defining grammatical difficulty a student teacher perspective

8/19/2019 Defining Grammatical Difficulty a Student Teacher Perspective

1/23

Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=rmla20

Download by: [5.100.166.255] Date: 23 February 2016, At: 03:06

Language Awareness

ISSN: 0965-8416 (Print) 1747-7565 (Online) Journal homepage: http://www.tandfonline.com/loi/rmla20

Defining grammatical difficulty: a student teacherperspective

Johan Graus & Peter-Arno Coppen

To cite this article: Johan Graus & Peter-Arno Coppen (2015) Defining grammaticaldifficulty: a student teacher perspective, Language Awareness, 24:2, 101-122, DOI:10.1080/09658416.2014.994639

To link to this article: http://dx.doi.org/10.1080/09658416.2014.994639

Published online: 20 Feb 2015.

Submit your article to this journal

Article views: 443

View related articles

View Crossmark data

http://crossmark.crossref.org/dialog/?doi=10.1080/09658416.2014.994639&domain=pdf&date_stamp=2015-02-20http://crossmark.crossref.org/dialog/?doi=10.1080/09658416.2014.994639&domain=pdf&date_stamp=2015-02-20http://www.tandfonline.com/doi/mlt/10.1080/09658416.2014.994639http://www.tandfonline.com/doi/mlt/10.1080/09658416.2014.994639http://www.tandfonline.com/action/authorSubmission?journalCode=rmla20&page=instructionshttp://www.tandfonline.com/action/authorSubmission?journalCode=rmla20&page=instructionshttp://dx.doi.org/10.1080/09658416.2014.994639http://www.tandfonline.com/action/showCitFormats?doi=10.1080/09658416.2014.994639http://www.tandfonline.com/loi/rmla20http://www.tandfonline.com/action/journalInformation?journalCode=rmla20


2/23

Dening grammatical difculty: a student teacher perspectiveJohan Graus a * and Peter-Arno Coppen b

a Graduate School of Education, HAN University of Applied Sciences, Nijmegen, Netherlands;b Graduate School of Education, Radboud University, Nijmegen, Netherlands

( Received 17 December 2013; accepted 25 November 2014 )

Numerous second language acquisition (SLA) researchers have tried to denegrammatical difculty in second and foreign language acquisition often as part of an attempt to relate the efcacy of different types of instruction to the degree of

difculty of grammatical structures. The resulting proliferation of denitions and thelack of a unifying framework have made the easy difcult distinction a muddled concept. This paper attempts to clear up some of the confusion by reviewing thedenitions SLA literature offers and relating these to (student) teachers’ cognitions ongrammatical difculty, a perspective largely ignored in SLA publications despite its potential for a more integrative and holistic view on the topic. To this end, a total of 727 undergraduate and postgraduate student teachers of English were surveyed in twoempirical studies. Quantitative and qualitative analyses in combination with thendings from SLA literature yielded a model comprising four interrelated categories of factors that determine grammatical difculty: (1) grammar feature; (2) pedagogical arrangement; (3) teacher quality; and (4) learner characteristics. Inaddition, the resulting model is discussed as a basis for future research into thechanging cognitions of student teachers as they become more experienced.

Keywords: teacher cognitions; form-focused pedagogy; SLA; language awareness;EFL; grammar

Introduction

In second language acquisitio n (SLA) research, there is a long tradition examining theeffect and effectiveness of L2 1 instruction (for an overview, see De Graaff & Housen,2009 ). Particularly, form-focused instruction a type of instruction that focuses learners’attention on linguistic form has been the object of investigation (see, for instance, Nor-ris & Ortega, 2000 ). One of the questions central to this line of inquiry is whether theeffectiveness of instruction is related to the degree of difculty of the target structure to be taught. Some researchers claim that only simple structures are amenable to instruction(e.g. DeKeyser, 1995 ; Krashen, 1981 , 1982 ; Reber, 1989 ; Robinson, 1996a , 1996b ), whileothers (tentatively) suggest that instruction is most effective for difcult structures (e.g.De Graaff, 1997 ; Housen, Pierrard, & Van Daele, 2005 ).

One of the problems of studies investigating effectiveness of instru ction in relation tothe degree of difculty of linguistic forms is the denition of difculty. 2 In a meta-analy-sis of 41 studies, Spada and Tomita ( 2010 ) found no clear evidence of interaction betweentype of instruction and the degree of complexity of a linguistic feature, but they readilyadmitted that results might have been different had they used a different set of criteria to

*Corresponding author. Email: [email protected]

2015 Taylor & Francis

Language Awareness , 2015Vol. 24, No. 2, 101 122, http://dx.doi.org/10.1080/09658416.2014.994639

w

mailto:[email protected]://dx.doi.org/10.1080/09658416.2014.994639http://dx.doi.org/10.1080/09658416.2014.994639mailto:[email protected]


3/23

distinguish between simple and complex structures; they encountered no less than eightdifferent denitions of the easy difcult distinction.

