School psychologists are rarely involved with
evaluation, as one of them remarked “I suspect that my advice is put into a drawer and never looked at
again”. This demonstrates that there might be discontinuity in targeted care.
Effectuation of care processes. As mentioned before, teachers report to special
care coordinator in case of an at-risk student. The steps undertaken by special care
coordinators differ very much. Answers like I administer several tests (24%), or consultation
to other experts (19%) were mentioned. They seldom assist in adapting instruction: half
of them indicate never inventing exercises for the classroom instruction, the other half
rarely.
According to teachers, the school psychologist comes into play when the school is unable
to provide appropriate intervention. 17% of the school psychologists immediately
administer an IQ test when a child is referred to them, 83% works differently. Several
CHAPTER 5
110
aspects, like classroom observation, parental or teacher consultation, are mentioned.
School psychologist, thus, work very differently with respect to at-risk students.
Future needs. Participants were asked about information they would need, and
currently did not receive. Teachers strongly suggest that school psychologists provide
information about care or interventions. Special care coordinators indicate the need for
a specialist within the school. School psychologists would like to have opportunities to
evaluate and take care of continuity of the care process. Currently, they cannot perform
a follow-up of the students after their reports.
ConclusionDue to the rather ambiguous answers to questions around the value of intelligence
tests for advice in educational practice, questions remained unanswered. Although
participants were positive overall about care processes, discontinuity in the care
processes was shown. Roles are unclear, and evaluation hardly takes place. Moreover,
questions arose about the type and format of advice (remedial) teachers get from school
psychologists who interpret the outcomes of an intelligence test. Why is an individualised
assessment of intelligence administered? What happens afterwards? These questions
needed to be answered in a second study. Therefore, a case study was carried out to
unfold the process between assessment and intervention.
VOICES FROM PRACTICE: WHEN IS THE GAP BETWEEN DIAGNOSIS AND INTERVENTION APPARENT?
111
5
Study 2: Case Study
MethodThe aim of the case study was to provide answers to the questions produced by the first
study, as mentioned above. The answers to these questions will provide an insight in the
gap between diagnosis and intervention.
Sample. Subject to the case study was the care process of a grade 1 student showing
persistent learning problems. The school in which the case study was carried out
received financial aid from the government, due to the diverse multicultural population
of students. This resulted in an additional grade 1 teacher for several days a week,
which allowed for teaching in smaller groups and individual assistances for students that
needed it the most.
Instruments. Observations during all key elements in the care process (such as
meetings, consultations, and test administrations), and observations in and outside
the classroom were undertaken. Moreover, all professionals involved, i.e. teacher and
additional teacher, school psychologist and special care coordinator, were interviewed
regularly and at specific moments (such as parental consultation) during the process
in order to gain insight in their beliefs and reasons for acting in the classroom. Finally,
student’s files and school’s policy documents were studied in depth.
Procedure. During a 4 month period the care process of the student has been followed
carefully. The study started in the second semester of grade 1 at the moment that the
school psychologist was consulted for screening the student (begin March 2012). The
study ended at the end of grade 1 (end of June 2012).
Analysis. For the trustworthiness of data (Lincoln & Guba, 1985) peer debriefing and
member checking techniques were carried out. During each step of the data collection
one or two other researchers critically reviewed subjective statements of the initial
CHAPTER 5
112
researcher. Information was checked with all participants involved. Moreover, data
triangulation from the multiple data sources mentioned above contributed to reliability
of results.
Results and interpretationWhy is an individualised assessment of intelligence administered in this case? The teacher, assistant
teacher, special care coordinator and the school psychologist decided together to start
an “individual assessment of intelligence” (IAI) to get more information about the
student’s problems. In the present case, the IAI comprised not only the administration
of a static intelligence test, but also an observation in the general classroom by the school
psychologist, and an analysis of the information provided by the teacher and special care
coordinator. However, in the school (i.e. teacher, special care coordinator, and assistant
teacher) the IAI is seen as “an estimation of intelligence of the student, and what to
expect from her” and is seen as a very valuable instrument. The IAI appeared to be a
tool to gain insight in the possible causes for learning delays of this student, the school
had high expectations for the outcomes of the IAI. The special care coordinator and
teacher expect advice from the school psychologist “in order to eliminate the student’s
learning delays, and how they should act with this student”.
Later on in the process, the IAI appeared to play a less important role in problem
identification. Student’s results to other instruments, such as curriculum related tests
and scores on nationwide screening instruments, and teacher’s observations appeared
to be more indicative for the ‘care’ process. The outcomes of the IAI were only additive
to these results. In this case, results of the IAI were surprising to the school, since
they “could expect more from the student than she shows at the moment”. At this
moment, the participants seem to have different ideas about the usefulness and goal
of the intelligence test administration. The assistant teacher indicates the influence of
contextual effects during the test administration by the school psychologist: “it is different
from the classroom situation”. Participants seem to pay less attention to the outcomes
of the test, since it is not in congruence with what they experienced in the classroom. In
VOICES FROM PRACTICE: WHEN IS THE GAP BETWEEN DIAGNOSIS AND INTERVENTION APPARENT?
113
5
sum, the diagnosis of the problems of this student seems not to be executed in a cyclic
process. Systematic evaluation of the outcomes to the care process of the child is, apart
from the meeting with the school psychologist, not evaluated between participants.
The reasons for the IAI are multiple: what can we expect from the student? What
are the causes for the problems? But specific answers to these questions do not seem
to be provided by the IAI. Other information sources (such as curriculum related
test, observations) seem to provide information that is needed to give answers to these
questions.
What happens afterwards? Already ahead of the results of the IAI it was clear that
the student needed more care in future: intensive individual remedial teaching was
planned. Currently, the student already experienced intensive support with regard to
Dutch (school language), and this support should be continued. In the care process,
however, no structure or vision was observed, nor did the care process change after the
administration of the IAI during the period of this study. Although the IAI showed that
they could expect better results from the student, the school did not change their ‘care’
after the IAI outcomes. Thus, a gap between diagnosis and intervention was observed.
Summary. Before the administration of the IAI, high expectations concerning the
outcomes of the IAI were set by the educational staff. At the end of grade 1, however,
the results of the IAI did not play a significant role anymore. The researcher observed
an overrated value of the IAI before the administration, and the gap between diagnosis
and intervention was not bridged by the IAI. On the one hand, the IAI seemed to
deliver important information for the school with regard to their expectations of the
student. On the other hand, however, the school could not deduce practical guidelines
from the IAI to develop a tailored program adapted to this particular student. While a
lot was expected from the results of the IAI, the effects of the test administration did
not translate to extra care and time for this student in practice, as could be observed
during our study.
CHAPTER 5
114
ConclusionTraditionally, an IQ test is regarded as the best instrument to find causes of learning
problems. One could state that a blind faith in intelligence testing could be observed
in the present case study. A lot was expected beforehand, but after the administration
nothing changed in the ‘care’ process around the student, which reflects a certain
discontinuity in the process. Different opinions about the usefulness of the IAI were
noticeable. Educational professionals act rather intuitively, on short term vision, with
respect to the care provided in this case.
General discussion
An overview of the current state of care to at-risk students in Dutch educational
practice is provided by both studies described above. The results of the first study, in
which questionnaires to school psychologists, special care coordinators, and teachers
were distributed, indicated an important role of intelligence tests. However, answers
were often ambiguous and large differences between participants were observed,
especially with regard to the gap between diagnosis and intervention. Some schools
use the information from an intelligence test solely, others use alternative methods
such as classroom observations. Moreover, different views about the role of the school
psychologists indicate a difference in ‘care’ processes between schools.
The significance of intelligence tests to the development of interventions remained
unknown. Therefore, a case study was carried out in the second study. The focus
of the case study was the use of intelligence tests and its consequences for the ‘care’
process of a grade 1 student. It appeared that before the administering took place, a
lot was expected: participants aspired to get a solution for, and clues for actions with,
the student and her specific problems. However, the care provided to the student did
not change after the administration of the intelligence test. In summary, there is a need
for guidelines, but these guidelines are not provided by the individualised assessment
VOICES FROM PRACTICE: WHEN IS THE GAP BETWEEN DIAGNOSIS AND INTERVENTION APPARENT?
115
5
of intelligence. Besides, during the period of time of our case study, evaluation of the
‘care’ process between professionals was not undertaken, which also indicates a gap
between diagnosis of problems and the interventions provided.
The studies underlined the need for other formats of assessment that provide specific
guidelines for practice. Generally, assessment procedures focus on estimation of cognition
of the at-risk student. It can be questioned, however, whether learning problems of at-
risk students always originate in cognition. Next to cognitive aspects, other aspects that
play a role in learning processes, such as metacognitive and motivational aspects, should
be taken into account when providing guidelines for adaptive education in the general
classroom. In the case study the school psychologist not only used information from
the intelligence test, but also information from the teacher and classroom observation.
Still, a gap between his advice and the actions of the educational professionals in the
school was apparent. This indicates the need for enhancement of professionals’ skills
with regard to interpretations of tests scores. Even though this school was known for its
quality of care, gaps in the process could be detected, for example evaluation was not
undertaken during the case study. As already highlighted by Rouse and Agbenu (1998),
teachers should develop their skills in observation and assessment in order to bridge the
gap between assessment and intervention.
It should be noted that both studies were carried out in a rather small, specific sample
which could have biased outcomes. The educational advisory agency has shown interest
in dynamic assessment, resulting in more knowledge among their school psychologists
compared to school psychologists working at other educational advisories. Moreover,
the case study has been carried out at a school which is able to provide more care due to
financial aids by the government. This is not representative of schools in other regions
of the Netherlands. It is estimated that the results described in this chapter are rather
positive compared to other Dutch schools.
CHAPTER 5
116
With respect to the questionnaires it must be mentioned that answers to questions
could have been subjected to individual cases and experiences of participants. When
answering, they might have thought about specific students, resulting in differences in
answers to the questions. Individual beliefs, thus, play a significant role in answers to these
questions. This relates to the issues of self-efficacy and implicit theories of intelligence
in educational practice. As described in the previous chapter of this dissertation, beliefs
about the malleability of intelligence have their impact on the actions of educational
professionals. Therefore, next to knowledge about (dynamic) assessment, these beliefs
should be taken into account when intervening in educational practice.
Several improvements are needed in current Dutch psycho-educational practice. Other
formats of assessment, knowledge about assessment and diagnosis, awareness of one’s
own beliefs and their influence to one’s own acting, and the need for evaluation between
significant others in the learning process of student, are examples of aspects to improve
in order to bridge the gap between diagnosis and intervention. Therefore, in the next
chapter, the results of an intervention study that aimed at the development of these
skills will be reported.
VOICES FROM PRACTICE: WHEN IS THE GAP BETWEEN DIAGNOSIS AND INTERVENTION APPARENT?
117
5
Appendix | Examples of open-ended items and categories in coding scheme
Item of questionnaire for teachers
Which information do you want from the school psychologist?
Assessment results
Advice on specific needs of student
Information about learning potential / intelligence of student
Information about level and manner in which instruction should be effectuated
I don’t need any information
Unclear or no answer
Otherwise specified, namely:
Item of questionnaire for special care coordinators
In what way do you provide support to the teacher when encountering students with learning problems?
Via conversations
I work with the student myself (e.g., assessments, observations)
I provide advise (about e.g., intervention programs)
I evaluate students’ results
Unclear or no answer
Otherwise specified, namely:
Item of questionnaire for school psychologists
Which information do you need to formulate your actions-based advise?
Information from teacher
Classroom observation
Information from parents (home situation)
Insight in opportunities at school
Test outcomes (information about student)
Unclear or no answer
Otherwise specified, namely:
Chapter 6The next step in bridging the gap: a pilot study
to enhance teacher’s skills with at-risk children
This chapter is based on:Tiekstra, M., Minnaert, A., Hessels, M., & Verdel, R. (in prep.).
The next step in bridging the gap: a pilot study to enhance teacher’s skills with at-risk children.
CHAPTER 6
120
Abstract
It has been demonstrated that teachers rarely provide instruction tailored to specific
needs of students. Teachers seem to lack these tailored oriented skills, especially in the
case of at-risk students. In this pilot study, an intervention was developed to enhance
teacher’s skills in the general classroom in order to foster students’ learning potential.
Since implicit theories of intelligence influence teacher’s behaviour these were also
taken into account. A short intervention of four sessions was implemented at a school
team (N=21) of one primary school. The sessions focused on mediational techniques,
learning potential, the use of (dynamic) assessment, and translation to classroom
practice. The results showed a significant change in beliefs of intelligence, but not
yet in teacher’s behaviour. Multilevel growth curve analysis showed the significant
contribution of self-efficacy to the increase of incremental views on intelligence. It is
recommended that in order to change teacher behaviour, beliefs and self-efficacy should
be taken into account.
THE NEXT STEP IN BRIDGING THE GAP: A PILOT STUDY
121
6
Introduction
Students that are at-risk for learning difficulties, or seen to be eligible for special education,
often lack problem solving skills. These students, therefore, need more guidance in the
classroom. Several studies demonstrated that it seems difficult for teachers to provide
guidance which is tailored to the students’ needs (Steenbeek, Jansen, & Van Geert, 2012;
Sutherland, Lewis-Palmer, Stichter & Morgan, 2008). Dymond and Russell (2004) showed
in their observational research that teachers rarely adapt their instruction to the needs of
students with mild intellectual disabilities, while they do for students with severe disabilities.
In the latter case, teachers are often supported by an educational psychologist or other
professional, and consequently, feel less responsible for the learning processes of these
students in the classroom. It is a striking conclusion that students with learning problems
are submitted to general classroom instruction without any adaptation. Generally, teachers’
actions should be based upon Individualised Educational Plans (IEP). However, problems
emerge when translating IEPs to the classroom. Clearly, teachers lack skills to adapt their
instruction to mild problems, which are eminent in the daily inclusive classroom practice.
Since teachers play a leading part in learning processes of students (e.g., Hattie, 2009),
there is a need for teacher professionalisation or interventions at teacher level that aim at
improving their skills for adequately teaching students with mild intellectual disabilities.
