lable at ScienceDirect

Teaching and Teacher Education 26 (2010) 259–266

Contents lists avai

Teaching and Teacher Education

journal homepage: www.elsevier .com/locate/ tate

The Supporting Effective Teaching (SET) project: The relationshipof inclusive teaching practices to teachers’ beliefs about disabilityand ability, and about their roles as teachers

Anne Jordan a,*, Christine Glenn a,1, Donna McGhie-Richmond b,2

a Ontario Institute for Studies in Education, University of Toronto, Department of Curriculum, Teaching and Learning, 252 Bloor St. W., Toronto, Ontario, Canada M5S 1V6b Educational Psychology & Leadership Studies, University of Victoria, A555, MacLaurin Bldg, P.O. Box 3010, Victoria, B.C., Canada V8W 3N4

a r t i c l e i n f o

Article history:Received 21 August 2008Received in revised form16 January 2009Accepted 17 March 2009

Keywords:Teacher opinionsTeaching skillsTeacher effectivenessSpecial needs studentsInclusive schoolsMainstreamingElementary

* Corresponding author. Tel.: þ1 (416) 978 0067; faE-mail addresses: ajordan@oise.utoronto.ca (A.

(C. Glenn), donnamr@uvic.ca (D. McGhie-Richmond).1 Tel.: þ1 (416) 978 0067; fax: þ1 (416) 926 4744.2 Tel.: þ1 250 721 7817; fax: þ1 250 721 6190.

0742-051X/$ – see front matter Crown Copyright � 2doi:10.1016/j.tate.2009.03.005

a b s t r a c t

The Supporting Effective Teaching (SET) project consists of studies that examine the relationshipbetween elementary general education teachers’ beliefs about disability and ability and their roles ininclusive classrooms, and how these are related to teaching practices. Teaching effectiveness isoperationally defined as multiple dimensions of teaching practices observed in inclusive classrooms. Thispaper examines previously reported and newly completed studies that investigate the characteristics ofteachers in inclusive classroom settings, what they believe about their roles and responsibilities andabout their students’ learning, and how their beliefs relate to their teaching effectiveness with studentsboth with and without disabilities.

Crown Copyright � 2009 Published by Elsevier Ltd. All rights reserved.

1. Introduction

Artiles, Kozleski, Dorn, and Christensen (2006) claim thatinclusive education is a far-reaching notion that concerns allstudents. They propose that inclusion focuses on the trans-formation of school cultures to 1) increase access (or presence) ofall students (not only marginalized and vulnerable groups),2) enhance the school personnel’s and students’ acceptance of allstudents, 3) maximize students’ participation in various domains ofactivity, and 4) increase the achievement of all students.

In a study of 11,000 students in the United States, Blackorbyet al. (2005) found that students with disabilities who spendmore time in regular classrooms had higher scores on achieve-ment tests, were absent less, and performed closer to grade levelthan their peers who were withdrawn for instruction. Overall,students with disabilities performed less well on achievementtests than those without disabilities. Yet, students with disabilities

x: þ1 (416) 926 4744.Jordan), steen72@vianet.ca

009 Published by Elsevier Ltd. All

in inclusive settings outperformed their segregated peers withdisabilities. In Canada, Demeris, Childs, and Jordan (2007) repor-ted that the number of students included in Grade 3 classrooms,and class size had no negative influence on the provincial testachievement scores of the students without disabilities, and mayindeed have contributed to a slight increase in scores on readingand mathematics.

At face value, the success of inclusive education would seem tobe inevitable. However the notion of inclusion has been poorlyaccepted in the schools. Scruggs and Mastropieri (1996) found thattwo thirds of teachers favour inclusion in principle, but less thanone third believe that inclusion can be successful with theresources available to them. Teachers’ ambivalence about inclusionincreases as they become more concerned with teaching subjectmatter, as the stakes for student achievement become moreprominent in secondary schools (Avramidis & Norwich, 2002).Ainscow (1999) and Gibbs (2007) notes that teachers’ views aboutinclusion may depend in part on the social learning processeswithin the school.

One issue is that the inclusion initiative may potentially be inconflict with policy initiatives that aim to identify teachereffectiveness and teaching quality in terms of student academicoutcomes (Florian & Rouse, 2001; Kelchtermans, 2005). In both the

rights reserved.

A. Jordan et al. / Teaching and Teacher Education 26 (2010) 259–266260

United States and Canada, work on the quality of teaching in regularclassrooms, and how instructional quality affects student achieve-ment has in part been driven by the focus on school improvementpremised on large-scale assessment that is used to hold teachersand schools accountable for student achievement. In both countriesa strong move toward local or school system-level accountabilityhas occurred in the last 20 years with the introduction of state- andprovince-wide testing of students at various grade levels, and theallocation of resources tied to the achievement outcomes ofsuch ‘‘high-stakes’’ assessments (McLaughlin & Jordan, 2005;McLaughlin & Rouse, 2000). Tensions have arisen as a result of thelimitations that the school improvement initiative has placed onthe definition of teacher quality at the expense of other valuedstudent outcomes such as inclusion. Teacher quality has becomenarrowly interpreted (Fenstermacher & Richardson, 2005), a viewof teacher quality that has ignored the value of teaching studentswho would otherwise become disenfranchised, such as those withdisabilities. In the terms used by Artiles et al. (2006) to defineinclusive education, cited above, only the fourth criterion, studentachievement, is valued as an educational outcome.

Teachers may be faced with apparently disparate messagesabout inclusive education. On the one hand they are told that theyare to meet teaching quality objectives by raising class averages instudent achievement, while on the other hand told that they are tobe responsible for diversifying instruction to meet a range oflearner needs. It is no surprise therefore that teachers expressambivalence about including students with disabilities in theirclasses.

Despite the gloomy picture, there are exemplary schools thatcontribute to high levels of inclusion and also rate highly on overallstudent achievement. In the U.K., Dyson, Farrell, Polat, Hutcheson,and Gallannaugh (2004) found that schools that are effective ininclusion develop unique ways to adapt to their local communities.Dyson, Polat, and Farrell (2004) suggest that effective schoolsdevelop an ‘‘ecology of inclusion’’ (p. 14). Florian and Rouse (2001)note that when schools have access to a variety of supports andteaching strategies they can be effective both in inclusion and insustaining high levels of student achievement.