The proliferation of denitions of difculty as well as the limited scope of some of these denitions seem to be obstacles in studying the differential effects of type of instruction. A possible solution would be to search for a broader and more integrative perspective to the notion of difculty. In this article, we turn to the language teacher’s perspective, which in traditional SLA literature is often ignored or only marginally repre-sented, to investigate how (student) teacher cognitions can shed more light on the matter.In the majority of instructed settings, teachers are often the link between the subject mat-ter and the learner. Their cognitions what teachers know, think, and believe (Borg,2006 ) are the result of myriad factors, for instance, teacher education, and their experi-ence as (L2) learners and as teachers (Phipps & Borg, 2007 ). These factors make the for-eign language teacher’s perspective a well-suited candidate for empirically examining theeasy difcult distinction, particularly the perspective of those (student) teachers whohave rst had to learn the language they teach themselves as is the case in the presentstudy.

Both authors are currently working on a larger research project investigating thechanging cognitions of student teachers of English as a foreign language (EFL) on form-focused instruction. The present study focuses on the cognitions of both beginning (pre-service) student teachers and advanced (in-service) student teachers, the latter of whomalready have an undergraduate degree in EFL teaching and some experience. The objec-tive of this study then is to examine student teachers’ cognitions about what constitutesgrammatical difculty, and relate them to the notions of difculty as discussed in theSLA literature. We chose to operationalise form-focused instruction as grammar instruc-tion because this is an aspect of EFL teaching that student teachers are fairly familiar

with, and the grammar syllabus in lower secondary school in the Netherlands the con-text of our study is a well-dened one. The following main research question and sub-questions guided our study:

How can grammatical difculty be dened taking into consideration relevant litera-ture as well as the (student) teacher’s perspective?

Which grammatical features of English do EFL (student) teachers perceive as rela-tively difcult and which as relatively simple?

Which reasons do they cite for this distinction? To what extent do pre-service and in-service student teachers differ in their cogni-tions of grammatical difculty?

Literature review

Method

To select literature for the review, we focused on SLA literature that either discussesgrammatical difculty as such or investigates the relationship between type of instruction and grammatical feature. As a starting point, three recent sources wereexamined for specic references to grammatical difculty as a factor inuencingSLA: Spada and Tomita ( 2010 ), Shiu ( 2011 ), and De Graaff and Housen ( 2009 ).Subsequently, the snowballing method was used to identify further relevant sources.Finally, we searched two databases ERIC and Web of Science for relevantsources published between 2000 and 2012.

102 J. Graus and P.-A. Coppen

w


4/23

Review

Many researchers and teachers assume there is a relationship between the relative effective-ness of type of instruction and the degree of complexity of the target feature (De Graaff &Housen, 2009 ). The problem, however, is how to dene complexity. What are the factors

that make a structure difcult or easy to learn/acquire? Here we present the ndings of our literature review, which yielded three broad categories. First, we look at complexity fromthe point of view of the grammar structure itself: its form, use, meaning, and salience. Next, we consider complexity in terms of the pedagogical rules needed to express the lin-guistic feature in question. Lastly, we focus on problematicity, i.e. whether or not learninga grammar point constitutes a problem from a learner’s perspective (Ellis, 2008 ).

Grammatical feature difculty

Salience. Grammatical complexity can be seen as a function of salience. In the narrowsense, salience is equated with the frequency with which a feature arises in the input alearner receives. Bardovi-Harlig ( 1987 ), for instance, discussed salience as the degree towhich data is available to learners. The argument goes that the more frequent a feature is,the less difcult it is to acquire. Later research, too, has found frequency to be a major con-tributor to salience (Collins, Tromovich, White, Cardoso, & Horst, 2009 ; DeKeyser, 2005 ;Goldschneider & DeKeyser, 2005 ; Skehan, 1998 ), the assumption being that the degree towhich learners infer regularities from input is dependent on frequency (Ellis, 2002 , 2003 ).