However, a focus on skills solely is not in congruence with theory, since teachers’ beliefs have
been demonstrated to also play a role in the instruction teachers provide. Teachers’ self-
efficacy has an impact on behaviour in the classroom (e.g., Tschannen-Moran & Woolfolk
Hoy, 2001). Self-efficacy refers to the belief in one’s own capacity to achieve a certain
performance, and is context dependent (Bandura, 1997). Imants and De Brabander (1996)
define teachers’ sense of efficacy as “the extent to which teachers believe they can affect
student learning” (p.179). More specifically, as demonstrated by Gibbs and Powell (2012),
teachers’ self-efficacy is composed of their belief in efficacy for classroom management,
children’s engagement, and instructional strategies. Previous experiences and feedback
from the environment influence self-efficacy.
CHAPTER 6
122
Furthermore, teachers’ actions in daily classroom practice are influenced by implicit
theories of intelligence (Tiekstra & Minnaert, submitted). Dweck and colleagues have
shown in their extensive research that people hold implicit theories of intelligence, with
either an incremental or entity view, and that these views affect people’s behaviour (e.g.
Dweck, Chiu, & Hong, 1995). In the incremental view intelligence is considered as
malleable, which is in contrast to the entity view in which intelligence is seen as something
stable and not subjective to change. Implicit theories of intelligence influence the
quality of interactions, as was demonstrated by Moorman and Pomerantz (2010) with
regard to mothers and their children. This also holds for teachers (Tiekstra & Minnaert,
submitted). For instance, teachers supporting the entity view tend to practice direct
instruction with, essentially, a focus on performances. Teachers with an incremental
view, on the contrary, support mastery-oriented learning and focus on the learning
process instead of the learning result. Jordan and Stanovich (2001) demonstrated that
teachers with an incremental view were more frequently involved with at-risk students
than teachers holding an entity view, since these teachers feel more responsible for
their students’ learning processes. Implicit theories of intelligence, thus, affect the way
in which instruction is effectuated in the classroom. Accordingly, students’ learning is
influenced by different teaching strategies (Rozendaal, Minnaert, & Boekaerts, 2005;
Timperly & Parr, 2009). This means that an incremental approach is important for the
classroom situation: intelligence should be seen as malleable, and every individual has
potential which could be fostered by teachers’ actions.
The intervention (theoretical background)The aim of this study was to explore the effectiveness of an intervention aiming at the
enhancement of teachers’ skills and incremental views in order to foster students’ learning
potential. An incremental view on intelligence is a prerequisite for fostering learning potential
in the classroom. Blackwell, Trzesniewski and Dweck (2007) demonstrated the effectiveness
of intervening on theories of intelligence. More specifically, students who were taught that
intelligence is malleable (incremental view) showed more positive motivational patterns
which, in turn, caused higher grades. This underlines the need for incremental views in
THE NEXT STEP IN BRIDGING THE GAP: A PILOT STUDY
123
6
education, and that intervening on such core beliefs can have huge impacts on students’
learning. The intervention in the study of Blackwell et al. (2007), however, was targeted at
students. In our intervention, in line with the above, we self-evidently intervene on the level
of teachers, since they play a crucial role in fostering a student’s learning potential.
To realise a student’s potential, teachers should explore and, accordingly, teach in the
student’s zone of proximal development (ZPD). According to Vygotsky (1978), only in this
way learning occurs. The teacher has a mediating role, and should be sensitive to the specific
needs of students. Feuerstein’s Instrumental Enrichment is an operationalisation of these
Vygotskian principles, and was originally developed for individualised intervention situations
(Feuerstein, Rand, & Hoffman, 1979). It is possible, though, to translate the intervention
principles to general classroom education. By asking mediating questions and focus upon
students’ learning processes, general cognitive teaching in the ZPD can be fostered.
Currently, teachers lack the skills and knowledge to execute such cognitive education, as
highlighted by Hessels and Hessels-Schlatter (2013).
Since teachers play a paramount role in students’ learning processes, they should be
reflective upon their own actions (Jennings & Greenberg, 2009; Steenbeek et al., 2012).
Insight in their own learning and acting has a positive influence on teachers’ effective
behaviour in individual differences of students (Rosenfeld & Rosenfeld, 2008). Moreover,
it has been demonstrated that reflecting and evaluating is an important feature of teacher
professionalisation (Meirink, Meijer, Verloop, & Bergen, 2009; Runhaar, Sanders, & Yang,
2010). Therefore, an important aspect of our intervention was to emphasise the role teachers
play in learning processes and to help reflect collaboratively on this issue.
In sum, the rationale of our intervention was that in order to develop a dynamic, (meta)
cognitive teaching style in teachers that fosters the students’ learning potential, one
should support an incremental view of intelligence. Since Meirink et al. (2009) showed
that changing teachers’ beliefs is a complex task, the focus of the intervention was
more on teachers’ activities in the classroom with at-risk children, than on changing
CHAPTER 6
124
their implicit theory of intelligence. In this way, the intervention was closely tailored to
teachers’ experiences in the classroom, accounting for the benefit of the implementation
of the intervention.
Interventions in educational contextsWhen implementing interventions one should take into account the ecology in which
the intervention takes place. Prince, Tiekstra, and Minnaert (2014) showed that the
educational context is composed of a web of agents who all have their own beliefs
and intentions, which causes a complex ecology for interventions. To eliminate
implementation bias, these authors stressed the necessity for collaboration, and a shared
vision and language among participants, including the researcher.
Yeager and Walton (2011) also emphasised the importance of taking contextual effects
into account when generalising effects of educational interventions. Several authors
(e.g., Boekaerts & Minnaert, 2003; Levin & O’Donnell, 1999; Sternberg et al., 2006)
recommend testing interventions in pilot studies first, before executing large scale
randomised control trials (for more information, see Prince et al., 2014). Accordingly,
the relevance and effectiveness of the content of our intervention was studied for the
first time by implementing it in one school only. In this way, the context remained
controllable over time and we could tailor the intervention to exclude contextual
influences and improve implementation. Although school-specific, the theoretical
principles of the intervention remained the same. The main research question in this
study addressed the extent to which the intervention accounted for a change in belief
and behaviour of the participants, taking teachers’ self-efficacy into account.
THE NEXT STEP IN BRIDGING THE GAP: A PILOT STUDY
125
6
Method
Participants The school that was selected for this study was a primary school in the western part
of the Netherlands. This area accommodates more immigrant families compared to
other parts of the Netherlands, implying a very diverse school population. This school,
however, can be characterised as a typical, average school.
Our emphasis on a “shared language” among school members demanded the
participation of the whole school team. The school board endorsed this viewpoint
and agreed to participate in the first training session to show their support. Generally,
support of directors is in favour of implementation (see Prince et al., 2014). Every
educational professional working at the school participated, which resulted in a total of
N= 21, all females.
The school team consisted of three elementary school teachers, seven middle school
teachers (grade 3 – 5), four teachers for the upper-level classes (grade 6 – 8), and for one
teacher it was unknown in which grade she taught. Next, six participants mentioned
that they taught in several classes, one of whom was a remedial teacher, one school
psychologist, the general coordinator (assistant director), and the special care coordinator
(assistant director). The mean working experience of the school team was 10,3 years.
Nine participants were very experienced (> 10 years), five had average experience (4 –
10 years), and five were less experienced (0 – 3 years). All participants will be referred to
as “teachers” in the rest of the article.
ProcedureThe intervention comprehended four two- hour sessions over a period of two weeks at
the beginning of the school year (September/October 2013). The study comprised a pre-
test, post-test, and a follow-up measure. The baseline measurement (pre-test, N =21) was
administered before starting the first training session (week 0). Immediately after the last
CHAPTER 6
126
training session, the post-test (N =15) was conducted (week 2). The follow-up measurement
(N =19) was carried out four weeks after the last training session (week 6).
Training. Several studies of A. Minnaert and colleagues (personal communication, March
11, 2013; May 23, 2014) showed that a short period of intensive training with a maximum
of two weeks provides more robust longitudinal effects than a training which lasts several
months. Moreover, Meirink et al. (2009) underlined the importance of a combination
of collaborative reflection with colleagues and immediate utilisation of new ideas in the
classroom when changing teacher’s beliefs. Therefore, four two-hour training sessions
distributed over a period of two weeks were carried out. In this way, learned material could
be tested in the classroom and each session started with a reflection on participants’ own
experiences. Sessions focused on mediational principles, reflection upon role of the teacher
in the learning processes, the use of assessment outcomes for setting up IEPs, and placed
emphasis on learning potential and the malleability of intelligence. See the appendix for a
more detailed description of the individual training sessions.
Implementation. In anticipation of the research, to warrant implementation, a series of
meetings was organised with a local educational advisory agency to develop the training.
In a next step, a meeting with the school board took place to tailor the intervention to the
specific needs of the school. As mentioned, the intervention aimed at the whole school team
to warrant a shared language within the school. The training was conducted by the first
author (who is an educational psychologist) together with two professional trainers from the
educational advisory agency. This assured control over the exact content of the training.
Moreover, observations and evaluations of each training session provided the feedback
needed to adjust the training where needed to guarantee its full applicability. The two
members of the school board who participated in the training provided the researchers with
feedback about the reactions within the school team after each training session, allowing
us to further adjust, or to attend to certain aspects more thoroughly.
THE NEXT STEP IN BRIDGING THE GAP: A PILOT STUDY
127
6
Measurements A questionnaire was administered to measure the effects of the intervention on several
variables listed below. Additionally, observations were carried out during the training
sessions and interviews after each training guided the implementation process.
Teacher behaviour. Teachers’ behaviours in the classroom were evaluated by
means of a questionnaire comprising statements with a vignette. Six questions followed
a vignette that described a specific classroom situation around a student in grade 3
with a learning problem. These focused on daily classroom practices, and had anchors
indicating either a dynamic (incremental) approach (4) or a static (entity) approach (1) of
the case. For example, item 2 addressed the individual approach to the specific student:
During a one-to-one moment with student M.,… I explain the exercise once again (static approach),
or …I ask the student how she is going to tackle the problem in the exercise (dynamic approach). The
other items referred to teacher’s behaviour in group wise instruction, student’s learning
problem, consultation of the special care coordinator, and setting up and evaluation of
an IEP. The higher the total score, the more dynamic is teacher behaviour. In this way
we could investigate whether teachers’ intentions in the daily classroom practice had
changed after our intervention. The vignette showed a reliability of Cronbach’s α .68 at
pre-test. A vignette similar to the pre-test (same classroom situation, different problems)
was administered six weeks later at the follow-up (α .61).
Implicit theories of intelligence. Next to behaviour, implicit theories of intelligence
(ITI) were assessed. The validated and widely used ITI-scale of Dweck et al. (1995) was
converted into items appropriate for Dutch teachers. The scale, comprising three items,
was administered at pre-test (α .85), post-test (α .82), and follow-up (α .88). A higher
score indicates a greater entity (static) view on intelligence (min. 3, max. 15).
Self-efficacy. Five items (pre-test, α .82) assessed self-efficacy of teachers. This scale
was retrieved from Opdenakker (2008) who tested the scale in different educational
contexts. Mean scale scores (min. 5, max. 25) were calculated to be used in the analysis.
CHAPTER 6
128
A higher score indicates a greater sense of self-efficacy.
Observations and interviews. Each training session has been observed and
evaluated by the trainers (including the researcher). Moreover, each session was
evaluated by the school board. They informed the researcher about reactions and
changes within their team during the intervention. After the administration of the
follow-up questionnaire (week 6) a final interview took place with several participants
of the training, and the school board in particular. This interview focused on perceived
differences within the school team.
AnalysisIn order to measure the effect of the training, descriptive analyses of scale and data
distributions were carried out first. Effect sizes (Cohen’s d) were calculated per item
measuring teacher behaviour. To minimise capitalisation on chance, it was decided that
only Cohen’s d’s of ≥ .50 accounted for a change.
The change in participants’ beliefs could be measured at post-test and follow-up.
Unfortunately, at post-test, we were confronted with a dropout rate of 33%. A detailed
analysis revealed that participants with a rather static view at pre-test did not participate
at post-test. A mixed model approach was employed to explore whether results were
independent of this dropout. Mixed models do not simply impute missing values, but
calculate a function to estimate these missing values, based on the functions that the
model estimated for the rest of the data. We conducted a specific kind of mixed models,
that is multilevel growth curve analyses in which measurements (time) were units within
participants, on the implicit theories variable. In this way, a trend analysis of the ITI
measure on pre, post and follow-up could be established independently of dropout.
It was further investigated whether self-efficacy influenced the outcomes in a next
multilevel growth curve model.
THE NEXT STEP IN BRIDGING THE GAP: A PILOT STUDY
129
6
The qualitative information from the final interview has been used to complement and
illustrate the results of these analyses.
Results
Change in behaviourVignette. The descriptive analysis of the pre-test showed that scores on the vignette-
scale were quite high already (M= 20,6 out of max. 24), and the effect size of difference
between pre-test and follow-up of the vignette as a scale appeared to be relatively
low (Cohen’s d -.09). However, detailed analysis of each item separately showed some
substantial differences in distributions of scores on pre-test and follow-up. Therefore,
effect sizes were calculated for each item separately. As showed in Table 1, these ranged
from -.43 to .65. The considerable effect of d. = .65 was found for item 2 referring to
individual instruction. Since a higher score implied more dynamic behaviour of the
teacher, a positive Cohen’s d was desired.
Table 1 | Overview of effect sizes per item measuring teacher behaviour
Item Cohen’s d
Group wise instruction -.43
Individual instruction .65
Student’s learning problem .09
Special care coordinator consultation -.41
Setting up IEP -.06
Evaluation of IEP -.27
CHAPTER 6
130
Interview. The final interview with the school board and several teachers revealed
that the school team had re-evaluated their group and individualised educational plans
several weeks after the intervention. Moreover, the school board remarked a difference
in some teachers with respect to their approach of students, and in the format of their
meetings about students. According to the school board, a trend was noticeable in the
change of behaviour of the school team.