In considering how to help educational systems become moreinclusive, the nature of teachers’ beliefs and how beliefs relate totheir consequent actions need to be understood (Gibbs, 2007). Howdo teachers cope with competing professional demands and dotheir responses influence who they teach and how effective theyare in meeting the range of student needs in their classrooms?What differences do they exhibit in their beliefs about their rolesand responsibilities for including students with disabilities, and forteaching students who are at risk of academic failure? In otherwords, what are their professional priorities for and beliefs abouttheir roles in promoting learning in inclusive classrooms and howdo these relate to how they practice?

2. The SET project

The Supporting Effective Teaching (SET) project commenced in1992. Its purpose is to develop a model of teacher characteristicsand school-related factors that predict the effectiveness of generaleducation (regular) elementary classroom teachers’ practices withstudents with disabilities included in their classrooms. In variouspublications the elements of the model have been explored (Jordan& Stanovich, 2001, 2003, 2004; McGhie-Richmond, Underwood, &Jordan, 2007; Stanovich & Jordan, 1998, 2002, 2004). Fourconstructs are central to the model: (1) teachers’ beliefs aboutdisability and their roles and responsibilities for students withdisabilities; (2) teachers’ practices in core subjects in theirclassrooms; (3) their practices in accommodating students with

disabilities and those at risk of school failure; and (4) the influenceof the school norm, that is, the collective beliefs or prevailing ethosof the administrators and staff in the school about their roles withand responsibilities for all students.

In the SET project, teaching effectiveness is operationallydefined by teachers’ scores on the Classroom Observation Scale(COS; Stanovich, 1994; Stanovich & Jordan, 1998), a scale derivedfrom the synthesis of effective teaching skills reported by Englert,Tarrant, and Mariage (1992). The COS contains criteria for ratingteachers on three groups of teaching characteristics as delivered tothe class as a whole: time management, classroom managementand lesson presentation and delivery (Appendix A). A discriminantfunctions analysis of the 32 items in the COS revealed that 5 itemsdiscriminate more from less effective teachers, resulting in a factorthat indicates the teachers’ engagement of learners (McGhie-Richmond et al., 2007). The COS also contains observational ratingsof the extent and quality of lesson-related interactions between theteacher and two or more students who are designated as havinga disability or are nominated by the teacher to be at risk of failure.Thus the operational definition of teaching effectiveness in thisproject is multi-dimensional. It includes a range of teaching skillsidentified as being effective in general education classrooms(Englert et al’s., 1992 synthesis, the student engagement factor), aswell as measures of the length and type of instructional interac-tions between the teacher and students both with and withoutdisabilities and those at risk in the classroom.

3. Research questions

The main purpose of this paper is to present the findings of theSET research program. Rather than follow a standard reportingformat, a series of research questions will be addressed from theresults of various studies:

1. Are teachers who are effective overall with all their studentsalso the most effective in including students with disabilitiesand those at risk in their general education classrooms?

2. Do teachers differ in what they believe about disabilities andtheir roles with and their responsibilities for their studentswith disabilities? If so, are differences in such beliefs related toa larger set of epistemological beliefs about the nature of abilityand how children learn?

3. Are differences in teachers’ epistemological belief structuresreflected in how they practice in inclusive settings? Is anydifference reflected both in their observed practices, and intheir self-reported instructional preferences?

4. How are differences in the beliefs and practices of teachersreflected in the opportunity to learn provided to students withdisabilities and students at risk in inclusive classrooms?

5. What are the implications of the project’s findings for guidingthe development of effective inclusive practices, such asinfluencing teachers’ beliefs about the nature of ability anddisability, and their roles and responsibilities in teachingstudents with diverse needs?

1. Are teachers who are effective overall with all their students alsothe most effective in including students with disabilities and thoseat risk in their general education classrooms?

This is perhaps the most difficult question to address, because,as noted above, teaching effectiveness can be measured throughachievement gains in non-disabled students, but is more difficult todetermine for students with disabilities and those who areunderachieving. The student outcomes question was addressed inthe SET project in several studies.

A. Jordan et al. / Teaching and Teacher Education 26 (2010) 259–266 261

Jordan, Lindsay, and Stanovich (1997) recorded and analysed thetalk of nine teachers in Grade 3 classrooms to their students withand without disabilities. During several lessons in each classroom,an observer rated the amount and type of teacher talk witha student who was regularly achieving, one who was of concern tothe teacher as being at risk of academic failure, and one formallydesignated as having a disability. A 5-point rating scale ranged fromno dialogue with the student, cursory management comments bythe teacher, and instructional transmissions, to interactive dialoguebetween the teacher and student. The amount of time was assessedin seconds, and was recorded as instructional or non-instructional(i.e., management, discipline, social etc.). The study showed that,first, the amount of instructional time is correlated to the time andclassroom management and lesson presentation skills of theteachers, and varies considerably from one teacher to the next.Teachers conserve instructional time by establishing clear routines,well understood roles for both teacher and students, and bydelivering well-designed lessons with a clear beginning, middleand end. Second, the more efficient the teacher at conservinginstructional time, the more the students receive extendeddialogues with the teacher, both individually and in small groups,that are intended to develop higher order thinking skills. Fewteachers regularly reached this level of instructional engagement ofstudents. It is a relatively rare phenomenon that was noted in lessthan a third of the group. Finally the teachers who scored highly onclassroom and time management and lesson presentation skillswere able to get around to their students at risk and those withdisabilities far more often, and for more minutes, than teacherswith lower scores on the COS items. Indeed, students withdisabilities and at risk in the classrooms of the three highestscoring, and arguably most effective, teachers received moreinstructional time at higher levels of cognitive engagement thanthe typically-achieving students in the classes of the low-scoringteachers.

An attempt was made to assess student outcomes using thePiers Harris Self-Concept Scale (Jordan & Stanovich, 2001).Although this instrument is a somewhat blunt and inadequatemeasure of student outcomes, as expected the students withdisabilities and at risk in all classrooms had lower Piers HarrisSelf-Concept Total scale scores than typically-achieving students.However, both groups of students had higher Self-Concept Totalscale scores in the classrooms of the more effective teacherscompared to those with the less effective teachers.

In effect how teachers use instructional time is related to howthey engage students in learning. This in turn is related to the depthand breadth of cognitive engagement of students with theinstructional material. More instructional time and greater studentengagement is also available to students who have learningdifficulties.