Salience can, however, also have a broader meaning. It can be viewed as ‘the propertyof a structure that is perceptually distinct from its environment’ (Ravid, 1995 , p. 117).Salience in this sense is related to the notions of attention and noticing (Long, 2007 ; Lyster,2007 ; Schmidt, 1990 , 1994 , 2001 ; Swain, 2005 ). It is presumed that for processing and sub-

sequent acquisition to take place, learners have to notice a feature and pay attention to it.Structures that are perceptually salient would facilitate such attentional allocation.

An even broader conceptualisation of salience can be found in Goldschneider and DeKeyser’s ( 2005 ) meta-analysis, in which they have attempted to explain the naturalorder often found in English L2 morpheme acquisition. They examined production datafrom 12 studies, investigating whether a combination of ve predictors could account for the order of acquisition of six grammatical functors (progressive -ing, plural -s, posses-sive -s, articles, third person singular -s, and regular past -ed). The ve predictors theydiscussed were:

(1) perceptual salience , which they operationalised as comprising three factors: thenumber of phones in a functor (phonetic substance), the presence/absence of avowel in the surface form (syllabicity), and the total relative sonority of a functor;

(2) semantic complexity , referring to the number of meanings expressed by a functor;(3) morpho-phonological regularity , i.e. the degree to which functors are affected by

their phonological context (e.g. allomorphy, contractibility, and redundancy);they also considered the number of phonological alternations a functor has and homophony with other grammatical functors;

(4) syntactic category , choosing functional category theory as a framework, Gold-scheider and DeKeyser divided grammatical functors into syntactic categories(lexical or functional) and they made a distinction between bound and freemorphemes;

(5) frequency .

Language Awareness 103

w


5/23

They found that these ve predictors explained a signicant portion of the variance inaccuracy scores ( R D .84, R2 D .71, p < .001). Concluding that the ve predictors do notmeasure completely different underlying constructs, they argued that they ‘all constituteaspects of salience in a broad sense of the word’ (p. 60).

Dening complexity in terms of salience, however, is not without problems. Saying thata feature is complex because it lacks salience can be considered a circular argument(VanPatten, 2007 ). Moreover, we still do not know what it is precisely that makes a formmore or less salient and what the role of a learner’s L1 is in determining salience (Shiu,2011 ).

Form, function, and meaning. Grammatical complexity can also be related to the form,function, and meaning of a grammar feature (or a combination thereof). Referring to lin-guistic form, Hulstijn and De Graaff ( 1994 ), for instance, dened complexity as contin-gent on ‘the number (and/or type) of criteria to be applied in order to arrive at the correctform’ (p. 103). In a meta-analysis of 41 studies investigating the differential effects of explicit and implicit instruction on the acquisition of complex and simple English gram-mar features, Spada and Tomita ( 2010 ) operationalised Hulstijn and De Graaff’s ( 1994 )criteria for determining complexity. Structures requiring two transformations or morewere considered to be complex, as a result of which they characterised as simple featuresin English: tense, articles, plurals, prepositions, subject verb inversion, possessive deter-miners, and participial adjectives.

Researchers have also made a distinction between functional and formal complexity.DeKeyser ( 1998 ) denes a functionally complex structure as requiring complicated mental processing operations, whereas formal complexity refers to the relationship between func-tion and form. Operationalising these notions in practical terms, however, is problematic.

A case in point is the third person singular -s, which Krashen ( 1982 ) classied as function-ally and formally simple, whereas Ellis ( 1990 ) argued it was formally complex because of the long-distance relationship with the grammatical number of the subject. Both Ellis and Krashen considered the third person singular -s functionally easy, but it can also be argued that it is functionally complex since a one-letter morpheme has to express simultaneouslythe notions of present tense, singular, and third person (DeKeyser, 1998 ).

In a later article, DeKeyser ( 2005 ) identied three factors that determine grammaticalcomplexity: complexity of form, complexity of meaning, and complexity of theform meaning relationship. He denes complexity of form as ‘the number of choicesinvolved in picking all the right morphemes and allomorphs . . . and putting them in theright place’ (pp. 5 6). Complexity of meaning may be a source of difculty because of what DeKeyser calls novelty or abstractness (or both). Articles, classiers, grammaticalgender, and verbal aspect are all examples of structures that are difcult to acquire for L2learners whose L1 does not have these or uses different systems (see, for instance, Jarvis,2002 ; Monstrul & Slabakova, 2003 ; Taraban, 2004 ), even in an instructional context (see,for instance, Ayoun, 2004 ; Butler, 2002 ; Leeman, 2003 ). Lastly, when the relationship between forms and meaning is not transparent, difculty can arise in form meaning map- ping, for instance because of redundancy (e.g. third person singular -s in English) or optionality (e.g. null subjects in Spanish).