Change in beliefImplicit theories of intelligence. Descriptive analysis revealed a trend in the mean
scores of the participants: from a relatively strong entity view on intelligence at pre-test
(M = 7,6) to a rather incremental view at post-test (M = 5,6). This trend flattened out at
follow-up (M = 6,4). We investigated whether these results were biased by the dropouts
at post-test, which turned out not to be the case. The multilevel quadratic growth curve
model fitted the data best. The decline in static view appeared to be significant (Time p
<.001), as was the upward trend between post-test and follow-up (Time2 p .003). Table
2 shows the modelling results. A picture of the trend in implicit theories of intelligence
can be seen in Figure 1.
THE NEXT STEP IN BRIDGING THE GAP: A PILOT STUDY
131
6
Table 2 | Multilevel growth curve model of implicit theories over time (N = 54)
Implicit theories of intelligence
Coefficient SE
Fixed effects
Intercept 7.619* .466
Time -2.600* .811
Time2 .755* .256
Random effects
Level 2 variance
Intercept 2.086 1.455
Time .049 .279
Intercept*time .080 .472
Time2 .000 .000
Intercept*time2 .000 .000
Time*time2 .000 .000
Level 1 variance
Residual 2.476* .938
Deviance 229.486
Note. * p ≤ .01
CHAPTER 6
132
Figure 1 | The development of implicit theories of participants at pre-test, post-test and follow-up (with confidence interval), independent from dropouts at post-test. The higher a score, the more entity oriented a view.
Dw
eck
fact
or
0,0 1,1 2,2
time
3,6
5,4
7,2
9,0Implicit theory of intelligence
Self-efficacy and implicit theories of intelligence. The question arose whether
self-efficacy influenced the curve of the model. Another quadratic multilevel growth
curve model, including self-efficacy intercept, and interactions of self-efficacy with time
and time2 (quadratic part), appeared to be the best fitting model to the data. Table 3
shows the outcomes of this analysis. The fixed effect of self-efficacy is not significant.
However, the interaction with time is nearly significant (self-efficacy*time p = .051) at
5% level. The upward trend is also significant at 10% level (self-efficacy*time2 p = .095).
Figure 2 provides more insight into this issue. The decline between pre-test and post-test
is different for participants with more or less self-efficacy.
THE NEXT STEP IN BRIDGING THE GAP: A PILOT STUDY
133
6
Table 3 | Multilevel growth curve model of implicit theories over time, taking self-efficacy into account (N = 51)
Implicit theories of intelligence
Coefficient SE
Fixed effects
Intercept 8.583** 3.214
Time 7.491 5.224
Time 2 -1.971 1.638
Self-efficacy -.221 .797
Self-efficacy * Time -2.549 † 1.31
Self-efficacy * Time 2 .686 † .411
Random effects
Level 2 variance
Intercept 1.791* .908
Time .000 .000
Intercept*time .000 .000
Time 2 .000 .000
Intercept*time 2 .000 .000
Time*time 2 .000 .000
Level 1 variance
Residual 2.404** .605
Deviance 210.494
Note. * p ≤ .05, and ** p ≤ .01, and † p ≤ .10
CHAPTER 6
134
Figure 2 | The influence of self-efficacy on scores on implicit theory of intelligence (the higher a score, the more entity oriented a view) on pre-test, post-test, and follow-up. The line at the top indicates a relative low mean score (3) on self-efficacy, whereas the line at the bottom implies a high mean score (5) on self-efficacy.
The higher self-efficacy at the start of the intervention, the more decline in static views
(nearly at 5% alpha level), and the more maintenance effect of this decline in the follow-
up measure. It should be noticed, however, that the interaction effect in the quadratic part
(upward line between post-test and follow-up) was significant at 10 % alpha level. This is
related to power problems: the small sample size affects the level of alpha (Carmines &
Zeller, 1979). The influence of self-efficacy in the enhancement of an incremental view is
shown in figure 2. These results indicate a differential effect in which self-efficacy makes a
difference in the long term in favour of the incremental view.
Dw
eck
fact
or
0,0 1,1 2,2
time
3,6
5,4
7,2
9,0SEgem.pred = 3SEgem.pred = 3,5SEgem.pred = 4SEgem.pred = 4,5SEgem.pred = 5
THE NEXT STEP IN BRIDGING THE GAP: A PILOT STUDY
135
6
DiscussionThe goal of this pilot study was to investigate whether teachers’ skills to effectively adapt
their instruction in order to foster student’s learning potential in the classroom could
be enhanced. A short, intensive, intervention was implemented in one school team (N=
21) to explore the extent to which their behaviour and beliefs would change during
the intervention. The results showed a substantial change in participants’ beliefs about
intelligence, but did not provide enough proof for a change in teacher’s behaviour.
Change in beliefMultilevel growth curve analysis showed a significant decline in the trend of static
implicit theories of intelligence after the training: the rather static (entity) views at the
beginning changed in more dynamic (incremental) views immediately after the training,
which remained present at follow-up, although to a lesser extent. The intervention
provided more incremental views on intelligence among participants which sustained
over (at least) six weeks of time.
The effectiveness of the training, however, was moderated by self-efficacy of participants.
More self-efficacy at the start of the project seemed to produce more decline in static
beliefs about intelligence over time. This illustrates the need for emphasis on self-efficacy
in teacher professionalisation, implying that actions in the classroom should be a key
aspect in interventions in order to achieve effects. Teachers learn from their actions,
which demands for a recursive design of studies (see also Boekaerts & Minnaert, 2003).
In this pilot study, teachers were trained in metacognitive teaching and reflection upon
their own teaching. Although the change in beliefs of participants was a promising
result, reflection, an explorative attitude towards students’ learning problems, and
metacognitive teaching should be embodied in teacher education. The teacher’s belief
in own actions in the classroom may be an important contributor to incremental views
on intelligence in educational practice, as well as it might be to skills for metacognitive
teaching.
CHAPTER 6
136
Change in behaviourDue to the restriction in measurement of change in behaviour in this pilot study, it remained
unclear whether these changes in beliefs affect changes in classroom situations. Some
outcomes indicated a change in behaviour. The vignette study showed a significant change
with a large effect size in a more dynamic approach of individual learning situations in
this school. Due to an unfortunate ceiling effect, however, the vignette can be regarded as
not discriminative enough. Moreover, it was a self-report measure, and there was a lack
of observations in the classroom that could have been used to assess change in teacher’s
behaviour. Nevertheless, the final interview with the school board and several teachers
revealed some promising indications for change of behaviour taking place in the school
team. Perhaps, the information of the intervention needed more time to be assimilated
by the teachers, which implies that the results in change of behaviour should have been
measured after a longer period of time.
The change in belief about IQ testing is an important finding with respect to the
implementation of dynamic assessment (DA) in practice. As highlighted by several
authors (e.g., Deutsch & Reynolds, 2000; Elliott, 2003), dynamic assessment is only
scarcely implemented in practice, regardless of many advantages over static testing in
at-risk populations. As a next step in bridging the gap between assessment and classroom
instruction, this pilot study showed that specific training is needed in order to change the
beliefs about IQ testing within school teams. In order to effectuate full implementation
of DA in educational practice, specific training of educational psychologists in dynamic
assessment is needed. Yet, being another important aspect to this issue, teachers too,
should embody a general “incremental attitude” towards learning (Hessels & Hessels-
Schlatter, 2013).
Future directionsThe current pilot study revealed some important aspects to consider in future research.
As Yeager and Walton (2011) underlined “an incremental mindset intervention might
have no effect if students believe that the person who tells them about their potential
THE NEXT STEP IN BRIDGING THE GAP: A PILOT STUDY
137
6
for growth and improvement does not believe this himself or herself.” (p.290). Although
our intervention focused on behaviour of teachers in the classroom, a positive change
at student level was desired. Therefore, we recommend taking students’ outcomes into
account in future research.
Furthermore, since the pilot study was conducted at one single school, the results could
be subject to specific characteristics of this school. Studies at multiple schools should
be performed to investigate the school effects. Schools in which the intervention is
implemented should be compared to schools in a control group. Observations in the
classroom context should be carried out instead of self-report measures. Moreover,
results should be measured in a long term perspective, to investigate the maintenance
effect of change in belief and behaviour.
This pilot study showed that a short intervention changed participants’ static beliefs
about intelligence into more incremental (dynamic) views. Despite the fact that a change
in one’s own beliefs is influenced by one’s individual pre-existing beliefs (Meirink,
et al., 2009), the whole school team demonstrated a decline in static views. This is
a promising outcome and proves ground for future research in the issue of fostering
student’s potential in the classroom. The decline in static views over time was subject
to participants’ self-efficacy. In order to change teacher behaviour, the contribution
of beliefs and self-efficacy to this issue should not be disregarded. Therefore, when
trying to change behaviour in interventions, beliefs and self-efficacy should be taken
into account.
CHAPTER 6
138
Appendix |Description of the training sessions
Training session 1
Goal To experience the connection between vision on individual differences and one’s own teaching actions in the classroom
Contents Participants’ views on intelligence How do you act upon individual differences in intelligence?
Intended aims
belief
behaviour
Individual differences demand an incremental view on intelligence
A shared language among members school team
Training session 2
Goal To learn about the role of the teacher in student’s learning process
Contents Zone of Proximal Development Role of the teacher as mediator in individual learning processesMetacognitive teaching in individual learning situations
Intended aims
belief
behaviour
Every student can learn (= incremental view on intelligence)
Enhancing skills in metacognitive teaching
THE NEXT STEP IN BRIDGING THE GAP: A PILOT STUDY
139
6
Training session 3
Goal To learn about the role of the teacher in the general classroom and at-risk students
Contents Steps in cognitive and motivational developmentSetting up group vs. individual educational plansSetting up “dynamic” educational plansAt-risk students in the classroom and the role of the teacher
Intended aims
belief
behaviour
Teachers play central role in intellectual development student
Setting up educational plans as guidance to tailored instruction; focus on student’s outcomes & teacher behaviour when reflecting upon plans
Training session 4
Goal To learn how to use (meta)cognitive teaching strategies (“mediation”) in daily classroom practice
Contents Mediational techniques during dynamic assessmentLink to classroom situation: mediational techniques in practice
Intended aims
belief
behaviour
Incremental view on learning
Enhancing metacognitive skills in group wise and individual learning situations
“I cannot teach anybody anything. I can only make them think.”(Socrates)
Chapter 7General Discussion
CHAPTER 7
142
GENERAL DISCUSSION
143
7
The first chapter of this dissertation started with the case of Yamina. This example from
daily classroom practice illustrated the gap that presently exists between assessment and
instruction. In an attempt to bridge this gap, more information was needed about the
consequential validity of assessment procedures. The aim of this thesis was to explore how
the consequential validity of (dynamic) assessment could be increased. This can be seen as a
methodological and theoretical approach to the “bridging the gap-issue” which is present in
educational practice. Therefore, the outcomes will be reviewed in this chapter by means of
another description of the case of Yamina:
When the problems of Yamina worsened, the teacher and remedial teacher decided to take action. The teacher
developed a series of measurements that linked with his instruction, and started to monitor Yamina’s progress to
gain more insight in her response to his instruction. He learned a lot with respect to his own teaching when he
interpreted Yamina’s results to these curriculum based measurements. Based on these evaluations the remedial
teacher in turn decided to provide reading instruction in small groups of students that make the same kind of
mistakes as Yamina. Although the adaptations in the instructions seemed to improve Yamina’s reading results,
she still had problems with finishing her tasks and keeping up with her classmates. The teacher and remedial
teacher put in a request for an individualised assessment by the school psychologist. The school psychologist
listened carefully to the experiences of the teacher and remedial teacher and decided to administer a dynamic
assessment procedure to gain more insight in Yamina’s problems. The learning potential test estimated Yamina’s
potential as much higher than expected. Moreover, the school psychologist remarked that Yamina responded well
to his metacognitive hints during the assessment. He also noticed that Yamina revived when she experienced her
successful responses to the test. The school psychologist was able to write a detailed report from the outcomes of
the dynamic assessment procedure. His advice was to especially support Yamina in the metacognitive aspects
of learning, such as planning and monitoring, and not to focus on her cognitive learning outcomes alone. Since
Yamina has a good potential for learning, the support in the metacognitive area may help her to further develop
her reading skills. The school psychologist predicted that Yamina would be much happier, and insisted upon
evaluation with the teacher and remedial teacher in three weeks.
This can be seen as a mindful and ideal situation based on a grounded assessment in which
the consequences of test outcomes serve as a guiding principle for classroom instruction and
CHAPTER 7
144
intervention: the gap between diagnosis and instruction has been bridged! This dissertation
explored the issues that are needed in order to achieve such an ideal situation in educational
practice.
Summary of findings As explained in chapter one bridging the gap not only involves a critical review of current (dynamic)
assessment procedures, but also a critical view on the significant others in educational practice.
Both sides have been investigated in this thesis. Chapters two and five focused at the consequences
of current (dynamic) assessment, chapters three, four, and six focused at significant others in the
learning processes of at-risk students.
Since dynamic assessment can be seen as of added value to traditional IQ testing in at-risk students
(see chapter 1), and the learning phase during administration could provide useful guidelines for
practice, the literature review in chapter two focused on current dynamic assessment procedures.
A critical review of 31 articles that met the inclusion criteria, in which 29 different assessment
procedures were described, revealed that, although promising, these procedures did not yet
provide explicit indications for practice. While proximal consequential validity was warranted
(i.e., test administration can be adapted to the responses of the student), it remained questionable
to what extent distal consequential validity was fostered: the majority of tests resulted in a number
(raw score) and only few also included an indication of the level of help needed, but no explicit
guidelines resulted from the procedures to underpin interventions. To assess the potential value
for bridging the gap, we focused at the learning phase of each procedure. It appeared that during
most of the procedures the examiner addressed cognitive and/or metacognitive aspects, whereas
motivational aspects never played an explicit role. This is a striking conclusion, since dynamic
assessment is known for the attention to non-intellective factors. In sum, although dynamic
assessment procedures have potential, they do - in practice - not contribute to bridging the gap
between assessment and instruction in their current formats.