This study led Stanovich (1994, Stanovich & Jordan, 1998) todevelop a broader observational instrument to gauge teachingeffectiveness, the Classroom Observation Scale (COS). Based ona synthesis of the literature on effective teaching (Englert et al.,1992), the Classroom Observation Scale contains 27 observer-ratedcriteria of teaching behaviours in three groups; time management

Table 1Correlations between COS total, student engagement, predominant teaching style, and t

Factor Classroom observationscale total

Student Engagement .824**Predominant Teaching Style .766**Teacher Interactions with Exceptional Students .372*Teacher Interactions with Students At-Risk .101

*p < .05. **p < .001.

(8 items), classroom management (8 items), and lesson presenta-tion (11 items) (Appendix A). The items in each section require theobserver to code whether the teacher exhibits each practiceconsistently (twice or more), inconsistently (once), or not at all,during a half-day of teaching core academic subjects such aslanguage arts, science and mathematics. Stanovich and Jordan(1998) report that the mean percentage agreement between twoobservers independently coding the COS across 33 teachers was78%. McGee (2004) reported an inter-rater reliability of .73 across30 teacher observations.

In addition to the 27-item ratings, the observer codes theinteractions of the teacher with the class as a whole (PredominantTeaching Style – PTS), and with one student with a disability, andone deemed by the teacher to be at risk of academic failure (TeacherInteraction scales; TID, TIAR). The teacher is not told whichstudents are being specifically tracked. The Teacher Interactionscales with an exceptional and at-risk student are considered toprovide a measure of the quality of teachers’ instructional inter-actions with each group. The three scales, the PTS, TeacherInteraction with a Student with a Disability (TID), and witha Student at Risk (TIAR) are 7-point scales reflecting the depth andbreadth of the teacher’s cognitive engagement of the class and ofthe individual students through dialogue. Low scores reflect little orno lesson-related interaction, while high scores reflect a dialogue inwhich teacher and students interact to develop student under-standing of lesson concepts (Jordan et al., 1997),

The relationship between effective teaching, as measured by theitems of the COS, and the quality of student–teacher academicinteraction, as measured by the PTS and the TID and TIAR measureswith students with disabilities and students at risk, has beenexamined in several studies. McGee (2001) found that the COS andinteraction scales observed with 13 teachers were significantlycorrelated (r ¼ .76, p < .01). Glenn (2007) reports that COS andinteraction measures during half-day classes of 32 elementaryteachers also showed a positive correlation between the PTS withthe total class and teachers’ interactions with students withdisabilities (r (31) ¼ .760, p < 001; r (30) ¼ .348, p < .05). In otherwords, teachers who are effective overall with their classes are alsoeffective in working individually with students with disabilities.Students at risk, however, did not fare as well, and seemed to beless successful in gaining teachers’ individual attention. Glenn’s(2007) results appear in Table 1.

It has been firmly established that an important contributor toteacher effectiveness is the teacher’s ability to engage students inlearning (e.g. Gersten, Fuchs, Williams, & Baker, 2001; Kucan &Beck, 1997). The results of the SET project’s observational studiesconfirms this. McGhie-Richmond et al. (2007) analysed the resultsof COS observations of 63 elementary classroom teachers, collectedover five years of the project, to identify which items discriminatedthe highest scoring from the lowest scoring teachers. A cluster offive items clearly discriminated the most effective teachers; fouritems relating to the establishment of clear lesson expectations andengaging student attention, and one to maintaining high rates ofstudent responses with prompting, error correction and feedback(Student Engagement, Table 2).

eacher interactions with exceptional and student at-risk (n ¼ 32).

Studentengagement

Predominantteaching style

Teacher interactionswith exceptional Students

1.000.606** 1.000.322 .441** 1.000.005 .209 .516**

Table 2COS items in discriminant function 1: student engagement.

Function 1 – Accounts for 95% of variance N ¼ 63, eigenvalue ¼ 6.93

B2 States expectations for seatwork in advanceB3 Establishes clear lesson routines that signal a beginning and an endB4 Gains students’ attention at the beginning of the lesson and maintains it during

instruction at 90% levelB5 Monitors transitions by scanning and circulating among studentsC7 Maintains high accurate responding rate in teacher-led activities:

- repeats practice opportunities until students not making errors- delivers instructional cues and prompts- provides error correction procedures- uses prompting and modelling following errors

A. Jordan et al. / Teaching and Teacher Education 26 (2010) 259–266262

The results led Stanovich and Jordan (2000, 2004) to posit thateffective teaching is akin to effective inclusion, and conversely thatacquiring teaching skills that are effective for students with disabil-ities benefits the development of overall teaching effectiveness.

2. Do teachers differ in what they believe about disabilities andtheir roles with and their responsibilities for their students withdisabilities? If so, are differences in such beliefs related to a largerset of epistemological beliefs about the nature of ability and howstudents learn?

a. Beliefs about Disability. Through the SET project studies,differences in the beliefs that teachers hold about the nature ofdisability have been identified. The instrument designed tomeasure beliefs is a narrative interview between the teacher andresearcher in which the teacher recounts the sequence of events fortwo students with whom the teacher has worked; the P-I interview.Approximately one quarter of the teachers interviewed hold‘pathognomonic’ (Path) beliefs, that disability is an internal, fixed,and pathological condition of the individual that is not amenable toinstruction. The term pathognomonic literally means ‘to name thesuffering’. As a result, these teachers emphasise the label ordesignated disability as the explanation for underachievement, anddeem students with disabilities and those who are underachievingto be the source of their own learning difficulties. Teachers withpredominantly Path beliefs tend to place the blame on the studentsthemselves and/or on their families for inability to progress.Consequently these teachers spend little time and effort workingwith their included students with disabilities and those who are atrisk of academic failure, compared to the rest of the students intheir classes. They prefer delivery models that withdraw studentsfrom their classes, and their concepts of instructional accommo-dations feature homework to be completed by the student withparents, and curriculum delivered in a resource or segregated classsetting (Jordan, Kircaali, & Diamond, 1993; White, 2007).

At the other end of the continuum, teachers with ‘interven-tionist’ (Int) beliefs view disability as created in part by a societythat is designed for the able, and that creates barriers for those whohave disabilities. Such teachers, who represent about 20% of oursamples, see their responsibility as being to create access tolearning, by reducing barriers to learning through accommodationsthat increase access and by working longer and at greater levels ofintensity with their students with learning difficulties. (Jordanet al., 1997; Jordan & Stanovich, 2001; White 2007). It should benoted that beliefs are not dichotomous, but are represented asa continuum, with about half of the teachers interviewedexpressing components of both views, and varying from oneclassroom and school situation to another.