Pedagogical rule difcultyQuantitative aspects. Housen et al. ( 2005 ) dene pedagogical rule complexity as related to ‘the degree of elaboration with which the rule is formulated, i.e. as the number of steps


w


6/23

the learner has to follow to arrive at the production of the intended linguistic structure,and the number of options and alternatives available at each step’ (p. 241). Some linguis-tic structures require a more elaborate rule than others, but one and the same structure canalso be described using a simple or a more complex rule. Dietz ( 2002 ) proposes a frame-work for determining rule complexity that is similar to Housen et al.’s ( 2005 ). Accordingto Dietz, three factors govern rule complexity: (1) the number of criteria a rule consistsof, (2) the number of subrules a domain requires, and (3) the number of conditions in theconditional part of a rule. For instance, the German Perfekt has two subrules, dependingon whether it is formed using the auxiliaries haben or sein . For both subrules, there areconditional rules that specify which type of verbs belongs to which auxiliary.

Conceptual clarity and metalanguage. Ellis ( 2009 ) takes another approach to exploring pedagogical rule difculty. In his view, rule difculty ‘needs to be understood in terms of how easy or difcult it is to verbalize a declarative rule’ (p. 148). This, in turn, depends

on two concepts: conceptual clarity and metalanguage. With regard to conceptual clarity,Ellis considers rules to be easy if they describe linguistic structures that are both formallyand functionally simple (Krashen, 1982 ), since these can be easily expressed usingHammerly’s ( 1982 ) formulation criteria (i.e. rules should be simple, concrete, non-techni-cal, close to traditional/popular notions, and in the form of rules-of-thumb). With refer-ence to metalanguage, Ellis ( 2009 ) considers those rules to be relatively complex thatrequire more ‘technical’ metalanguage (James & Garrett, 1992 ). In some cases, use of technical metalanguage can be avoided to a certain extent. So instead of saying that theindenite article in English is not used with uncountable nouns, one could also say ‘Don’t“use a/an” before a word that cannot be made plural’ (Ellis, 2009 , p. 150).

Scope and reliability. Finally, Hulstijn and De Graaff ( 1994 ) consider a rule’s probabi-listic nature its scope and reliability as a factor that can interact with rule complexity.A rule may be large or small in scope referring to the number of cases it covers. Further, arule can be said to be high or low in reliability, indicating whether there are many excep-tions. Although grammatical complexity on the one hand and scope and reliability on theother are conceptually different, Hulstijn and De Graaff do concede that ‘in the practiceof language pedagogy [they] often go hand in hand’ (p. 104).

The learner’s perspective

The learner’s L1. The role of the L1 in learning and acquiring a second language has been discussed and debated over and again in SLA research. Back in 1957, Lado pre-sented a straightforward view on the interaction between L1 and L2: the contrastive anal-ysis hypothesis. The idea was that learners would make the most mistakes in those areaswhere the L1 and the L2 differed. Where L1 and L2 were the same, few learning difcul-ties were to be expected. We now know that the role of the L1 through positive and nega-tive transfer is anything but straightforward. Cross-linguistic inuence no longer seenfrom its original behaviourist perspective is now considered a complex and multiface-ted phenomenon, and generalisations based on ndings from particular sets of languagesmay not hold for other languages (see, for instance, Odlin, 1989 , 2003 ).

Taking a different universal grammar perspective on the role of L1, White ( 1991 )argued that certain parametric differences between languages can cause learnability prob-lems for L2 learners, since ‘L2 learners use L1 settings of UG parameters as an interim


w


7/23

theory about the L2’ (p. 137). White tested 138 French-speaking learners of English and indeed found support for the hypothesis that learners used the L1 parameter settings.

No matter the perspective taken, L2 acquisition does seem to interact with thelearners’ L1. Theoretical accounts as well as empirical studies (see, for instance,Izquierdo & Collins, 2008 ; Luk & Shirai, 2009 ) have shown that L2 learning may becurbed by L1 transfer. Framing grammatical complexity as partly contingent on L1 inu-ence is therefore a tenable position. The intricacies of the manner in which the L1 pre-cisely affects L2 acquisition, however, make it difcult to make predictions and generalisations.