In chapter three we elaborated on the extent to which a student responded to quality
instruction. During dynamic assessment qualitative one-to-one instruction is provided,
GENERAL DISCUSSION
145
7
and the extent to which a student profits from this instruction is being measured. In
other words, the extent to which a student is didactically resistant, one of the criteria for
the diagnosis of learning disabilities, is being assessed. But how can this be translated
to general classroom practice? The RTI model seemed to be an instrument to make
this translation possible. In a qualitative multi-method study, the RTI model was
implemented in Dutch educational practice. The results showed that the RTI model
can be a valuable tool, but that the way in which the model is implemented is the
decisive factor. Not only behaviour of participants needed to change (i.e., they executed
the model), but more importantly, their beliefs about teaching and didactic resistance
needed to change to make meaningful interpretations about student’s responses to their
instruction. We concluded that the RTI model can be a valuable tool to measure didactic
resistance, but only if teachers have a dynamic view on learning and instruction, and
thus, acknowledge their role in student’s learning processes.
We regarded this division between the change of belief and the change of behaviour
of participants as an important aspect in improving the consequential validity of
assessment procedures. Therefore, we wanted to explore the influence of beliefs on
teaching behaviour of professionals in the Dutch educational context more in-depth.
More specifically, the extent to which implicit theories of intelligence influence behaviour
in the case of at-risk children has been investigated in a quantitative study in chapter
four. It appeared that implicit theories of intelligence play a prominent role in daily
classroom practice. In this sample, 34% of the total variance in the actions of teachers
can be explained by their implicit theories of intelligence, and in denominational
schools this percentage is even 61%. This means that teaching in the Zone of Proximal
Development is influenced by teachers’ implicit theories of intelligence. Furthermore,
a full mediational structural equation model was found in the group of support
professionals: implicit theories predict the belief about IQ tests, and, in turn, explains
the way people act according to the outcomes of these tests. The consequential validity
of tests is, thus, caused by people’s implicit theories of intelligence.
CHAPTER 7
146
Since implicit theories of intelligence, and herewith the faith in testing results, seemed to be
important for improving consequential validity, we wanted to investigate what role tests and
testing results play in practice to date. Therefore, chapter five reported on two qualitative in-
vivo studies. In the first study, an open-ending questionnaire amongst 36 school psychologists,
21 special care coordinators, and 44 teachers, revealed ambiguous answers about the use of
test results, and the tasks of significant others in educational practice. Therefore, a case study
has been set up in the second study. This study also confirmed the gap between assessment
and instruction: the teacher as well as the special care coordinator indicated a need for
guidelines. However, the results of the IQ test did not provide specific information to guide
their instruction. More specifically, the care for and instruction to the student did not change
after the administration of the IQ test. To conclude, these in-vivo studies underlined the
necessity for other formats of assessment in current Dutch educational practice.
The above-mentioned studies showed the need for more knowledge and skills among
educational professionals in order to foster students’ learning potential in the general
classroom. The influence of implicit beliefs concerning the malleability of intelligence
should be recognised. In chapter six we, therefore, reported on a pilot study in which a short
intensive intervention was implemented with one school team (N=21). This pilot study
provided insight in the processes that are needed to change current psycho-educational
practice in order to bridge the gap between assessment and instruction. It appeared that
the short intervention accounted for a change in beliefs of participants: a decline in static
views was visible among the whole school team. Besides, the multilevel growth curve model
showed that the decline in static views was subject to participants’ self-efficacy: the more
self-efficacy one has, the more decline in static views, even in the long term. A change in
behaviour could, unfortunately, not yet be proven due to restrictions in the measurements. It
was concluded that if we want to change behaviour of educational professionals, we should
also take beliefs and self-efficacy into account.
All studies underlined the complexity and interrelatedness of several factors with respect to
fostering student’s potential to learn in the general classroom. In the next sections theoretical
GENERAL DISCUSSION
147
7
and methodological considerations will be addressed. Furthermore, implications for
practice and policy will be highlighted. Finally, conclusions will be listed. The ideal case
of Yamina described above will be used to illustrate these considerations.
Theoretical considerationsMany types of, and approaches to, validity exist (Kerlinger & Lee, 2000). This thesis is
centred on the concept of ‘consequential validity’. Although Messick (1989) underlined
the fact that this concept should not be seen as a separate type of validity, but rather
as an aspect of construct validity, emphasis should be put on the importance of his
message, and we therefore used the term separately. Messick (1995) stated that “validity
is not a property of the test or assessment as such, but rather the meaning of test scores”
(p. 5). The application, and herewith the consequence(s), of test scores is a type of
validity that should not be neglected. Especially within the context of learning and
development, where the ultimate purpose of assessment is to guide actions in the
classroom, consequences are of paramount importance in validity research. Therefore,
consequential validity can be seen as the ultimate type of validity, while predictive and
content validity can be considered as conditional stages to be reached before one can
say something about the extent of consequential validity. Therefore, test developers
should never disregard the consequences in validity studies.
Currently, consequential validity does not play a central role in validity studies of
assessment. This is a missed opportunity, since consequential validity can be seen as
the key element in bridging the gap between assessment and instruction in practice.
This gap, as shown in chapter five, is still present in daily classroom practice. Dynamic
assessment seems promising for bridging this gap. During dynamic assessment the
proximal consequential validity is warranted. After all, the examiner adapts hints
and feedback to the responses of the examinee in the learning phase: instruction is
influenced by test performance. A careful administration of these adaptations could
enhance the distal consequential validity of the test procedure. The administration
of hints and feedback that the student needs to solve the task relate to adaptations
CHAPTER 7
148
that could be made in instruction to optimise student’s learning. In short, the learning
phase during dynamic assessment plays a prominent role in bridging the gap between
assessment and instruction.
If dynamic assessment procedures strive to measure learning potential, then significant
others need to develop skills in fostering the learning potential. The beliefs of these
significant others play a prominent role in fostering learning potential. People’s implicit
theories of intelligence influence the way test scores are used. Furthermore, test scores
affect the way people think about abilities of the student. With respect to the case of
Yamina, the consequences of the individual assessment by the school psychologist could
have been different to the individual assessment (same test, same student) administered
by another school psychologist. It is, therefore, important that tests provide valid
scores. Moreover, educational professionals who work with (at-risk) students should
have an incremental view on intelligence, i.e., intelligence being seen as malleable. An
incremental view on intelligence can be seen as a prerequisite for the implementation
of dynamic assessment in current educational practice.
Moreover, the incremental mindset is a prerequisite for the operationalisation of
‘didactic resistance’. The concept should be measured at both student and teacher level.
After all, it is hard to say anything meaningful about the extent to which a student is
didactically resistant without knowing what kind of instruction has been offered by the
teacher or remedial teacher. In the ideal situation described above, Yamina’s teacher
decided to undertake action in order to examine Yamina’s didactic resistance. The
teacher evaluated his own teaching by curriculum based measurements, the remedial
teacher tried to adapt small group instruction to Yamina’s needs, and an evaluation
moment with the school psychologist will take place in three weeks. In other words,
an incremental view of educational professionals accounts for the acknowledgment of
their influence to the learning processes of their students. Only in this way, the extent
of ‘didactic resistance’ can be measured.
GENERAL DISCUSSION
149
7
Methodological considerationsThe two-sided approach to the issue of consequential validity, i.e. the focus upon tests on
the one hand, and significant others on the other hand, demanded for multiple studies to
be carried out. This explorative approach can be regarded as a methodological strength
of this thesis. The combination of qualitative and quantitative methods accounted for
a strong focus upon contextual influences. Therefore, the conclusions can be seen as
ecologically sound. Moreover, the different studies are strongly connected, accounting
for the internal validity of the thesis. Each study was anchored on the previous one,
exploring an issue more in-depth, or using a different approach to the problem.
A recurrent problem, however, was the attrition of participants. Apparently, the studies
addressed emotive subjects for participants in the field. After all, views on learning and a
critical view on actions undertaken with at-risk children come very close to participants’
actions and their sense of self-efficacy. In some participants this might have provoked
resistance, resulting in an avoidance of interest in the studies undertaken. As addressed
in chapter three and six, the group that dropped out of the studies did not adhere to
or understand a dynamic view on learning. I will further address this issue in the next
section.
The strengths of the thesis, the multiple explorative studies and a strong focus on the
context, leads us to the weaknesses of the thesis. Since the studies were carried out in
small sample sizes, one could question the generalisability of the outcomes to educational
practice in the Netherlands, or even abroad. The tension between statistics (focus on
generalisability) and need for individual approach in the case of at-risk children is
emphasised in this thesis. Although quantitative information provides knowledge about
general processes in at-risk students (e.g., the role of working memory in many learning
deficits), it does not provide information that is needed in at-risk cases that are often
surrounded by individual and specific characteristics. Qualitative information is needed
to reveal the specific processes that cause the problems of individual at-risk cases: for
example, one dyslexic student can be very different from another, and consequently,
CHAPTER 7
150
different interventions are needed. Accordingly, a mixed method approach, in which
both quantitative and qualitative methods are applied, is needed when studying consequential
validity. After all, the context plays an important role with respect to consequential validity,
which can be best studied by qualitative methods.
The explorative studies showed the importance of beliefs when changing behaviour. This
can be seen as a first step in ameliorating the implementation of dynamic assessment in
educational practice. The matter is, however, not yet settled. If we wanted to improve
consequential validity, the change in behaviour should have been measured, too. The
next step is, therefore, to focus on the transfer of beliefs to behaviour. More importantly,
future research should focus on the measurement of teaching behaviour of educational
professionals, especially in the case of at-risk children. However, unmanageable processes
that are characteristic for educational practice make it hard to study behavioural change.
In this light, observations could be combined with other, more generalisable, instruments.
Hence, the use of mixed methods is the best way to investigate the complex system of beliefs
and behaviour.
Implications for practice and policyCurrent educational policies demand for inclusion of students with all kind of disabilities in
the regular classroom (UNESCO, 2005). In the Netherlands, more particularly, the policy
of “Adaptive Education” forces schools to provide targeted care (a term in line with the
DSM-V) for all their students. If we want this targeted care to be effectuated properly, then
changes are needed in current educational practice in the Netherlands. In this process,
bridging the gap between assessment and instruction is an absolute requirement to improve
educational practice. After all, adaptive education demands for an individual approach to
students’ learning processes. Moreover, adaptive instruction demands for a critical approach
to the instruction provided. Besides, as demonstrated by this dissertation, if we want to
change behaviour, beliefs should be addressed first. Next to educational professionals in
the schools, such as teachers, remedial teachers, and care coordinators, this involves school
psychologists, test developers, and policy makers.
GENERAL DISCUSSION
151
7
The process from assessment to instruction is composed of several steps. Test scores
result from the test administration. These need to be interpreted and translated to
the educational practice accurately. In order to provide quality instruction, finally,
instruction should be adapted to the outcomes of the measurements. Several stakeholders
interact in these different steps. If we want to bridge the gap between assessment and
instruction, and accordingly, increase the consequential validity of assessment, the next
recommendations should be followed.
Firstly, recommendations at the level of test developers will be addressed. Test developers
should think about the consequences strived for when their test is administered. With
this, they should consider as well how to prevent bias in interpretations of the results of
their tests. With respect to the development of dynamic assessment procedures, progress
can be made to result in explicit clues for practice by administering the processes
during learning phases carefully. There should be more emphasis on motivational and
metacognitive aspects. A qualitative analysis of the aspects deployed and the hints being
given will provide explicit guidelines for the significant others around the student being
tested. The focus upon these consequences in educational practice will contribute to an
improvement of the quality of tests, and should, therefore, have a prominent place in
test evaluation reports (COTAN).
This paves the way for the second level of recommendations. School psychologists
have the responsibility to communicate the outcomes to educational professionals
in the schools, such as (remedial) teachers. They should be able to make accurate
interpretations of the test scores, and more prominently, to translate these outcomes to
educational practice, in such way that instruction can be adapted to individual needs.
This leads us to a critical review of the role of the school psychologist in practice.
Currently, tasks and actions of school psychologists are rather focused on classification,
diagnosing, and decision making. The focus is not on providing specific guidelines for
practice, whereas professionals “IN” the schools urgently need this information. This
indicates a need for specific training in translating test results to guidelines during early
CHAPTER 7
152
career, as well as professional courses later on. Professional organisations at national
level (such as the NVO in The Netherlands) should take a leading role in this. Moreover,
school psychologist should be aware of their own beliefs about intelligence and learning
when administering and interpreting tests.
Next to school psychologists, all professionals in the schools (i.e., teachers, special care
coordinators, and even school administrators) should acquire a disposition focused
on the malleability of intelligence. During initial teacher training emphasis should
be placed upon the crucial role teachers play in the learning processes of students.
Teachers’ implicit theories of intelligence influence the amount of teaching in, and
exploration of, the Zone of Proximal Development of a student. This implies that the
quality of teaching a student receives depends on which teacher provides the instruction.
The instruction is influenced by implicit theories of intelligence. The third level of
recommendations, thus, addresses the educational professionals who are responsible for
adapting their instruction regularly to the individual needs of students. Skills need to
be developed of all these professionals. The case of Yamina can be helpful to illustrate
this statement. Since Yamina appeared to have a good potential to learn, and currently
does not show enough progress, it should be evident that instruction should be adapted.
It might be the case that the remedial teacher and the teacher respond differently to
this advice. It could be that, after a while, the teacher is not convinced about Yamina’s
learning potential anymore since her school results do not improve despite several
adaptations he made. It could be that the remedial teacher does not know how to
include metacognitive teaching during the sessions with Yamina. It might be that the
teacher knows how to do this, but is not sure about his own skills. If they both do not
acknowledge their own expertise in the learning processes of Yamina, and do not feel
responsible enough, then nothing will change in the instruction provided to Yamina. As
a result, her future learning outcomes will not progress, despite her potential to learn.
This dissertation underlined the urgent need for more emphasis on beliefs and
behaviour of educational professionals in the case of at-risk students. Only in this way
GENERAL DISCUSSION
153
7
the consequential validity of (dynamic) assessment will be warranted. Eventually this
will lead to bridging the gap between assessment and instruction, resulting in quality
instruction, especially to at-risk students in the general classroom.
References
156
Altheide, D. L., & Johnson, J. M. (2011). Reflections on interpretive adequacy in qualitative research. In N. K. Denzin & Y. S. Lincoln (Eds.), The SAGE handbook of qualitative research (4th ed., pp. 581-594). Los Angeles, CA: Sage.