The P-I interview administration, scoring and analysis have beenreported elsewhere (Stanovich & Jordan, 1998). Inter-rater reliabilitybetween raters independently scoring the transcripts of the inter-views was reported by Jordan et al. (1997) to be .88, and by McGee

(2001) to be .91. Stanovich and Jordan (1998) reported a Cronbach’sa co-efficient of .89 for the scoring criteria. Glenn (2007) reports thatamong a sample of 33 teachers, the P-I scores ranged from 1.40 to3.00, with a mean score of 2.46 (SD ¼ .50). This is consistent withother samples using the P-I measure (McGee, 2004).

Two subscales of the P-I rating are Teacher Responsibility, theview that the teacher is primarily responsible for developingopportunities to learn, and Teacher Attribution, whether theteacher ascribes learning difficulties either to the student andfamily or to prior and current opportunity to learn. Descriptivestatistics for these measures appears in Appendix A. The correla-tions between the total P-I score and the Responsibility subscale isr(32) ¼ .66 (p < .01), and between Responsibility and Attribution isr(32) ¼ .55 (p < .01).

b. Beliefs about Ability. Teacher beliefs about the nature ofdisability and about their roles in working with students withdisabilities share many of the attributes of beliefs about ability, andthe larger issues of epistemological beliefs, about the nature ofknowledge, knowing, and how knowledge is acquired. Two studiesexamined whether there was such a relationship, such that the P-Iconstruct of beliefs about disability might be part of a larger set ofepistemological beliefs held by teachers about the nature ofknowledge, ability and learning.

Glenn (2007; Glenn, Schwartz, & Jordan, 2007) developed andadministered a questionnaire, the Beliefs about Learning andTeaching Questionnaire (BLTQ). The BLTQ was based on a survey ofteacher beliefs about mathematics (Stipek, Givvin, Samon, &McGyvers, 2001) and was administered to 280 teachers andteacher-candidates. The results provided a four-factor structure andhigh item reliability co-efficients. Replicating Stipek at al’s factorsolution, Glenn was able to identify teachers’ personal epistemo-logical theories about the nature of ability as either ‘an entity’ (E) oras ‘incremental’ (I). Entity beliefs (Dweck, 2000; Stipek et al., 2001)hold that ability is a fixed and static trait, likely present at birth,and, particularly at the lower end, has a limited responsiveness tolearning (not unlike the pathognomonic view of disability). Incre-ment beliefs view ability as malleable, influenced by learning,amenable to improvement and developing incrementally under theright learning conditions (similar to the views held by those with aninterventionist view of disability). The mean total E-I score for the33 teachers was moderate (M¼ 18.86, SD¼ 3.42). The continuum ofEntity-Increment beliefs about ability appears to share character-istics of the Pathognomonic-Interventionist beliefs about disability.

Glenn administered the BLTQ to 33 teachers who had previouslybeen observed and scored on the COS, and who had completed theP-I interview. The teachers’ scores on the self-rated E-I factor of theBLTQ was moderately, positively correlated with the teachers’beliefs about their roles and responsibilities in working withstudents with disabilities (P-I); r(32) ¼ .372, p < .05. Descriptivestatistics of Glenn’s measures appear in Appendix B.

A significant correlation was also found between the P-Isubscale of Attribution and E-I, suggesting that teachers whotended to attribute a student’s learning difficulties to characteris-tics internal to the student (e.g., ability, IQ, or exceptionality) heldmore entity beliefs as measured by the E-I factor. A significantcorrelation between Responsibility and E-I suggests that teacherswho tended to use a student’s exceptionality as justification for notbeing responsible for the instruction of that student also tended tohold entity beliefs.

Evidently the research suggests that there is a relationshipbetween elementary classroom teachers’ beliefs about the fixed ormalleable nature of both disability and ability, possibly indicatingan underlying beliefs construct about the nature of ability to learn,as broadly conceived. The source of such beliefs may be contextualrather than an attribute of the individual, reflecting the cultural

Table 3Correlations between P-I, responsibility, attribution, E-I, COS, student engagement, predominant teaching style, and teacher interactions with exceptional and students at-risk(N ¼ 186).

Factor Pathognomonic-interventionist

Responsibility Attribution Entity-increment

Classroom Observation Scale .077 .440* .304 .261Student Engagement .185 .431* .397* .381*Predominant Teaching Style .280 .428* .361* .284Teacher Interactions with Exceptional Students .178 .318 .378* .189Teacher Interaction with Student At-Risk .054 .045 .205 .346*

*p < .05. **p < .01.

A. Jordan et al. / Teaching and Teacher Education 26 (2010) 259–266 263

norm or beliefs of the staff of the school as a whole (Gibbs, 2007;Salisbury, 2006).

3. Are differences in teachers’ epistemological belief structuresreflected in how they practice in inclusive settings? Is any differ-ence reflected both in their observed practices, and in their self-reported instructional preferences?

Two sources suggest that there may be a relationship betweenthe belief structures of teachers about ability and disability and theway they practice. In Glenn’s (2007) study, the correlation betweenE-I and P-I and overall classroom practices as measured by the COSfailed to reach significance. The Student Engagement subscales ofthe COS did however correlate with E-I, and with the Responsibilityand Attribution subscales of the P-I (Table 3). That is, those teacherswho rated high on the student engagement items of the COS thatdiscriminated high- from low-scoring teachers also describedtaking more responsibility for the learning of their students withdisabilities, and tended to attribute learning difficulties to factorswithin their influence, such as using accommodations and adaptinginstructional techniques.

Predominant Teaching Style in the COS, the prevalent interactivestyle of the teacher with the class as a whole, also correlated withthe Responsibility and Attribution subscales of the P-I. Teacherswho engage their students in higher levels of instructional dialoguealso expressed taking responsibility for their students’ learning andattributed their students’ learning difficulties to factors that theteacher is able to influence (Table 3).

3.1. Teachers’ described practices and instructional styles

The Beliefs about Learning and Teaching Questionnaire – BLTQ,which was used to obtain teachers’ E-I beliefs ratings, containedthree further factors that illuminate teachers’ preferences forinstructional styles. In a factor analysis of the BLTQ with 280teachers, Glenn (2007) reports the following factors:

1. Teacher-led instruction, (item example: It is important forstudents to complete assignments exactly as the teacherplanned). High scores on this factor reflect a transmissive styleof teaching in which the teacher is the source of knowledge.