Learner characteristics. Ultimately, whether a particular grammar point is difcult or easy is relative to the learners’ prociency level and developmental stage (Ellis, 2008 ).So in order to ascertain complexity, a researcher must have some knowledge of thelearners’ stage of development and what is difcult or easy at that particular stage. SLAliterature has a rich tradition in studies that have examined developmental stages learnersgo through (for overviews of grammatical morpheme studies for English, seeGoldschneider & DeKeyser, 2001 ; Larsen-Freeman & Long, 1991 ). As Shiu ( 2011 ) points out, in the case of L2 English several researchers have found evidence for a pre-dictable development in the acquisition of negation (e.g. Dulay, Burt, & Krashen, 1982 ), pronouns (e.g. Zobl, 1985 ), relativisation (e.g. Pavesi, 1986 ), possessive determiners (e.g.White, 1998 ; White, Munoz, & Collins, 2007 ), and tense and aspect (e.g. Bardovi-Harlig,2000 ; Shirai, 2004 ).

Saying that a structure is acquired late because it is difcult, however, is a circular argument. It does not explain why some structures are acquired later than others. Piene-mann and colleagues (e.g. Pienemann, 1989 ; Pienemann, Johnson, & Brindley, 1988 ),

investigating a number of syntactic and morphological features in English, took a psycho-linguistic approach in accounting for the early late conundrum. Pienemann ( 1989 , 2003 ) proposed processability theory (PT), which posits that psycholinguistic processing abili-ties and constraints are responsible for the predictable order in which learners acquire cer-tain structures. Underlying PT is the idea that a learner can only comprehend and producethose L2 features that the language processor at that point in time can handle.

Apart from prociency level and developmental stage, there is one other importantlearner characteristic that may play a role in why some structures are perceived as moredifcult than others: aptitude. The concept was originally operationalised by Carroll(Carroll & Sapon, 1959 ), who devised the Modern Language Aptitude Test (MLAT),which consists of ve parts that measure four underlying aptitude factors: phonemic cod-ing ability, grammatical sensitivity, inductive language learning ability, and associativememory. The MLAT, however, has been criticised for its validity (see, for instance,Sawyer & Ranta, 2001 ), as a result of which several researchers have proposed revised versions (e.g. Robinson, 2005 ; Skehan, 2002 ). Nevertheless, aptitude remains a conceptthat is difcult to operationalise.

Exploring difculty empirically

As shown in Figure 1 , grammatical difculty as discussed in SLA literature is a combination of myriad factors such as formal and functional complexity and pedagogical ruledifculty, both of which interact with learner characteristics. The multifaceted nature of grammatical difculty makes it hard for researchers to include all possible variables gov-erning difculty in empirical studies. It is therefore quite understandable that over the


w


8/23

years several researchers turned to studying language users to operationalise the conceptof grammatical difculty (Bialystok, 1979 ; Green & Hecht, 1992 ; Robinson, 1996b ).

Schefer ( 2011 ) is a recent example of empirically exploring grammatical dif-culty. He relied on teacher intuitions to assess the concept. As an explanation for hischoice, he cites Ellis ( 2006 ) in saying that such an empiric approach may be inevita- ble since ‘it may prove impossible to arrive at [objective linguistic] criteria that willensure a reliable and valid assessment’ (Ellis, 2006 , p. 439). Schefer (2011 ) asked 20 Polish teachers of English to rank 12 structures on a one-to-ve scale measuringgrammatical complexity. Then he asked a group of 50 Polish learners of English(upper secondary school) to write sample sentences using the correct forms in a cor-rect context. He found a strong association between the teachers’ predictions and thelearners’ scores ( r s D ¡ .9, p < .01).

Figure 1. Grammatical difculty in SLA literature.


w


9/23

Although SLA literature already offers a comprehensive discussion of grammaticaldifculty, what is lacking is the (student) teacher’s perspective. In the following sections,we discuss two empirical studies that explore how (student) teachers view the concept of grammatical difculty.

Method

In order to answer the research questions, we undertook two empirical studies: a pilotstudy and a main study. The structure of the study is summarised in Figure 2 .

Participants

The participants in the two empirical studies were selected from two groups: undergradu-ate and postgraduate student teachers of English matriculated at seven different Dutchuniversities of applied sciences. The undergraduate students were all enrolled in four-year full-time programmes that lead to a bachelor’s degree in education and a so-called Grade II certication, allowing them to teach English in lower secondary school. The bachelor degree programme is a pre-service teacher education course in which subjectknowledge, subject methodology, pedagogy, and teaching practice are combined.

The postgraduate students were enrolled in three-year work study programmes uponcompletion of which they are awarded a master’s degree in education and a Grade I certi-cation, which is required to teach upper secondary school in the Netherlands. Unlike the bachelor programme, the master’s course is an in-service track for teachers who alreadyhave a Grade II certication in English and who in most cases already have considerable

experience. Further details about the participants can be found in Table 1 .