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.
Ames, C. (1992). Classrooms: Goals, structures, and student motivation. Journal of Educational Psychology, 84, 261-271. doi:10.1037/0022-0663.84.3.261
Antón, M. (2009). Dynamic assessment of advanced second language learners. Foreign Language Annals, 42, 576-598.
ATLAS.ti (Version 6.2) [Computer software]. Berlin: Scientific Software Development GmbH.
Azjen, I. (2005). Attitudes, personality and behavior (2nd ed.). Milton Keynes, England: Open University Press.
Bandura, A. (1993). Perceived self-efficacy in cognitive development and functioning. Educational Psychologist, 28, 117-148. doi:10.1207/s15326985ep2802_3
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.
Baudson, T. G., & Preckel, F. (2013). Teachers' implicit personality theories about the gifted: An experimental approach. School Psychology Quarterly, 28, 37-46. doi:10.1037/spq0000011
Bazeley, P. (2013). Qualitative data analysis: Practical strategies. London: Sage.
Beckmann, J. F. (2006). Superiority: Always and everywhere? On some misconceptions in the validation of dynamic testing. Educational and Child Psychology, 23(3), 35-49.
Blackwell, L. S., Trzesniewski, K. H., & Dweck, C. S. (2007). Implicit theories of intelligence predict achievement across an adolescent transition: A longitudinal study and intervention. Child Development, 78, 246-263. doi:10.1111/j.1467-8624.2007.00995.x
Boekaerts, M., & Minnaert, A. (2003). Measuring behavioral change processes during an ongoing innovation program: scope and limits. In E. De Corte, L. Verschaffel, N. Entwistle, & J. Van Merriënboer (Eds.), Powerful learning environments: Unravelling basic components and dimensions. (pp. 71- 87). Oxford, England: Pergamon/Elsevier Science.
Bol, L., Stephenson, P. L., O'Connell, A. A., & Nunnery, J. A. (1998). Influence of experience, grade level, and subject area on teachers' assessment practices. Journal of Educational Research, 91, 323- 330. doi:10.1080/00220679809597562
REFERENCES
157
Ref
eren
ces
Borkowski, J. G., Carothers, S. S., Howard, K., Schatz, J., & Farris, J. R. (2007). Intellectual assessment and intellectual disability. In J. W. Jacobson, J. A. Mulick, J. Rojahn, J. W. Jacobson, J. A. Mulick & J. Rojahn (Eds.), Handbook of intellectual and developmental disabilities. (pp. 261-277). New York: Springer Publishing Co. doi:10.1007/0-387-32931-5_14
Bosma, T., Hessels, M. G. P., & Resing, W. C. M. (2012). Teachers’ preferences for educational planning: Dynamic testing, teaching’ experience and teachers’ sense of efficacy. Teaching and Teacher Education, 28, 560-567. doi:10.1016/j.tate.2012.01.007
Bosma, T., & Resing, W. C. M. (2006). Dynamic assessment and a reversal task: a con tribution to needs- based assessment. Educational and Child Psychology, 23(3), 81-98.
Bosma, T., & Resing, W. C. M. (2008). Bridging the gap between diagnostic assessment and classroom practice. Journal of Cognitive Education and Psychology, 7, 174-198. doi:10.1891/194589508787381854
Büttner, G., & Hasselhorn, M. (2011). Learning disabilities: Debates on definitions, causes, subtypes, and responses. International Journal of Disability, Development and Education, 58, 75-87. doi:10.1080/1 034912X.2011.548476
Carlson, J. S., & Wiedl, K. H. (1992). Principles of dynamic assessment: The application of a specific model. Learning and Individual Differences, 4, 153-166. doi:10.1016/1041-6080(92)90011-3
Carmines, E. G., & Zeller R. A. (1979). Reliability and validity assessment. In J. L. Sullivan & R. G. Niemi (Series Ed.), Sage University Paper series on Quantitative Applications in the Social Sciences, 07-017. Beverly Hills, CA: Sage Publications.
Cavendish, W. (2013). Identification of learning disabilities: Implications of proposed DSM-5 criteria for school-based assessment. Journal of Learning Disabilities, 46, 52-57. doi:10.1177/0022219412464352
Chak, A. (2001). Adult sensitivity to children's learning in the zone of proximal development. Journal for the Theory of Social Behaviour, 31, 383-395. doi: 10.1111/1468-5914.00166
Compton, D. L., Fuchs, D., Fuchs, L. S., Bouton, B., Gilbert, J. K., Barquero, L. A., . . . Crouch, R. C. (2010). Selecting at-risk first-grade readers for early intervention: Eliminating false positives and exploring the promise of a two-stage gated screening process. Journal of Educational Psychology, 102, 327-340. doi:10.1037/a0018448
Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78, 98-104. doi:10.1037//0021-9010.78.1.98
Daniel, M. H. (1997). Intelligence testing: Status and trends. American Psychologist, 52, 1038-1045. doi:10.1037/0003-066X.52.10.1038
158
Datnow, A., & Castellano, M. (2000). Teachers' responses to success for all: How beliefs, experiences, and adaptations shape implementation. American Educational Research Journal, 37, 775-799. doi:10.3102/00028312037003775
de Beer, M. (2010a). Longitudinal predictive validity of a learning potential test. Journal of Psychology in Africa, 20, 225-231.
de Beer, M. (2010b). A modern assessment psychometric approach to dynamic assessment. Journal of Psychology in Africa, 20, 241-246.
Delclos, V. R., Burns, M. S., & Kulewicz, S. J. (1987). Effects of dynamic assessment on teachers’ expectations of handicapped children. American Educational Research Journal, 24, 325-336. doi:10.3102/00028312024003325
Den Brok, P. J., Van Tartwijk, J. W. F., Wubbels, T., & Veldman, I. M. J. (2010). The differential effect of the teacher-student interpersonal relationship on student outcomes for students with different ethnic backgrounds. British Journal of Educational Psychology, 80, 199-221. doi:10.1348/000709909X465632
Deutsch, R., & Reynolds, Y. (2000). The use of dynamic assessment by educational psychologists in the UK. Educational Psychology in Practice, 16, 311-331. doi:10.1080/02667360020006381
Dina, F., & Efklides, A. (2009). Metacognitive Experiences as the Link between Situational Characteristics, Motivation, and Affect in Self-Regulated Learning. In M. Wosnitza, S. Karabenick, A. Efklides, & P. Nenniger (Eds.), Contemporary Motivation Research: from global to local perspectives (pp. 117-146). Cambridge, MA: Hogrefe & Huber.
Dinger, F. C., & Dickhäuser, O. (2013). Does implicit theory of intelligence cause achievement goals? Evidence from an experimental study. International Journal of Educational Research, 61, 38-47. doi:10.1016/j.ijer.2013.03.008
Dolores Calero, M., Belen, G. M., & Auxiliadora Robles, M. (2011). Learning potential in high IQ children: The contribution of dynamic assessment to the identification of gifted children. Learning and Individual Differences, 21, 176-181. doi:10.1016/j.lindif.2010.11.025
Dweck, C. S., Chiu, C., & Hong, Y. (1995). Implicit theories and their role in judgments and reactions: A word from two perspectives. Psychological Inquiry, 6, 267-285. doi:10.1207/s15327965pli0604_1
Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation and personality. Psychological Review, 95, 256-273. doi:10.1037//0033-295X.95.2.256
Dymond, S. K., & Russell, D. L. (2004). Impact of grade and disability on the instructional context of inclusive classrooms. Education and Training in Developmental Disabilities, 39(2), 127-140.
REFERENCES
159
Ref
eren
ces
Elleman, A. M., Compton, D. L., Fuchs, D., Fuchs, L. S., & Bouton, B. (2011). Exploring dynamic assessment as a means of identifying children at risk of developing comprehension difficulties. Journal of Learning Disabilities, 44, 348-357. doi:10.1177/0022219411407865
Elliott, J. G. (2003). Dynamic assessment in educational settings: Realising potential. Educational Review, 55, 15-32. doi:10.1080/0013191022000037830
Espin, C., Wallace, T., Lembke, E., Campbell, H., & Long, J. D. (2010). Creating a progress-monitoring system in reading for middle-school students: tracking progress toward meeting high-stakes standards. Learning Disabilities Research & Practice, 25, 60-75. doi:10.1111/j.1540- 5826.2010.00304.x
Fabio, R. A. (2005). Dynamic assessment of intelligence is a better reply to adaptive behavior and cognitive plasticity. Journal of General Psychology, 132, 41-64.
Feuerstein, R., Rand, Y., & Hoffman, M. B. (1979). The dynamic assessment of retarded performers. The Learning Potential Assessment Device: Theory, instruments, and techniques. Jerusalem, Israel: ICELP Press.
Flyvbjerg, B. (2006). Five misunderstandings about case-study research. Qualitative Inquiry, 12, 219-245. doi:10.1177/1077800405284363
Francis, D., Fletcher, J., Stuebing, K., Lyon, G., Shaywitz, B., & Shaywitz, S. (2005). Psychometric approaches to the identification of LD: IQ and achievement scores are not sufficient. Journal of Learning Disabilities, 38, 98-108. doi:10.1177/00222194050380020101
Freeman, L., & Miller, A. (2001). Norm-referenced, criterion-referenced, and dynamic assessment: What exactly is the point? Educational Psychology in Practice, 17, 3-16. doi:10.1080/02667360120039942
Fuchs, D., Compton, D. L., Fuchs, L. S., Bouton, B., & Caffrey, E. (2011). The construct and predictive validity of a dynamic assessment of young children learning to read: Implications for RTI frameworks. Journal of Learning Disabilities, 44, 339-347. doi:10.1177/0022219411407864
Fuchs, L. S., Compton, D. L., Fuchs, D., Hollenbeck, K. N., Craddock, C. F., & Hamlett, C. L. (2008). Dynamic assessment of algebraic learning in predicting third graders' development of mathematical problem solving. Journal of Educational Psychology, 100, 829-850.
Fuchs, L. S., Compton, D. L., Fuchs, D., Hollenbeck, K. N., Hamlett, C. L., & Seethaler, P. M. (2011). Two- stage screening for math problem-solving difficulty using dynamic assessment of algebraic learning. Journal of Learning Disabilities, 44, 372-380. doi:10.1177/0022219411407867
Fuchs, D., & Fuchs, L. S. (2006). Introduction to response to intervention: What, why, and how valid is it? Reading Research Quarterly, 41, 93-99. doi:10.1598/RRQ.41.1.4
Fuchs, D., Fuchs, L. S., & Vaughn, S. (Eds.) (2008). Response to Intervention: A framework for reading educators. Newark, DE: International Reading Association
160
Fuchs, D., Mock, D., Morgan, P. L., & Young, C. L. (2003). Responsiveness-to-intervention: Definitions, evidence, and implications for the learning disabilities construct. Learning Disabilities Research & Practice, 18, 157-171. doi:10.1111/1540-5826.00072
Fuchs, L. S. , & Vaughn, S. (2012). Responsiveness-to-Intervention: A decade later. Journal of Learning Disabilities, 45, 195-203. doi:10.1177/0022219412442150
García-Cepero, M.C., & McCoach, D.B. (2009). Educators' implicit theories of intelligence and beliefs about the identification of gifted students. Universitas Psychologica, 8, 295-310.
Ghesquière, P. (2002). Perspectives in learning potential assessment: Conclusive remarks. In G. M. van der Aalsvoort, W. C. M. Resing, & A. J. J. M. Ruijssenaars (Eds.), Advances in cognition and educational practice Volume 7. Learning potential assessment and cognitive training (pp. 385-388). New York: Elsevier Science Ltd.
Gibbs, S., & Gardiner, M. (2008). The structure of primary and secondary teachers' attributions for pupils' misbehaviour: A preliminary cross-phase and cross-cultural investigation. Journal of Research in Special Educational Needs, 8, 68-77. doi:10.1111/j.1471-3802.2008.00104.x
Gibbs, S., & Powell, B. (2012). Teacher efficacy and pupil behaviour: The structure of teachers' individual and collective beliefs and their relationship with numbers of pupils excluded from school. British Journal of Educational Psychology, 82, 564-584. doi:10.1111/j.2044-8279.2011.02046.x
Gielen, S., Dochy, F., & Dierick, S. (2003). Evaluating the Consequential Validity of New Modes of Assessment: The Influence of Assessment on Learning, Including Pre-, Post-, and True Assessment Effects. In M. Segers, F. Dochy, & E. Cascallar (Eds.), Optimising New Modes of Assessment: In search of Qualities and Standards (pp. 37-54). Dordrecht, The Netherlands: Kluwer Academic Publishers.
Gillam, S. L., Fargo, J., Foley, B., & Olszewski, A. (2011). A nonverbal phoneme deletion task administered in a dynamic assessment format. Journal of Communication Disorders, 44, 236-245. doi:10.1016/j. jcomdis.2010.11.003
Gresham, F. M., & Witt, J. C. (1997). Utility of intelligence tests for treatment planning, classification, and placement decisions: Recent empirical findings and future directions. School Psychology Quarterly, 12, 249-267. doi:10.1037/h0088961
Grigorenko, E. L., & Sternberg, R. J. (1998). Dynamic Testing. Psychological Bulletin, 124, 75-111. doi:10.1037/0033-2909.124.1.75
Gummersall, D. M., & Strong, C. J. (1999). Assessment of complex sentence production in a narrative context. Language Speech and Hearing Services in Schools, 30, 152-164. doi:10.1044/0161- 1461.3002.152
REFERENCES
161
Ref
eren
ces
Hallahan, D. P., Kauffman, J. M., & Pullen, P. C. (Eds.) (2012). Exceptional Learners. An introduction to special education. Boston, MA: Pearson.
Hallahan, D. P., & Mock, D. R. (2003). A brief history of the field of learning disabilities. In H. Swanson, K. Harris, & S. Graham (Eds.), Handbook of Learning Disabilities (pp. 16-29). New York: The Guilford Press.
Hamers, J. H. M., Hessels, M. G. P., & Pennings, A. H. (1996). Learning potential in ethnic minority children. European Journal of Psychological Assessment, 12, 183-192.