Table 4Correlations between P-I, responsibility, attribution, E-I, teacher-controlled instruction, s

Factor Pathognomonic-interventionist

Responsibility Attr

Responsibility .662** 1.000Attribution .287 .553** 1.00Entity-Increment .372* .539** .47Teacher-Controlled Instruction �.087 �.339 �.34Student-Centered Instruction .438* .463** �.01Attaining Standards �.541** �.571** �.54

*p < .05. **p < .01.

2. Student-centered instruction, (item example: Good teachersgive students choices in their learning tasks). High scores onthe items in this factor reflect a focus on student-initiatedlearning, with the teacher as guide.

3. Attaining Standards (item example: Giving grades is a goodstrategy for getting students to work). This measure distin-guishes a preference for techniques of extrinsic over intrinsicmotivation, such as getting good grades.

The correlations between teachers’ P-I beliefs and the remainingthree factors were examined. A significant positive correlation wasfound between Student-Centered Instruction and P-I, r(32) ¼ .438,p < .05, and between Attaining Standards and P-I, r(32) ¼ �.541,p < .01. The correlation between P-I and Teacher-ControlledInstruction did not reach significance. Teachers with Interventionistbeliefs about disability were more likely to report that theypreferred student-centered instruction, and were less likely toreport motivating their students through extrinsic sources such asgrades.

Further support for the relationship between P-I beliefs andpractices was provided by the Attribution and Responsibilitysubscores of the P-I and the three remaining BLTQ factors,revealing a strong negative correlation between Attribution andAttaining Standards, r(32) ¼ �.545, p < .01, and betweenResponsibility and Attaining Standards, r(32) ¼ �.571, p < .01.While no significant correlations were found between Attributionand Student-Centered Instruction, or Attribution, Responsibilityand Teacher-Controlled Instruction, a moderate positive relationwas found between Responsibility and Student-Centered Instruc-tion, r(32) ¼ .463, p < .01. On the BLTQ, there was a significantnegative correlation between E-I and teacher-controlled instruc-tion (r(32) ¼ �.69, p < .01) (Table 4). Teachers who reporteda student-centered approach to learning in the BLTQ alsodescribed taking greater responsible for student learning in theirP-I interviews. Teachers who reported a preference for teacher-ledinstruction were likely to report less incremental, more entityviews of ability.

There is some evidence to suggest therefore that teachers seemto differ in their self-reported preferences for different practicessuch as teacher-led instruction or student-centered instruction,and methods of motivating students that are consistent with theirbeliefs about the fixed or malleable nature of ability and disability.

tudent-centered instruction and attaining standards (N ¼ 36) (Glenn, 2007).

ibution Entity-increment Teacher-controlledinstruction

Student-centeredinstruction

09** 1.0007 �.692** 1.0001 �.251 �.369 1.0005** �.368 �.525** �.042

A. Jordan et al. / Teaching and Teacher Education 26 (2010) 259–266264

Further, as shown in Table 4, there seems to be a link, as yettentative, between the practices that we observed and theexpressed beliefs and instructional preferences that teachersreported. Glenn’s findings add support to our contention that thereare general epistemological belief structures about the nature ofability, disability and learning that are linked to the decisionsteachers make about how they teach and to whom, and to theirpreferences for teaching styles.

4. How are differences in the beliefs and practices of teachersreflected in the opportunity to learn provided to students withdisabilities and students at risk in inclusive classrooms?

While a moderate correlation was found between effectivepractices as measured by the COS and interactions with excep-tional students, (Table 1, r(32) ¼ .372, p < .05), the COS total scoredid not correlate with interactions with students at risk. Therelationships between Student Engagement and interactions withstudents who are exceptional and those at risk were not signifi-cant, although teachers’ predominant teaching style correlatedwith their interactions with students with disabilities (Table 1,r(32) ¼ .44, p < .01). It appears that effective teachers pay atten-tion to their students with disabilities although they may overlookthose at risk.

In summary, the results of Glenn’s study must be considered aspreliminary, since the variables in the study need to be developedwith larger numbers of teachers to confirm the relationshipbetween them. The data do however suggest a link between howteachers interact with their students both with and withoutdisabilities and their beliefs about ability and disability and abouttheir roles in fostering learning. These are intriguing findings.Teachers’ epistemological beliefs, beliefs about disability, and abouttheir responsibilities for students with disabilities, may be relatedto the quality of their overall teaching practices. This is contrary tothe claim that teaching students with disabilities in inclusiveclassrooms detracts from the teachers’ time with non-disabledstudents. Our evidence suggests that the most effective teacherswork more with their students with disabilities at higher levels ofengagement. They also engage their other students at higher levelsof thinking and responding, through dialogical interactions thatfeature questioning, providing cognitive extension and elaboratestudents’ thinking.

In her doctoral study, Schwartz (2008; Schwartz, Glenn, &Jordan, 2008,) analysed in depth the statements of 12 teachersduring the P-I interview, mapping them onto the theoreticalmodels of epistemological beliefs (Baxter Magolda, 2002; Hofer,2001; Hofer & Pintrich, 1997; Schommer-Aikins, 2002; Schraw &Olafson, 2002). Schwartz discovered that clusters of beliefs not onlyvalidate the dimension of P-I, but also reflect diversity andcomplexity of teachers’ understanding of the nature of knowing,knowledge, and learning, and their relationships to differences inteachers’ preferred styles of teaching, setting assignments andmotivating learning. Schwartz posits clusters of beliefs that eitherfocus on the fixed, universal nature of knowledge or on the fluid,relative nature of knowledge, and whether knowing is a collectiveor personal attribute. Teachers’ described teaching styles flow fromthese distinctions. Schwartz showed that teachers’ descriptions ofthe decisions they make, the steps they take to seek support forstudents, and the extent to which they accommodate learningdifficulties are aligned with differences in their profiles of episte-mological beliefs.

An important question is how malleable are the beliefs of theteachers themselves. A qualitative thesis study (White, 2007)shows how intransigent are the beliefs of teachers about their roleswith students with disabilities and the nature of disability. Whiteexamined the beliefs and practices of five teachers who were part of

the SET project in 1999 and continued to participate in 2004, to seehow they changed over time and as a result of teaching experi-ences. All teachers were employed by a school system that placeda great deal of emphasis on inclusion and provided considerableresources and professional development to ensure its success.White found that the thinking of teachers over the intervening timeperiod was largely predicted by the beliefs with which they enteredthe project.