Figure 2. Method.


w


10/23

Pilot study

To get a better understanding of what factors make English grammar difcult from the

perspective of student teachers, we surveyed 157 undergraduate and postgraduate stu-dents. They were asked to cite as many reasons as possible why English grammar in thecontext of lower secondary school can be difcult. The choice of focusing on lower sec-ondary school was a deliberate one, since the undergraduate students were being trained to become teachers of English in lower secondary and the postgraduate students had atleast two years of experience in that sector. Additionally, in Dutch lower secondaryschool most if not all basic elements of English grammar are taught.

The responses were analysed qualitatively through a process of open and axial coding.Additionally, we performed several quantitative tests. First, we calculated the frequencyof responses, which resulted in ranked lists. Next, to nd out whether undergraduate and postgraduate students differed signicantly in how frequent a response was given, wedivided the participants into three subgroups year 1 and 2 undergraduates, year 3 and 4undergraduates, and postgraduates and performed a Kruskal Wallis one-way analysisof variance (ANOVA) (and step-down follow-up analyses where appropriate). Finally,the Jonckheere Terpstra test 3 was used to check for signicant trends in the data.

Main study

For the main study we devised a questionnaire to determine how the participants esti-mated the difculty of a number of grammar points and which factors contributed to the perceived level of difculty. To select appropriate grammatical structures, we analysed the grammar points dealt with in the three best-selling course book series for English inDutch lower secondary school (personal communication, E. de Wijs, April 24, 2013) atotal of 21 textbooks. Subsequently, three topics were chosen (tense & aspect, word order,determiners & quantiers) based on the following criteria: (1) the topics comprise at leastsix individual grammar points; (2) they are taught at all three main levels of Dutch sec-ondary education; and (3) they are taught in each of the three years of lower secondaryschool. Table 2 presents an overview of the selected grammar points (31 in total).

In the rst part of the questionnaire, respondents were asked to judge the difculty of the 31 grammar points that we selected in relation to the learners’ level. This level wasspecied using two dimensions: (1) level of education and (2) form (year 1, 2, or 3).Dutch secondary education consists of three main levels: VMBO or pre-vocational sec-ondary education, HAVO or senior general secondary education, and VWO or pre-univer-sity education. So for each grammar point, respondents had to decide whether they would

Table 1. Descriptive statistics of participants.

Pilot study Main study

n Experience a n Age b Experience a

Undergraduate students Year 1 2 75 303 21.57Year 3 4 49 180 26.73

Postgraduate students 33 12.79 85 38.98 10.66Total N 157 570

a Mean number of years of experience. b Mean age in years.


w


11/23

teach it at VMBO, HAVO, and/or VWO, and in which year. For instance, the followingscore

future continuous VMBO: 4 j HAVO: 3 j VWO: 2

means that a respondent would teach the future continuous (for the rst time) in the sec-ond year of VWO and in the third year of HAVO; a score of 4 means that the respondentwould not teach the structure in question in the rst three years of that type of education(in the case of the example, he or she would not teach the future continuous in the rstthree years of VMBO).

For the purpose of facilitating comparison, the respondents’ answers were treated asscores, which were averaged and subsequently ranked in order to determine when studentteachers would teach each item (for the rst time) and at which level. The same statisticaltests as in the pilot study were performed to check for signicant differences between sub-groups. Finally, a Spearman’s rank order correlation matrix was calculated to determinethe degree of similarity between the rankings of the subgroups.

The second part of the questionnaire focused on the perceived causes of grammati-cal difculty. Based on an additional, small-scale pilot of 13 experienced teachers of English (mean number of years of experience: 10.38; range: 5 34 years), who wereasked to rate the difculty of the 31 grammar points that feature in the rst part of thequestionnaire, we chose nine grammar points (three per topic) that were perceived asdifcult (see the structures marked with an a in Table 2 ). For each of these grammar points the respondents in the main study were asked to indicate to what extent theyconsidered the following factors to contribute to the difculty of the grammar featurein question: (1) complexity of form; (2) complexity of use; (3) complexity of peda-gogical rule; (4) inuence of L1; and (5) frequency of input. We selected these vefactors from the literature review and the results of the pilot study because we consid-ered them to have a possible differential effect on individual grammar points rather than inuence grammar learning in general. For instance, aptitude is more likely to

Table 2. Grammar points included in main study.