Hattie, J. (2009). Visible Learning. A synthesis of over 800 meta-analyses relating to achievement. London: Routledge.
Haywood, H.C. (2012). Dynamic assessment: A history of fundamental ideas. Journal of Cognitive Education and Psychology, 11, 217-229. doi:10.1891%2F1945-8959.11.3.217
Haywood, H. C., Tzuriel, D., & Vaught, S. (1992). Psychoeducational assessment from a transactional perspective. In H. C. Haywood & D. Tzuriel (Eds.), Interactive as sessment. New York: Springer Verlag. Pp. 38-63.
Hessels, M.G.P. (1993). Leertest voor ethische minderheden: Theoretische en empirische verantwoording [Learning potential test for ethnic minorities: Theoretical and empirical foundation]. Rotterdam, The Netherlands: RISBO.
Hessels, M. G. P. (2000). The learning potential test for ethnic minorities (LEM): A tool for standardized assessment of children in kindergarten and the first years of primary school. In C. S. Lidz, & J. Elliott (Eds.), Dynamic Assessment: Prevailing Models and Applications (pp.109-131). New York: Elsevier.
Hessels, M. G. P., & Hessels-Schlatter, C. (2013). Current views on cognitive education: A critical discussion and future perspectives. Journal of Cognitive Education and Psychology, 12, 108-124. doi:10.1891/1945-8959.12.1.108
Hessels, M. G. P., & Tiekstra, M. (2010). Evaluation des capacités de raisonnement et prédiction des apprentissages dans un domaine scolaire nouveau chez des élèves présentant une déficience intellectuelle légère [Evaluation of reasoning capacity and prediction of learning in a novel school domain of students with mild mental retardation]. In M. G. P. Hessels, & C. Hessels- Schlatter, (Eds.), Evaluation et intervention auprès d'élèves en difficultés [Evaluation and intervention with students with difficulties] (pp. 51-66). Bern, Switzerland: Peter Lang.
Hessels-Schlatter, C., & Hessels, M. G. P. (2009). Clarifying some issues in dynamic assessment: Comments on Karpov and Tzuriel. Journal of Cognitive Education and Psychology, 8, 246-251. doi:10.1891/1945- 8959.8.3.246
Ikeda, M. J. (2012). Policy and practice considerations for Response to Intervention: Reflections and commentary. Journal of Learning Disabilities, 45, 274-277. doi:10.1177/0022219412442170
162
Imants, J. G. M., & De Brabander, C. J. (1996). Teachers’ and principals’ sense of efficacy in elementary schools. Teaching and Teacher Education, 12, 179-195. doi:10.1016/0742-051X(95)00053-M
Imants, J., Van der Aalsvoort, G. M., De Brabander, C. J., & Ruijssenaars, A. J. J. M. (2001). The role of the special services coordinator in Dutch primary schools: A counterproductive effect of inclusion policy. Educational Management Administration & Leadership, 29, 35-48. doi:10.1177/0263211x010291003
Jeltova, I., Birney, D., Fredine, N., Jarvin, L., Sternberg, R. J., & Grigorenko, E. L. (2011). Making instruction and assessment responsive to diverse students' progress: Group-administered dynamic assessment in teaching mathematics. Journal of Learning Disabilities, 44, 381-395. doi:10.1177/0022219411407868
Jennings, P. A., & Greenberg, M. T. (2009). The prosocial classroom: Teacher social and emotional competence in relation to student and classroom outcomes. Review of Educational Research, 79, 491-525. doi:10.3102/0034654308325693
Jitendra, A. K., & Kameenui, E. J. (1996). Experts' and novices error patterns in solving part-whole mathematical word problems. Journal of Educational Research, 90, 42-51.
Jones, R. E., & Ball, C. R. (2012). Introduction to the special issue: Addressing response to intervention implementation: Questions from the field. Psychology in the Schools, 49, 207-209. doi:10.1002/ pits.21590
Jordan, A., & Stanovich, P. (2001). Patterns of teacher–student interaction in inclusive elementary classrooms and correlates with student self-concept. International Journal of Disability, Development & Education, 48, 33-52. doi:10.1080/10349120120036297
Kanevsky, L., & Geake, J. (2004). Inside the zone of proximal development: Validating a multifactor model of learning potential with gifted students and their peers. Journal for the Education of the Gifted, 28, 182-217.
Katz, N., Golstand, S., Bar-Ilan, R. T., & Parush, S. (2007). The dynamic occupational therapy cognitive assessment for children (DOTCA-ch): A new instrument for assessing learning potential. American Journal of Occupational Therapy, 61, 41-52.
Kavale, K. A. (2005). Identifying specific learning disability: Is responsiveness to intervention the answer? Journal of Learning Disabilities, 38, 553-562. doi:10.1177/00222194050380061201
Kavale, K. A., & Spaulding, L. S. (2008). Is response to intervention good policy for specific learning disability? Learning Disabilities Research & Practice, 23, 169-179. doi:10.1111/j.1540- 5826.2008.00274.x
Kavale, K. A., Spaulding, L. S., & Beam, A. P. (2009). A time to define: Making the specific learning disability definition prescribe specific learning disability. Learning Disability Quarterly, 32, 39-48.
REFERENCES
163
Ref
eren
ces
Kerlinger, F. N., & Lee, H. B. (2000). Foundations of Behavioral Research (4th edition). New York: Harcourt Brace College Publishers.
Kirkwood, M. W., Weiler, M. D., Bernstein, J. H., Forbes, P. W., & Waber, D. P. (2001). Sources of poor performance on the rey-osterrieth complex figure test among children with learning difficulties: A dynamic assessment approach. Clinical Neuropsychologist, 15, 345-356.
Konstantopoulos, S., & Chung, V. (2011). The persistence of teacher effects in elementary grades. American Educational Research Journal, 48, 361-386. doi:10.3102/0002831210382888
Larsen, J. A., & Nippold, M. A. (2007). Morphological analysis in school-age children: Dynamic assessment of a word learning strategy. Language, Speech, and Hearing Services in Schools, 38, 201-212.
Lauchlan, F., & Elliott, J. (2001). The psychological assessment of learning potential. British Journal of Educational Psychology, 71, 647-665.
Lauermann, F., & Karabenick, S. A. (2011). Taking teacher responsibility into account(ability): Explicating its multiple components and theoretical status. Educational Psychologist, 46, 122-140. doi:10.1080/ 00461520.2011.558818
Lebeer, J., Birta-Székeley, N., Demeter, K, Bohács, K., Araujo Candeias, A., Sønnesyn, G., . . . Dawson, V. (2011) Re-assessing the current assessment practice of children with special education needs in Europe. School Psychology International, 33, 69-92. doi:10.1177/0143034311409975
Leroy, N., Bressoux, P., Sarrazin, P., & Trouilloud, D. (2007). Impact of teachers' implicit theories and perceived pressures on the establishment of an autonomy supportive climate. European Journal of Psychology of Education, 22, 529-545. doi:10.1007/BF03173470
Levin, J. R., & O’Donnell, A. M. (1999). What to do about educational research’s credibility gaps? Issues in Education, 5, 177-229. doi:10.1016/S1080-9724(00)00025-2
Lidz, C. S. (1995). Dynamic assessment and the legacy of L. S. Vygotsky. School Psychology International, 16, 143-153.
Lidz, C. S., & Macrine, S. L. (2001). An alternative approach to the identification of gifted culturally and linguistically diverse learners: The contribution of dynamic assessment. School Psychology International, 22, 74-96.
Lincoln, Y. S., & Guba, E. G. (1985). Establishing Trustworthiness. In Y. Lincoln & E. Guba (Eds.), Naturalistic Inquiry (pp. 289-331). Beverly Hills, CA: Sage.
Meijer, J. (2001). Learning potential and anxious tendency: Test anxiety as a bias factor in educational testing. Anxiety, Stress and Coping, 14, 337-362.
164
Meijer, J., & Elshout, J. J. (2001). The predictive and discriminant validity of the zone of proximal development. British Journal of Educational Psychology, 71, 93-113.
Meijer, W., & Tijhaar, M. (1995). Begeleidingsvaardigheden bij consultatieve leerlingbegeleiding [Support skills in teachers’ consultation on student behavior]. School en Begeleiding [School and Counseling], 13(1), 26-29.
Meirink, J. A., Meijer, P. C., Verloop, N., & Bergen, T. C. M. (2009). Understanding teacher learning in secondary education: The relations of teacher activities to changed beliefs about teaching and learning. Teaching and Teacher Education, 15, 89-100. doi:10.1016/j.tate.2008.07.003
Messick, S. (1989). Meaning and values in test validation: The science and ethics of assessment. Educational Researcher, 18(2), 5-11.
Messick, S. (1995). Standards of validity and the validity of standards in performance assessment. Educational Measurement: Issues and Practice, 14(4), 5-8.
Minnaert, A. (2013). Goals are motivational researchers’ best friend, but to what extent are achievement goals and achievement goal orientations also the best friend of educational outcomes? International Journal of Educational Research, 61, 85-89. doi:10.1016/j.ijer.2013.08.002
Minnaert, A., & Vermunt, J. D. (2006). 25 jaar Onderwijspsychologie in Nederland en Vlaanderen in de periode 1980 tot 2005: Trends, pendels en grensverleggers [25 years of Educational psychology in The Netherlands and Flanders during 1980 till 2005]. Pedagogische Studiën, 83, 260-277.
Moorman, E. A., & Pomerantz, E. M. (2010). Ability mindsets influence the quality of mothers’ involvement in children’s learning: An experimental investigation. Developmental Psychology, 46, 1354-1362. doi:10.1037/a0020376
Nirmalakhandan, N. (2009). Use of computerized dynamic assessment to improve student achievement: Case study. Journal of Professional Issues in Engineering Education and Practice, 135, 75-80.
Opdenakker, M.-C. (2008). Vragenlijst over de leeromgeving – versie docenten [Questionnaire about learning environment – teacher version]. Groningen, The Netherlands: University of Groningen.
Opdenakker, M., & Van Damme, J. (2006). Differences between secondary schools: A study about school context, group composition, school practice, and school effects with special attention to public and catholic schools and types of schools. School Effectiveness and School Improvement, 17, 87-117. doi:10.1080/09243450500264457
Pajares, M. F. (1992). Teachers' beliefs and educational research: Cleaning up a messy construct. Review of Educational Research, 62, 307-32. doi:10.3102/00346543062003307
REFERENCES
165
Ref
eren
ces
Pakarinen, E., Kiuru, N., Lerkkanen, M., Poikkeus, A., Siekkinen, M., & Nurmi, J. (2010). Classroom organization and teacher stress predict learning motivation in kindergarten children. European Journal of Psychology of Education, 25, 281-300. doi:10.1007/s10212-010-0025-6
Pelletier, L. G., Seguin-Levesque, C., & Legault, L. (2002). Pressure from above and pressure from below as determinants of teachers' motivation and teaching behaviors. Journal of Educational Psychology, 94, 186-196. doi:10.1037//0022-0663.94.1.186
Peltenburg, M., van den Heuvel-Panhuizen, M., & Doig, B. (2009). Mathematical power of special-needs pupils: An ICT-based dynamic assessment format to reveal weak pupils' learning potential. British Journal of Educational Technology, 40, 273-284.
Peltenburg, M., van den Heuvel-Panhuizen, M., & Robitzsch, A. (2010). ICT-based dynamic assessment to reveal special education students' potential in mathematics. Research Papers in Education, 25, 319- 334. doi:10.1080/02671522.2010.498148
Perry, N., Turner, J.C., Meyer, D.K. (2006). Classrooms as contexts for motivated learning. In P. A. Alexander & P. H. Winne (Eds.), Handbook of Educational Psychology (2nd ed., pp. 327-348). Mahwah, NJ: Lawrence Erlbaum.
Prince, A., Minnaert, A., & Opdenakker, M.-C. (2014). Motivation in educational interventions: A systematic review exploring the concept in relation to intervention outcomes. In A. Prince, Motivation: How to tame the elephant in the classroom? (Unpublished doctoral dissertation). Groningen, The Netherlands: University of Groningen.
Prince, A., Tiekstra, M., & Minnaert, A. (2014). When theory meets practice becomes political: Can we combine scientific credibility and societal credibility? In A. Prince, Motivation: How to tame the elephant in the classroom? (Unpublished doctoral dissertation). Groningen, The Netherlands: University of Groningen.
Resing, W. C. M. (1990). Intelligentie en leerpotentieel. Een onderzoek naar het leerpotentieel van jonge leerlingen uit het basis- en speciaal onderwijs [Intelligence and learning potential. A study on the learning potential of young children from primary schools and special education]. Lisse, The Netherlands: Swets & Zeitlinger.
Resing, W. C. M. (2006). Using children’s ‘ability to learn’ in diagnosis and assessment. Educational & Child Psychology, 23(3), 6-11.
Resing, W. C. M., Tunteler, E., de Jong, F. M., & Bosma, T. (2009). Dynamic testing in indigenous and ethnic minority children. Learning and Individual Differences, 19, 445-450. doi:10.1016/j. lindif.2009.03.006
Reynolds, C. R., & Shaywitz, S. E. (2009). Response to intervention: Ready or not? Or, from wait-to-fail to watch-them-fail. School Psychology Quarterly, 24, 130-145. doi:10.1037/a0016158
166
Rosenfeld, M., & Rosenfeld, S. (2008). Developing effective teacher beliefs about learners: the role of sensitizing teachers to individual learning differences. Educational Psychology, 28, 245-272. doi:10.1080/01443410701528436
Rouse, M., & Agbenu, R. (1998). Assessment and special educational needs: teachers’ dilemmas. British Journal of Special Education, 25, 81-87. doi:10.1111/1467-8527.t01-1-00062
Rozendaal, J. S., Minnaert, A., & Boekaerts, M. (2005). The influence of teacher perceived administration of self-regulated learning on students’ motivation and information-processing. Learning and Instruction, 15, 141-160. doi:10.1016/j.learninstruc.2005.04.011
Ruijssenaars, A. J. J. M. (2001). Leerproblemen en leerstoornissen. Remedial teaching en behandeling. Hulpschema’s voor opleiding en praktijk [Learning difficulties and learning disabilities. Remedial teaching and treatment. Frameworks for education and practice]. Rotterdam, The Netherlands: Lemniscaat.