5. What are the implications of the project’s findings for guiding thedevelopment of effective inclusive practices, such as influencingteachers’ beliefs about the nature of ability and disability, and theirroles and responsibilities in teaching students with diverse needs?

The implications for of the SET project are considerable forteacher training and development (Jordan, Schwartz, & McGhieRichmond, 2009). They contribute to the literature (Fang, 1996;Fenstermacher, 1993; Kagan, 1992; Munby, Russell, & Martin, 2001;Pajares, 1992; Stipek et al., 2001; Yoon, 2008) supporting the notionthat the overall epistemological understanding that teachers bringto their teaching practices may have tremendous implications forhow teachers develop their skills and practice as educators. Thedevelopment of interventionist beliefs and inclusive practices is noteasily achieved. Pajares (1992) and Berry (2006, 2008) reflectWhite’s (2007) conclusion that teacher beliefs about inclusion arerelated to other beliefs about equity and fairness and are relativelyhard to change. Our findings, that beliefs about disability and abilityshare a common understanding in general, and are based ondiffering perspectives of how learning occurs, suggest that teachersmay need a broader understanding of learning and how knowledgeis acquired.

The school environment may have a significant influence onteachers’ beliefs and their views about their roles and responsi-bilities in inclusive settings. Factors such as leadership andcollegial perspectives may especially influence newer teachers.Although this variable, termed school norm, was significant inpredicting teaching practices in earlier studies (Stanovich, 1994;Stanovich & Jordan, 1998) it has not been reported here and wasnot featured in recent studies. However, as noted above, theinfluence of school norm, the culture or ethos of the school (Dyson,Polat, et al., 2004), may be highly influential in shaping teachers’beliefs. Therefore, beliefs may need to be interpreted as sociallycontextualized by the views held in the school rather than aspersonal characteristics. In order to effect change in teacher beliefsabout inclusion, Gibbs (2007) proposes that there needs to bea synergy between teacher perception of their self efficacy and thecollective efficacy within their school. The research of Dyson,Farrell, et al. (2004) notes how fragile this synergy may be:

Where schools have relatively high proportions of [SpecialEducation needs] pupils, there appears to be a delicate balancebetween the resources they can bring to bear on the task ofteaching and the demands which the presence of these childrencreate. Through good policies, good leadership, and the skillsand energies of teachers, this balance is often maintained.However it would seem that it does not take much to disturbit – a shortfall in classroom support, a weakness in teacher skillsor managerial planning, a lack of funding, and so on (p. 105).

Despite competing professional demands and variableresources, some teachers are able to cope with both high achieve-ment standards and inclusive practices by espousing a beliefsystem and adopting a series of instructional practices that areeffective for all their students. Those teachers were deemed to bethe most effective among their peers when observed in theirinclusive classrooms. The research reported here suggests thateffective teachers conserved instructional time by preparing

A. Jordan et al. / Teaching and Teacher Education 26 (2010) 259–266 265

lessons that clearly communicated expectations that engage allstudents in learning. They expected high standards for all students.They established routines that enabled them to spend largeamounts of instructional time with individuals and small groups.These teachers then use their time to engage students individuallyand in small groups, including those with disabilities and thosestruggling with the lesson, in dialogical interactions designed tofoster thinking skills and student understanding. It may thereforebe not only possible but also desirable for teachers to align theirsystems of epistemological beliefs with their inclusive practices andthe requirement to raise achievement standards by applying highlyeffective instructional methods to the benefit of all their students.

Acknowledgements

Role of the funding source. This research was supported bygrants from the Social Sciences and Humanities Research Council ofCanada. This funding source had no role in the design, execution orreporting of this study.

Appendix A

Item samples from the Classroom Observation Scale (Stanovich,1994; Stanovich & Jordan, 1998) developed from Englert et al.(1992).

(Items marked * were identified by McGhie-Richmond et al.,2007 as the student engagement factor; items that discrimi-nated the highest scoring teachers from their lower scoringpeers).

A. Classroom Management

1. Arranges physical space to maintain minimally disruptivetraffic patterns and procedures.

2. Rules and procedures exist for non-instructional events (e.g.,movement about room, student talk, distributing materials,bathroom use, etc.) and for instructional events (e.g., gettingready for lessons, expected behaviour of instructional group,obtaining help, seatwork procedures, out-of-seat proce-dures, etc.)

3. Evidence of rules that involve respect for other members ofclass and/or provides verbal reminders to students about howto treat each other.

4. Consequates rule non-compliance quickly; cites rule or proce-dure in responding to disruptive behaviours.

5. Positions self in room to provide high degree of visibility (e.g.,can make eye contact with all students).

6. Scans class frequently.7. Uses non-verbal signals whenever possible to direct students

in a non-disruptive manner when teaching other groups ofstudents.

8. Administers praise contingently and uses specific praisestatements.

B. Time Management

1. Allocates generous amounts of time for instruction (limitstime spent on behaviour management, recess, and non-academic activities, keeps transition time between lessonsshort).

2. States expectations for seatwork and transitions in advance(e.g., prepares students for transitions in advance by statingbehavioural expectations and informing students that lesson isdrawing to a close).*

3. Establishes clear lesson routines that signal a beginning andend.*

4. Gains students’ attention at the beginning of the lesson andmaintains attention during instruction at 90% level.*

5. Monitors transition by scanning and circulating amongstudents.*

6. Maintains students’ attention during seatwork at 86% or higher.7. Circulates frequently among seatwork students to assist

students and to monitor progress.8. Provides active forms of seatwork practice clearly related to

academic goals.

C. Lesson Presentation

1. Provides review of previous day’s concepts at beginning oflesson; actively tests students’ understanding and retention ofprevious day’s lesson content.

2. Provides a clear overview of the lesson:3. Actively model and demonstrate concepts, learning strategies,

and procedures related to effective problem solving in thecontent area:

4. Maintains a brisk pace during the lesson.5. Provides frequent questions to evaluate students’ mastery of

lesson concepts.6. Evaluates students’ understanding of seatwork tasks and

cognitive processes by asking students ‘‘what, how, when,where, why’’ questions related to the targeted skill or strategy.