Tense & aspect Word order Determiners & quantiers

Present simple Transformations Demonstrative determinersPresent continuous Negations (do support) Interrogative determinersPast simple Questions (do support) a Possessive determinersPast continuous Negations (auxiliaries) ; denite article C school/church etc. a

Present perfect a Questions (auxiliaries) a Indenite article C professionPresent perfect continuous a Ordinal numeralsPast perfect a Basic word order Quantifying phrasesFuture: will/shallFuture: to be going to

Adverbials of placeand/or time

Some/anyLittle/few a

Future: present simple Adverbs of frequency a Many/muchFuture: present continuous Genitive a

Future continuous

Note: The above grammar points were included in the rst part of the main study. Only the points marked withan a also featured in the second part.


w


12/23

affect a learner’s ability to learn grammar across the board, whereas the inuence of complexity of use is likely to be contingent on the grammar point in question.

A ve-point Likert scale was used to measure agreement with statements expressingthe ve factors mentioned above. In order to facilitate comparison, average scores werecalculated for each factor, and a Friedman’s ANOVA was used to search for signicantdifferences within each grammatical structure. Whenever such a difference occurred, astep-down follow-up analysis was performed to create homogenous subsets.

Results

Pilot study

In the pilot study, 157 undergraduate and postgraduate students gave a total of 515answers to a question about what makes English grammar difcult in the context of lower secondary school. Categorising these answers, qualitative analysis yielded 31 differentcodes. Upon further examination, common topics were identied, enabling us to groupthe codes into 13 categories, which in turn were clustered into four major themes. Table 3 presents the original codes and the result of subsequent axial coding.

The rst theme, grammatical feature, comprises complexity of form and use, the number of forms and structures learners have to master and can choose from, and the qualityof the input learners are confronted with outside the classroom, which according to therespondents often contains instances of incorrect English, slang, and colloquialisms. Thenext theme, pedagogical arrangement, is a combination of rule complexity, practice(insufcient practice and recap opportunities, too few transfer exercises, and lack of timeto practise), and method, which consists of codes such as teaching methods that are

criticised for lacking a communicative or meaningful context, being too teacher-focused,a lack of testing or conversely an excessive focus on testing, and the quality of the materi-als presented in textbooks. The next theme concerns the quality of the teachers them-selves, who sometimes do not have an adequate grasp of the subject matter, resulting inunclear instructions to learners. The nal theme focuses on learner characteristics, suchas the L1 of the learner, their motivation, level, aptitude, and the number of years of instruction they had.

In quantitative terms, the primary reason why grammar is difcult according to therespondents was the learners’ rst language: 61.2% mentioned the differences betweenEnglish and Dutch as one of the causes underlying many errors. The number of excep-tions came in second (35%), and motivation (22.3%) in third place. Table 4 presents anoverview of the top 15 reasons that were cited by the respondents.

Further analysis comparing undergraduate students and postgraduate studentsrevealed that the latter group considered grammatical difculty mainly in terms of pedagogical arrangement assuming that the number of responses that constitute a particular theme is an indication of relative importance. The third- and fourth-year undergraduates also emphasised pedagogical arrangement as the most important fac-tor. Learner characteristics were considered the most important by the rst- and sec-ond-year undergraduates, whereas this theme occupied the second place in the other two groups. See Table 5 for a complete overview of the mean number of responses per group.

The differences between the groups were only signicant for the teacher theme ( a D.05), H (2) D 10.08, p D .006. Step-down follow-up analysis showed that the postgraduateand third- and fourth-year undergraduate students clustered together and differed from


w


13/23

the year 1 and year 2 undergraduate group. The differences for pedagogical arrangementwere almost signicant, H (2) D 5.909, p D .052.

To detect trends in the data, the Jonckheere Terpstra test was carried out. It showed

signicant trends for the themes pedagogical arrangement and teacher. For pedagogicalarrangement, membership of a more advanced group meant on average more answers for this theme, J D 4507.500, z D 2.147, p D .032, r D .17. Conversely, membership of a

Table 3. Qualitative analysis of reasons why student teachers think English grammar can bedifcult ( N D 157).