Runhaar, P, Sanders, K., & Yang, H. (2010). Stimulating teachers’ reflection and feedback asking: An interplay of self-efficacy, learning goal orientation, and transformational leadership. Teaching and Teacher Education, 26, 1154-1161. doi:10.1016/j.tate.2010.02.011
Ryba, K. (1998). Dynamic assessment and programme planning for students with intellectual disabilities. New Zealand Journal of Psychology, 27(2), 3-10.
Scanlon, D. (2013). Specific learning disability and its newest definition: Which is comprehensive? And which is insufficient? Journal of Learning Disabilities, 46, 26-33. doi:10.1177/0022219412464342
Schlatter, C., & Büchel, F. P. (2000). Detecting reasoning abilities of persons with moderate mental retardation: the analogical reasoning learning test (ARLT). In C. S. Lidz, & J. Elliott (Eds.), Dynamic Assessment: Prevailing Models and Applications (pp.155-186). New York: Elsevier.
Shamir, A., & Tzuriel, D. (2004). Children’s Mediational Teaching Style as a Function of Intervention for Cross-Age Peer-Mediation. School Psychology International, 25, 59-78.
Shapiro, E. S. (1987). Behavioral assessment in school psychology. Hillsdale, NJ: Lawrence Erlbaum.
Shapiro, E. S., & Kratochwill, T. R. (2000). Behavioral assessment in schools. Theory, research, and clinical foundations. New York: Guilford Press.
Skinner, E. A., & Belmont, M. J. (1993). Motivation in the classroom: Reciprocal effects of teacher behavior and student engagement. Journal of Educational Psychology, 85, 571-581. doi:10.1037/0022- 0663.85.4.571
Steenbeek, H., Jansen, L., & Van Geert, P. (2012). Scaffolding dynamics and the emergence of problematic learning trajectories. Learning and Individual Differences, 22, 64-75. doi:10.1016/j.lindif.2011.11.014
REFERENCES
167
Ref
eren
ces
Sternberg, R. J. (2004). Culture and intelligence. American Psychologist, 59, 325-338. doi:10.1037/0003- 066X.59.5.325
Sternberg, R. J., Birney, D., Jarvin, L., Kirlik, A., Stemler, S., & Grigorenko, E. L. (2006). From molehill to mountain: The process of scaling up educational interventions In M. A. Constas, & R. J. Sternberg (Eds.), Translating Educational Theory and Research into Practice: Developments in content domains, large-scale reform, and intellectual capacity. (pp. 205-222). Mahwah, NJ: Lawrence Erlbaum Associates.
Sternberg, R. J., Grigorenko, E. L., Ngorosho, D., Tantufuye, E., Mbise, A., Nokes, C., . . . Bundy, D. A. (2002). Assessing intellectual potential in rural Tanzanian school children. Intelligence, 30, 141- 162.
Stuebing, K. K., Barth, A. E., Molfese, P. J., Weiss, B., & Fletcher, J. M. (2009). IQ is not strongly related to response to reading instruction: A meta-analytic interpretation. Exceptional Children, 76, 31-53.
Sutherland, K. S., Lewis-Palmer, T., Stichter, J., & Morgan, P. L. (2008). Examining the influence of teacher behavior and classroom context on the behavioral and academic outcomes for students with emotional or behavioral disorders. The Journal of Special Education, 41, 223-233. doi:10.1177/0022466907310372
Swanborn, P. (2010). Case study research: What, why and how? London: SAGE
Swanson, H. L. (1995a). Effects of dynamic testing on the classification of learning-disabilities - the predictive and discriminant validity of the Swanson cognitive processing test (S-cpt). Journal of Psychoeducational Assessment, 13, 204-229.
Swanson, H. L. (1995b). Using the cognitive processing test to assess ability: Development of a dynamic assessment measure. School Psychology Review, 24, 672-693.
Swanson, H. L. (2011). Dynamic testing, working memory, and reading comprehension growth in children with reading disabilities. Journal of Learning Disabilities, 44, 358-371. doi:10.1177/0022219411407866
Swanson, H. L., & Howard, C. B. (2005). Children with reading disabilities: Does dynamic assessment help in the classification? Learning Disability Quarterly, 28, 17-34. doi:10.2307/4126971
Tannock, R. (2013). Rethinking ADHD and LD in DSM-5: Proposed changes in diagnostic criteria. Journal of Learning Disabilities, 46, 5-25. doi:10.1177/0022219412464341
Teo, P., & Roodenburg, J. (2013). Extensions of cognitive ability assessment with dynamic testing using self-regulated learning. Journal of Cognitive Education and Psychology, 12, 230-250. doi:10.1891 %2F1945-8959.12.2.230
Thorndike, E. L. (1924). An introduction of the theory of mental and social measurements. New York: Wiley.
168
Thurlow, M. L., & Ysseldyke, J. E. (1982). Instructional planning: Information collected by school psychologists versus information considered useful by teachers. Journal of School Psychology, 20, 3-10.
Tiekstra, M., Hessels, M. G. P., & Minnaert, A. E. M. G. (2009). Learning capacity in adolescents with mild intellectual disabilities. Psychological Reports, 105, 804-814. doi:10.2466/PR0.105.3.804-814
Tiekstra, M., & Minnaert A. (2015). At-risk students and the role of implicit theories of intelligence in educational professionals’ actions. Manuscript submitted for publication.
Tiekstra, M., Minnaert, A., & Hessels, M. G. P. (2016). A review scrutinising the consequential validity of dynamic assessment. Educational Psychology, 36, 112-137. doi:10.1080/01443410.2014.915930
Timperley, H. S., & Parr, J. M. (2009). What is this lesson about? Instructional processes and student understandings in writing classrooms. Curriculum Journal, 20, 43-60. doi:10.1080/09585170902763999
Tschannen-Moran, M., & Woolfolk Hoy, A. (2001). Teacher efficacy: Capturing an elusive construct. Teaching and Teacher Education, 17, 783-805. doi:10.1016/S0742-051X(01)00036-1
Tzuriel, D. (2000). The seria-think instrument: Development of a dynamic test for young children. School Psychology International, 21, 177-194.
Tzuriel, D., & Kaufman, R. (1999). Mediated learning and cognitive modifiability: Dynamic assessment of young Ethiopian immigrant children to Israel. Journal of Cross-Cultural Psychology, 30, 359-380. UNESCO (1994). The Salamanca statement and framework for action on special needs education. Paris: UNESCO.
UNESCO (2005). Guidelines for Inclusion: ensuring access to education for all. Paris: UNESCO.
Van de Ven, A. H., Polley, D., Garud, R., & Venkaraman, S. (1999). The innovation journey. New York: Oxford University Press.
Van der Aalsvoort, G. M., Resing, W. C. M., & Ruijssenaars, A. J. J. M. (Eds.). (2002). Advances in cognition and educational practice Volume 7. Learning potential assessment and cognitive training. New York: Elsevier Science Ltd.
Vaughn, S., & Denton, C. A. (2008). Tier 2: The role of intervention. In D. Fuchs, L. Fuchs, & S. Vaughn (Eds.), Response to Intervention: A framework for reading educators (pp. 51-70). Newark, DE: International Reading Association.
Vermunt, J. D. (1996). Metacognitive, cognitive and affective aspects of learning styles and strategies: A phenomenographic analysis. Higher Education, 31, 25-50.
REFERENCES
169
Ref
eren
ces
Viljoen, J. L., Odgers, C., Grisso, T., & Tillbrook, C. (2007). Teaching adolescents and adults about adjudicative proceedings: A comparison of pre- and post-teaching scores on the MacCAT-CA. Law and Human Behavior, 31, 419-432.
Vygotsky, L. S. (1978). Mind in society: The development of Higher Psychological Processes. In M. Cole, V. John- Steiner, S. Scribner, & E. Souberman (Eds.). Cambridge, MA: Harvard University Press.
Wiedl, K. H. (1980). Paradigmatische Orientierungen von Lehrern zum Problem psychischer Normalität and Störung: theoretische Explikation und empirische Analyse [Paradigm orientations of teachers about psychological normality and disorder: theoretical explication and empirical analysis]. Sonderpädagogik und Therapie, 286-295.
Wigfield, A., Byrnes, J. P., & Eccles, J. S. (2006). Development during early and middle adolescence. In P. A. Alexander & P. H. Winne (Eds.), Handbook of Educational Psychology (2nd ed., pp. 87-113). Mahwah, NJ: Lawrence Erlbaum.
Wilson, B. (1998). Matching Student Needs to Instruction. In S. A. Vogel & S. Reder (Eds.), Learning Disabilities, Literacy and Adult Education (pp. 191-212). Baltimore, MD: Brookes.
Woide, N., Beckmann, J. F., Elliott, J. G., & Guthke, J. (2005, July). Effects of feedback: individual differences in the ability to benefit from feedback. Paper presented at the 10th Conference of the International Association on Cognitive Education and Psychology, Durham, England.
Wolters, C. A., & Daugherty, S. G. (2007). Goal structures and teachers' sense of efficacy: Their relation and association to teaching experience and academic level. Journal of Educational Psychology, 99, 181-193. doi:10.1037/0022-0663.99.1.181
Yeager, D. S. , & Walton, G. M. (2011). Social-psychological interventions in education: They’re not magic. Review of Educational Research, 81, 267-301. doi:10.3102/0034654311405999
Ysseldyke, J. (2001). Reflections on a research career: Generalizations from 25 years of research on assessment and instructional decision making. Exceptional Children, 67, 295-309. doi:10.1177/001440290106700301
SamenvattingSummary in Dutch
172
NEDERLANDSE SAMENVATTING
173
Het leerpotentieel benutten van risicoleerlingen in de klasEen onderzoek naar de consequentiële validiteit van dynamische tests
In de reguliere onderwijspraktijk is het afnemen van tests gebruikelijk geworden wanneer
een leerling aanhoudende (leer-)problemen in de klas vertoont. Bij problemen in het
cognitieve gebied (dat wil zeggen, de leerling loopt wat betreft schoolresultaten voor of
achter op klasgenoten) wordt veelal uitgeweken naar het afnemen van een intelligentietest.
Deze testafnames zijn bedoeld om juiste diagnoses1 te stellen, met als uiteindelijk doel dat
er een passend onderwijsaanbod aan de leerling geboden kan worden. Echter, de vraag
blijft in hoeverre instructiegedrag van de leerkracht nou daadwerkelijk verandert na
resultaten van deze testafnames. Het blijkt voor leerkrachten moeilijk om de uitkomsten van
intelligentietests te vertalen in concrete veranderingen in hun onderwijs (Freeman & Miller,
2001; Ysseldyke, 2001). Kortom, er bestaat een kloof tussen het stellen van een diagnose
(assessment) en het aanpassen van leerkrachtgedrag op deze diagnose (Haywood, 2012;
Shapiro & Kratochwill, 2000).
De onderzoeken in deze dissertatie richten zich op deze kloof. Om uiteindelijk gerichte
interventies toe te kunnen passen die recht doen aan het leerpotentieel van de leerling, is
er onderzoek nodig dat inzicht geeft in de wijze waarop de uitkomsten van tests vertaald
worden in concrete acties. Met andere woorden, de consequenties van testafnames zouden
beter bestudeerd moeten worden (zie Messick, 1989, 1995). De problematiek wordt vanuit
een methodologisch perspectief benaderd: als we de kloof willen overbruggen, dan zal de
consequentiële validiteit van (dynamische) testafnames verhoogd moeten worden. Op deze
manier zal de instructie aan leerlingen bij wie het leren niet vanzelfsprekend is gebeuren op
een wijze die overdacht en gefundeerd is. Tests met consequentiële validiteit hebben immers
als doel het onderwijskundige en orthopedagogische handelen van docenten te verbeteren
om op deze wijze het leren bij leerlingen weer vlot te trekken of, in geval van hardnekkige
problematiek, verdere achterstand te voorkomen.
1 Met het stellen van een diagnose wordt bedoeld dat de problematiek van een leerling onderzocht wordt, en niet het etiketteren van (leer-)problemen van de leerling.
174
Het verhogen van de consequentiële validiteit van tests, en daarmee het overbruggen
van de kloof tussen assessment en instructie in de klas, vraagt om een tweezijdige
benadering: niet alleen de tests zijn hierin belangrijk, maar ook de context waarin de
tests afgenomen en geïnterpreteerd worden. In deze context spelen de zogenoemde
“significante anderen” een grote rol in het ontwikkelen van het potentieel van leerlingen
en zijn zij degene die testuitslagen interpreteren.
Op dit moment vindt de afname van tests meestal plaats op een geprotocolleerde
manier om aan criteria zoals betrouwbaarheid en validiteit te voldoen. Dit betekent bij
IQ tests dat de testafnemer wel af en toe een vraag mag stellen, maar geen additionele
uitleg mag geven tijdens de testafname. In het geval van risicoleerlingen kunnen hierbij
vraagtekens gezet worden, en een dynamische manier van testen is in dit soort gevallen
een beter alternatief. In een studie voorafgaand aan deze dissertatie is aangetoond dat
een dynamische test onderscheid kan maken tussen leerlingen die op een standaard
IQ test allen laag scoren (Tiekstra, Hessels, & Minnaert, 2009). Tijdens de dynamische
testafname vindt er een fase plaats waarin feedback en/of hints gegeven worden. Deze
“leerfase” lijkt een belangrijke rol te kunnen spelen bij het overbruggen van de kloof
tussen assessment en instructie, omdat deze inzicht verschaft in de processen die de
leerling aanwendt om taakjes op te lossen. Ondanks deze voordelen worden dynamische
tests nauwelijks in de dagelijkse praktijk gebruikt. Er dient onderzocht te worden hoe
we de kloof tussen assessment en instructie kunnen overbruggen en of dynamische tests
daarbij kunnen helpen.