7. Maintains high accurate responding rate (70–90%) in teacher-led activities:*a. Repeats practice opportunities until students are not

making errors.b. Delivers instructional cues and prompts.c. Provides error correction procedures.d. Using prompting or modelling following errors rather than

telling the answer.8. Provides error drill on missed concepts or review of difficult

concepts during the end of each lesson.9. Gives summary of the lesson content and integrates lesson

content with content of other lessons or experiences.10. Summarizes the lesson accomplishments of individuals and

group.11. Forecasts upcoming lesson content.

D. Predominant Teaching Style

1 2 3 4

1 2. INC 3. CON 4. INC 5. CON 6. INC 7. CON CHECKING TRANSMITTING ELABORATES No interactions

with studentson lessoncontent.

Teacher circulates,checking workbriefly andmoving on(brief and cursory).

Teacher circulates,transmitting anddirecting lessonresponses(tells studentswhat to work on,how to correct it,and moves on).

Teacherelaborates(asks studentsquestions aboutlesson materialconcepts thatrequire responses;interactionrequiring studentparticipation).

Inc – observed as inconsistent (once only); Con – observed to be consistent (twice ormore).

E. Teacher Interaction with a) an exceptional student,b) a student at risk:

(Same scale as in D.)

A. Jordan et al. / Teaching and Teacher Education 26 (2010) 259–266266

Appendix B

Table B1. Descriptive Statistics for the P-I, COS and BLTQ(Glenn, 2007)

Factor N Scale Range Score Range Mean SD

P-I Total Score

33 1.00–3.00 1.40–3.00 2.46 .50 Responsibility 33 1.00–5.00 1.50–5.50 4.22 1.15 Attribution 33 1.00–5.00 1.50–4.50 2.82 .87 COS 36 .00–64.00 23.00–64.00 49.33 10.63 Student Engagement 36 .00–10.00 2.00–10.00 8.00 2.29 Predominant Teaching Style 35 .00–7.00 2.00–7.00 5.44 1.32 Teacher Interactions with

Exceptional Student

35 .00–7.00 1.00–7.00 4.71 1.87

Teacher Interactions withStudent At-Risk

36

.00–7.00 2.00–7.00 4.36 1.88

Teacher-ControlledInstruction

36

6.00–36.00 8.00–29.00 17.34 4.97

Entity-Increment

36 4.00–24.00 9.00–24.00 18.86 3.42 Student-Centered

Instruction

36 6.00–36.00 24.00–37.00 30.81 2.98

Attaining Standards

36 4.00–24.00 4.00–20.00 11.38 3.56

References

Ainscow, M. (1999). Understanding the development of inclusive schools. London:Falmer Press.

Artiles, A. J., Kozleski, E. B., Dorn, S., & Christensen, C. (2006). Chapter 3: learning ininclusive education research: re-mediating theory and methods with a trans-formative agenda. Review of Research in Education, 30, 65–108.

Avramidis, E., & Norwich, B. (2002). Teachers’ attitudes toward integration/inclusion:a review of the literature. British Journal of Special Needs Education, 17(2),129–147.

Baxter Magolda, M. (2002). Epistemological reflection: the evolution of episte-mological assumptions from age 18 to 30. In B. Hofer, & P. Pintrich (Eds.),Personal epistemology: The psychology of beliefs about knowledge and knowing(pp. 89–102). Mahwah, New Jersey: Erlbaum Associates, Inc.

Berry, R. W. (2006). Beyond strategies: teacher beliefs and writing instruction intwo primary inclusion classrooms. Journal of Learning Disabilities, 39(1), 11–24.

Berry, R. W. (2008). Novice teachers’ conceptions of fairness in inclusion class-rooms. Teaching and Teacher Education, 24(5), 1149–1159.

Blackorby, J., Wagner, M., Cameto, R., Davies, L., Levine, P., & Newman, L. (2005).Engagement, academics, social adjustment, and independence. Palo Alto: CA: SRI.

Demeris, H., Childs, R. A., & Jordan, A. (2007). The influence of students with specialneeds included in grade-3 classrooms on the large-scale achievement scores ofstudents without special needs. Canadian Journal of Education, 30(3), 609–627.

Dyson, A., Farrell, P., Polat, F., Hutcheson, G., & Gallannaugh, F. (2004). Inclusion andpupil achievement. London, UK: Department for Education and Skills. (ResearchRep. No. RR578) Available from. http://www.dfes.gov.uk/research/data/uploadfiles/ACFC9F.pdf Retrieved 28.06.07.

Dyson, A., Polat, F., & Farrell, P. (2004, April). Inclusion and student achievement.Paper presented at the annual meeting of the American Educational ResearchAssociation, San Diego, CA.

Dweck, C. S. (2000). Self theories: Their role in motivation, personality and develop-ment. Philadelphia: Taylor and Francis.

Englert, C. S., Tarrant, K., & Mariage, T. (1992). Defining and redefining instructionalpractice in special education: perspectives on good teaching. Teacher Educationand Special Education, 15(2), 62–86.

Fang, Z. (1996). A review of research on teacher beliefs and practices. EducationalResearch, 38, 47–65.

Fenstermacher, G. D., & Richardson, V. (2005). On making determinations of teacherquality. Teachers College Record, 107(1), 186–213.

Fenstermacher, G. D. (1993). The knower and the known: the nature of knowledgein research on teaching. Review of Research in Education, 20, 3–56.

Florian, L., & Rouse, M. (2001). Inclusive practices in English secondary schools:lessons learned. Cambridge Journal of Education, 31(3).

Gersten, R., Fuchs, L. S., Williams, J. P., & Baker, S. (2001). Teaching readingcomprehension strategies to students with learning disabilities: a review ofresearch. Review of Educational Research, 71(2), 279–320.

Gibbs, S. (2007). Teachers’ perceptions of efficacy: beliefs that support inclusion orsegregation. Educational and Child Psychology, 24(3), 47–53.

Glenn, C. (2007). The impact of teachers’ epistemological beliefs and their beliefs aboutdisability on their teaching practices in inclusive classrooms. Unpublished Ph.D.dissertation, University of Toronto.

Glenn, C., Schwartz, E., & Jordan, A. (2007). Teacher beliefs and inclusive teaching practices:Who is affected?. Paper presented at the meeting of the International Study Asso-ciation of Teachers and Teaching, Brock University, St. Catharines, Ontario, July.