Resultaten Omdat dynamische testprocedures gezien kunnen worden als een waardevol alternatief
voor de standaard IQ tests, richtte het eerste onderzoek (zie hoofdstuk 2) zich op deze
vorm van tests. Een literatuurstudie van 31 wetenschappelijke artikelen waarin in totaal
29 verschillende dynamische testprocedures werden beschreven, leverde een aantal
interessante bevindingen op. Na een grondige analyse kon vastgesteld worden dat
dynamische tests wellicht voldoen aan het criterium van validiteit voor consequenties
NEDERLANDSE SAMENVATTING
175
direct volgend op de testafname (immers, de aanwijzingen van de testafnemer
worden aangepast op de resultaten van de zojuist gemaakte test van de leerling), maar
richtlijnen voor instructie of interventie na de testafname worden nauwelijks gegeven;
de consequentiële validiteit die verder van de test af zit, wordt amper benut bij de
huidige tests. Daarom was het belangrijk om een grondige analyse te maken van de
leerfases die bij ieder instrument beschreven waren. Het bleek dat de testafnemer hints
en feedback geeft die voornamelijk op het cognitieve gebied, en in mindere mate ook op
metacognitief gebied, gericht zijn. Echter, motivationele aspecten worden nooit expliciet
genoemd in het afnemen van deze tests. Dat is opmerkelijk, omdat dynamische tests
zich ook richten op de “warme” kant van het leren. Kortom, ondanks het potentieel van
dynamische tests om de kloof tussen assessment en instructie te overbruggen, voegen ze
op dit gebied in de praktijk nog niets extra’s toe.
Tijdens een afname van dynamische tests wordt de mate waarin een kind didactisch
resistent is gemeten. Immers, er wordt nagegaan in hoeverre een kind reageert op de
kwaliteits- één-op-één instructie die geboden wordt tijdens de testafname. De vraag blijft
hoe men dit zou kunnen vertalen naar de praktijk in de reguliere klas. Het Response to
Instruction/Intervention model (beide termen worden gebruikt) lijkt een manier te zijn
om deze vertaling mogelijk te maken. Het kwalitatieve multipele methode onderzoek dat
in hoofdstuk drie beschreven wordt, richtte zich op de implementatie van het RTI-model
in de Nederlandse context. In het bijzonder werd er gekeken naar de mogelijkheid om
met dit model didactische resistentie te meten van leerlingen in de reguliere klas. De
resultaten wezen uit dat het een instrument kan zijn om didactische resistentie te meten,
mits het model zodanig geïmplementeerd wordt dat naast de resultaten van de leerling
ook de resultaten (geboden instructie) van de leerkracht in kaart gebracht worden. Dit
vraagt om een aanpassing in hoe men denkt over het leren (namelijk: de leerkracht
speelt een net zo grote rol in de mate van didactische resistentie) en niet alleen een
verandering in gedrag (d.w.z. het model uitvoeren).
176
De scheiding tussen gedrag en opvattingen van professionals die zo duidelijk uit
dit onderzoek naar voren kwam, leek ons belangrijk voor het verbeteren van de
consequentiële validiteit. In hoofdstuk 4 wordt er verder ingegaan op het verschil tussen
gedrag en opvattingen van onderwijsprofessionals. In een kwantitatieve studie werd de
invloed van impliciete intelligentietheorieën op het gedrag van leerkrachten in de klas
en onderwijsprofessionals buiten de klas bij risicoleerlingen onderzocht. Het bleek dat
deze impliciete intelligentietheorieën een grote rol spelen in de dagelijkse praktijk van het
onderwijs. Ongeveer 34% van de totale variantie in handelingen van de leerkrachten kon
voorspeld worden door impliciete theorieën over intelligentie, bij leerkrachten van christelijke
scholen was dit zelfs 61%. Dit betekent dat het onderwijzen in de Zone van de Naaste
Ontwikkeling beïnvloed wordt door ideeën van leerkrachten over de veranderbaarheid van
intelligentie. In de groep van onderwijsprofessionals buiten de klas (zoals intern begeleiders
en schoolpsychologen of orthopedagogen) bleek dat er een model uit de data afgeleid kon
worden: impliciete intelligentietheorieën voorspellen de waarde die wordt gehecht aan IQ-
tests, en dit voorspelt vervolgens de handelingen die worden gedaan naar aanleiding van
de testuitkomsten. In het kort gezegd, de consequentiële validiteit van tests wordt voorspeld
door de impliciete intelligentietheorie van de professional.
Omdat impliciete intelligentietheorieën, en hiermee het op waarde schatten van testresultaten,
een dusdanig grote rol bleken te spelen in de mate van consequentiële validiteit, wilden
we uitzoeken welke rol tests en testgegevens spelen in de dagelijkse praktijk. In hoofdstuk vijf
wordt gerapporteerd over twee kwalitatieve studies die gericht zijn op een beschrijving van
de praktijk. In de eerste in-vivo studie zijn de antwoorden van 36 schoolpsychologen, 21
intern begeleiders en 44 leerkrachten op open vragen geanalyseerd. De resultaten wezen
uit dat er geen eenduidig beeld bestond over het gebruik van testuitkomsten en de rol van
alle onderwijsprofessionals in de dagelijkse praktijk. Daarom is er in de tweede in-vivo studie
een casestudy uitgevoerd. Ook in deze studie werd de kloof tussen testafnames en instructie
bevestigd. Samenvattend laten deze studies zien dat er in de huidige onderwijspraktijk
behoefte is aan een andere soort van tests die (passende) richtlijnen voor het handelen in
de praktijk kunnen geven.
NEDERLANDSE SAMENVATTING
177
Alle bovengenoemde studies toonden aan dat onderwijsprofessionals meer kennis en
vaardigheden nodig hebben om het leerpotentieel van leerlingen te kunnen benutten in
de reguliere klas. De invloed van impliciete intelligentietheorieën moet hierbij zeker niet
vergeten worden. Als laatste studie hebben we een kortstondige interventie ontwikkeld
om te kijken of we deze vaardigheden en kennis kunnen verhogen en de impliciete
theorieën zouden kunnen beïnvloeden. De resultaten zijn beschreven in hoofdstuk zes.
Een pilotstudie onder een schoolteam (N=21) heeft meer inzicht kunnen verschaffen in
welke processen nodig zijn om de huidige onderwijspraktijk te veranderen en de kloof
tussen assessment en instructie te kunnen overbruggen. Het bleek dat de kortdurende
interventie een effect had op de mate waarin mensen geloven in de veranderbaarheid
van intelligentie: binnen het gehele schoolteam namen de “statische” opvattingen over
intelligentie af. Bovendien lieten groeicurvemodellen zien dat deze afname beïnvloed
wordt door de mate van het geloof in eigen handelen van leerkrachten (“self-efficacy”).
Hoe zekerder men is over het eigen handelen, des te sterker de afname in statische
opvattingen, zelfs op de lange termijn. Echter, hoewel de training zorgde voor een
verandering in het denken over intelligentie, kon de verandering in handelen niet
gemeten worden door restricties in de meetinstrumenten (plafondeffect). Dit aspect zou
dan ook in een vervolgstudie beter gemeten moeten worden. Concluderend kan men
stellen dat als we gedrag van leerkrachten willen veranderen, dat opvattingen en de
zekerheid over het eigen handelen meegenomen moeten worden in de interventies.
Implicaties en aanbevelingen Tests worden afgenomen met als doel het handelen van leerkrachten te veranderen.
Daarom kan ‘consequentiële validiteit’, waarin de validiteit van consequenties centraal
staat, gezien worden als ultieme vorm van validiteitseisen van tests. Helaas wordt deze
vorm van validiteit op dit moment niet meegenomen in oordelen over betrouwbaarheid
en validiteit van tests. De studies in deze dissertatie laten zien dat juist deze vorm van
validiteit de brug kan slaan tussen diagnose en interventie. De studies hebben laten zien
dat dynamische tests een groot potentieel hebben om meer consequentiële validiteit te
waarborgen dan de huidige statische tests.
178
Daarnaast onderschrijven de studies de behoefte aan een dynamisch perspectief op
leren van alle betrokkenen: van leraren tot schooldirecteuren, van orthopedagogen tot
beleidsmakers. In Nederland is recent het Passend Onderwijs beleid ingevoerd. Met dit
beleid is de zorgplicht voor zorgleerlingen verschoven naar de scholen. Dit betekent dat
onderwijsbetrokkenen op een juiste manier zouden moeten kunnen handelen in het geval
van risicoleerlingen. Als we willen dat zit zal slagen, zijn er veranderingen nodig.
Ten eerste moet er meer nadruk komen te liggen op de overtuigingen en het geloof in
eigen handelen (self-efficacy) van alle onderwijsprofessionals die met risicoleerlingen werken.
Meerdere studies in deze dissertatie hebben aangetoond dat opvattingen voor een groot
deel bepalen hoe professionals handelen, en grote gevolgen kunnen hebben voor het
optimaal benutten van het leerpotentieel van leerlingen. In initiële opleidingen van deze
professionals zou een dynamische opvatting over intelligentie en leren aangeleerd dan wel
ontplooid moeten worden, zodat er recht gedaan kan worden aan het (leer)potentieel van
alle leerlingen.
Ten tweede zou er in plaats van de huidige overwaardering voor (statische) intelligentietests
in de onderwijspraktijk meer aandacht besteed moeten worden aan alternatieven,
zoals dynamische tests. Ondanks het grote potentieel van dynamische tests voor de
onderwijspraktijk, blijven kennis en kunde van onderwijsprofessionals achter. Indien
onderwijsprofessionals de bovengenoemde dynamische opvatting over intelligentie
ontwikkelen, zal de implementatie van dynamische tests in de praktijk soepeler verlopen.
Ten derde zou consequentiële validiteit centraal moeten komen te staan in testvalidering.
Naast de aanbevelingen voor onderwijsprofessionals, hebben ook testontwikkelaars
de verantwoordelijkheid om stil te staan bij mogelijke consequenties die volgen uit de
testresultaten van hun test. Zij zouden zich moeten buigen over vraagstukken met betrekking
tot de interpretatie van hun testuitkomsten door onderwijsprofessionals en er zorg voor
moeten dragen dat dit op een valide manier gebeurt.
Als deze aanbevelingen in acht genomen worden, zal de huidige kloof tussen diagnose en
NEDERLANDSE SAMENVATTING
179
interventies overbrugd kunnen worden. Uiteraard vraagt dit tijd en aanpassing van alle
betrokkenen, maar zijn het kleine stappen die - op termijn - een groot verschil kunnen
maken in het professionele handelen van docenten en daarmee in de leerprocessen en
-resultaten van leerlingen.
About the author
188
About the author
Marlous Tiekstra (1985) was born in Delft. She attended primary school in Marum
(Gr.) and secondary school in Leek (Gr.), The Netherlands. In 2003 she started her
Bachelor Degree in Pedagogical Sciences at the University of Groningen and finished
in 2006. In 2004 she was board member of the students’ association of the Pedagogical
Sciences department and was chairwoman of the committee that organised a national
day for Pedagogical Sciences students in The Netherlands. She finished her Masters
studies in 2007, while spending nearly the full academic year at the University of
Geneva, Switzerland. In this year she acquired experience and conducted research in
the field of dynamic assessment and worked in l’Atelier d’Apprentissage, where she was
trained in metacognitive interventions and guided students with learning disabilities in
their learning processes. Beginning 2008, Marlous started her doctoral studies at both
universities mentioned before. The field of learning and learning processes is still of
great interest to her. Next to the researcher’s part of her PhD studies, she developed
and conducted different trainings and interventions for teachers and coaches in
primary education, and collaborated with several Dutch educational advisory agencies.
During her PhD Marlous was member and chairwoman of several PhD committees.
Furthermore, she was committee member of several international conferences, and
a scientific reviewer for the Journal of Cognitive Education and Psychology. Finally, in 2015
she completed her PhD studies. Marlous is currently working at the Dutch Ministry of
Education in The Hague.
ABOUT THE AUTHOR
189
Publications
Tiekstra, M., Hessels, M. G. P., & Minnaert, A. E. M. G. (2009). Learning capacity in adolescents with mild intellectual disabilities. Psychological Reports, 105, 804-814. doi:10.2466/PR0.105.3.804-814
Hessels, M. G. P., & Tiekstra, M. (2010). Evaluation des capacités de raisonnement et prédiction des apprentissages dans un domaine scolaire nouveau chez des élèves présentant une déficience intellectuelle légère [Evaluation of reasoning capacity and prediction of learning in a novel school domain of students with mild mental retardation]. In M. G. P. Hessels, & C. Hessels- Schlatter, (Eds.), Evaluation et intervention auprès d'élèves en difficultés [Evaluation and intervention with students with difficulties] (pp. 51-66). Bern, Switzerland: Peter Lang.
Schölvink, M., Jansen, L., Van der Aalsvoort,D., Ahlers, L., & Tiekstra, M. (2011). Scenario "Responsiveness To Intervention". In M. Schölvinck, L. Jansen, & A. Minnaert (Eds.), Passend onderwijs, anders beschikken. Budgetverdeling op basis van goed onderwijs en effectief leerkrachtgedrag [Adaptive Education, rearrangement of special education. Budgeting based on quality teaching and effective teacher behavior] (pp. 89-108). Amersfoort, The Netherlands: CPS.
Prince, A., Tiekstra, M., & Minnaert, A. (2014). When theory meets practice becomes political: Can we combine scientific credibility and societal credibility? In A. Prince, Motivation: How to tame the elephant in the classroom? (Unpublished doctoral dissertation). Groningen, The Netherlands: University of Groningen.
Tiekstra, M., Minnaert, A., & Hessels, M. G. P. (2016). A review scrutinising the consequential validity of dynamic assessment. Educational Psychology, 36, 112-137. doi:10.1080/01443410.2014.915930
Tiekstra, M., & Minnaert A. (under review). At-risk students and the role of implicit theories of intelligence in educational professionals’ actions. Manuscript submitted for publication.
Tiekstra, M., Prince, A., & Minnaert, A. (in prep.). Can didactic resistance be measured by using the Response-to- Instruction model? A pilot study.
Tiekstra, M., Minnaert, A., Hessels, M., & Verdel, R. (in prep.). The next step in bridging the gap: a pilot study to enhance teacher’s skills with at-risk children.
ABOUT THE AUTHOR