Hofer, B. (2001). Personal epistemological research implications for learning andteaching. Educational Psychology Review, 13(4), 353–383.

Hofer, B., & Pintrich, P. (1997). The development of epistemological theories: beliefsabout knowledge and knowing and their relation to learning. Review ofEducational Research, 67(1), 88–140.

Jordan, A., Kircaali, G., & Diamond, C. T. P. (1993). Who has the problem, the studentor the teacher? Differences about teachers’ beliefs about their work with at-riskand integrated exceptional students. International Journal of Disability, Devel-opment and Education, 40(1), 45–62.

Jordan, A., Lindsay, L., & Stanovich, P. J. (1997). Classroom teachers’ interactions withstudents who are normally achieving, at-risk and exceptional. Remedial andSpecial Education, 18, 82–93.

Jordan, A., Schwartz, E., & McGhie Richmond, D. (2009). Training general educationelementary teachers for inclusive classrooms. Teaching and Teacher Education,25, 535–542.

Jordan, A., & Stanovich, P. (2001). Patterns of teacher–student interaction ininclusive elementary classrooms and correlates with student self-concept.International Journal of Disability, Development and Education, 48(1), 43–62.

Jordan, A., & Stanovich, P. J. (2003). Teachers’ personal epistemological beliefs aboutstudents with disabilities as indicators of effective teaching practices. Journal ofResearch in Special Educational Needs, 3. Available [on-line]. http://www.nasen.org.uk.

Jordan, A., & Stanovich, P. J. (2004). The beliefs and practices of Canadian teachersabout including students with special needs in their regular elementaryclassrooms. Exceptionality Education Canada, 14, 25–46.

Kagan, D. (1992). Implications of research on teacher belief. Educational Psycholo-gist, 27(1), 65–90.

Kelchtermans, G. (2005). Professional commitment beyond contract: Teachers’ self-understanding, vulnerability and reflection. Paper presented at the InternationalStudy Association on Teachers and Teaching (ISATT), Sydney Australia.

Kucan, L., & Beck, I. L. (1997). Thinking aloud and reading comprehension research:inquiry, instruction, and social interaction. Review of Educational Research, 67,271–299.

McGee, M. (2001). Measuring effective teaching in inclusive classrooms. UnpublishedMA thesis, University of Toronto, Toronto, ON.

McGee, M. R. (2004). Teacher and school variables associated with the academicand social outcomes of students with special needs in general educationclassrooms. Unpublished Doctoral Dissertation, OISE/University of Toronto,Toronto, Ontario, Canada.

McGhie-Richmond, D., Underwood, K., & Jordan, A. (2007). Developing effectiveinstructional strategies for teaching in inclusive classrooms. ExceptionalityEducation Canada, 17(1), 27–52.

McLaughlin, M., & Jordan, A. (2005). Push or pull: the forces that shape inclusion inthe U.S.A. and Canada. In P. Mitchell (Ed.), Contextualising inclusive education(pp. 89–113). London: RoutledgeFalmer.

McLaughlin, M. J., & Rouse, M. (2000). Introduction: the struggles for reform. InM. J. McLaughlin, & M. Rouse (Eds.), Educational reform in the United states andBritain (pp. 1–10). London: Routledge.

Munby, H., Russell, T., & Martin, A. (2001). Teacher’s knowledge and how itdevelops. In V. Richardson (Ed.), Handbook of research on teaching (4th ed., pp.877–904). Washington,DC: American Educational Research Association.

Pajares, F. (1992). Teachers’ beliefs and educational research: cleaning up a messyconstruct. Review of Educational Research, 62(3), 307–332.

Salisbury, C. L. (2006). Principals’ perspectives on inclusive elementary schools.Research and Practice for Persons with Severe Disabilities, 31(1), 70–82.

Schommer-Aikins, M. (2002). An evolving theoretical framework for an epistemo-logical belief system. In B. Hofer, & P. Pintrich (Eds.), Personal epistemology: Thepsychology of beliefs about knowledge and knowing (pp. 103–118). Mahwah, NewJersey: Erlbaum Associates, Inc.

Schraw, G., & Olafson, L. (2002). Teachers’ epistemological world views andeducational practices. Issues in Education, 8(2), 99–149.

Schwartz, E. (2008). Elementary classroom teachers’ epistemological beliefs andtheir practices with students with disabilities and at risk of academic failure ininclusive classrooms. Unpublished Ed.D. dissertation, University of Toronto.

Schwartz, E., Glenn, C., & Jordan,A. (2008).Analytic framework for understanding teachers’epistemological beliefs about students with disabilities in elementary inclusive class-rooms. New York, New York: American Educational Research Association.

Scruggs, T. E., & Mastropieri, M. A. (1996). Teacher perceptions of mainstreaming/inclusion, 1958–1995: a research synthesis. Exceptional Children, 63(1), 59–74.

Stanovich, P. J. (1994). Teachers’ sense of efficacy, beliefs about practice andteaching behaviors as predictors of effective inclusion of exceptional andat-risk students. Unpublished Doctoral Dissertation, University of Toronto,Toronto, Ontario.

Stanovich, P. J., & Jordan, A. (1998). Canadian teachers’ and principals’ beliefs aboutinclusive education as predictors of effective teaching in heterogeneous class-rooms. Elementary School Journal, 98, 221–238.

Stanovich, P., & Jordan, A. (2000). Effective teaching as effective intervention.Learning Disabilities: A Multi-disciplinary Journal, 10(4), 235–238.

Stanovich, P. J., & Jordan, A. (2002). Preparing general educators to teach in inclusiveclassrooms: some food for thought. The Teacher Educator, 37, 173–185.

Stanovich, P., & Jordan, A. (2004). Inclusion as professional development. Excep-tionality Education Canada, 14(2–3), 169–188.

Stipek, D. J., Givvin, K. B., Samon, J. M., & McGyvers, V. L. (2001). Teachers’ beliefsand practices related to mathematics instruction. Teaching and Teacher Educa-tion, 17, 213–226.

White, R. (2007) Characteristics of classroom teachers which contribute to theirprofessional growth in implementing inclusive practices. Unpublished M.A.thesis, OISE, University of Toronto.

Yoon, B. (2008). Uninvited guests: the influence of teachers’ roles and pedagogieson the positioning of English language learners in the regular classroom.American Educational Research Journal, 45(2), 495–522.