Theme Category Associated codes

Grammatical feature Complexity of form Complexity of formComplexity of use Complexity of use

Complexity of choice Number of structures Large number of structures to be mastered Quality of input Poor quality of input

Pedagogical arrangement Rule complexity Complexity of rules a

AbstractnessLarge number of exceptionsMetalanguage

Practice Poor quality of practice b

Too few recap possibilitiesToo few transfer exercisesLack of time

Method Teaching method c

Not student-oriented Lack of communicative/meaningful contextPlace of grammar in the classroomTarget language as language of instructionToo much or not enough testingPoor quality of textbook

Teacher Teacher quality Unsatisfactory teacher quality

Learner L1 L1L1 is not Dutch

Motivation Low motivationLow relevance

Level LevelDevelopmental stage

Aptitude AptitudeProblem of intuitionMemory

Experience Time of rst instruction

abc These codes may seem redundant since they summarise the codes immediately below. Nevertheless we haveincluded them because a number of respondents directly mentioned them without further specication. For instance, some respondents named ‘complexity of rules’ as a reason for grammatical difculty without referringto ‘abstractness’, ‘large numbers of exceptions’, or ‘metalanguage’.


w


14/23

more advanced group meant on average fewer answers for the teacher theme, J D 3543.5, z D ¡ 3.079, p D .002, r D ¡ .25.

Main study

Table 6 presents the results of the rst part of the main survey. We only include the scoresfor HAVO (senior general secondary education) because each grammar topic showed thesame pattern: VMBO (pre-vocational secondary education) scores were highest, VWO(pre-university education) scores were lowest, and HAVO scores were in between. Theoverall scores indicate that the ve most difcult structures were: (1) present perfect con-tinuous; (2) future continuous; (3) past perfect; (4) ; denite article C school, church,etc.; and (5) present perfect. The ve easiest grammar topics according to the respondentswere: (1) possessive determiners; (2) present simple; (3) past simple; (4) negations (dosupport); and (5) ordinals. Comparing subgroups, two distinct patterns emerged. In mostcases that yielded signicant differences, the more advanced the subgroup the lower thescores. In some cases, however, the pattern was reversed most notably in those tensesthat fall on the difcult end of the spectrum.

As Table 7 shows, the rank order correlations among the subgroups were very high

with values between r s D .899 and r s D .948, and statistically signicant ( p < .001).

Table 4. Top 15 of reasons why student teachers think English grammar can be difcult ( N D157).

Reason Percentage a Reason Percentage a

L1 61.2 Memory 9.6Large number of exceptions 35.0 Teaching method 9.6Low motivation 22.3 Complexity of rules 8.9Large number of structures 16.6 Low relevance 8.3Too few transfer exercises b 12.1 Unsatisfactory teacher quality 7.6Complexity of use 10.8 Poor quality of practice 7.6Abstractness 10.2 Poor quality of textbook 7.6Metalanguage 10.2

a Percentage of respondents who mentioned the reason in question. b Transfer exercises require learners to apply a grammar topic that they have practised with in controlled exer-cises (e.g. drills, gap-lls) to free-production exercises.

Table 5. Mean number of answers per theme.

UndergraduatesY1 Y2 ( n D 75)

UndergraduatesY3 Y4 ( n D 49)

Postgraduates(n D 33)

M M M

Grammatical feature 0.37 0.29 0.30Pedagogical arrangement 1.17 1.18 1.73Teacher 0.15 0.02 0Learner 1.27 1.04 1.21

H (2) D 1.711, p D .425; H (2) D 5.909, p D .052; H (2) D 10.08, p D .006; H (2) D 2.277, p D .320.


w


15/23

In the second part of the questionnaire, respondents were asked to indicate the per-ceived causes of difculty on a ve-point Likert scale for nine grammatical features.

Table 8 reports the results. In the tenses category, difculty of use had the highest scores;low frequency came in last. The word order category showed a different picture: adverbsof frequency and questions with do support scored highest on L1, whereas questions with

Table 6. Degree of difculty ( N D 570).

HAVO scores

OverallUndergraduates

Y1 Y2Undergraduates

Y3 Y4 Postgraduates

Possessive determiners 1.17 (1) 1.23 b (1) 1.14 b (2) 1.06 a (1)Present simple 1.17 (2) 1.25 b (3) 1.11 a (1) 1.08 a (2)Past simple 1.21 (3) 1.25 (2) 1.20 (3) 1.15 (5) Negations (do support) 1.23 (4) 1.26 (4) 1.21 (4) 1.15 (6)Ordinals 1.23 (5) 1.28 (5) 1.22 (6) 1.13 (4)Questions (do support) 1.30 (6) 1.39 b (6) 1.24 a (7) 1.17 a (8)Demonstrative determiners 1.33 (7) 1.47 b (9) 1.21 a (5) 1.16 a (7)Present continuous 1.33 (8) 1.43 c (7) 1.31 b (8) 1.10 a (3)Indenite article C profession 1.42 (9) 1.50 (11) 1.32 (9) 1.42 (12)

Interrogative determiners 1.53 (10) 1.46 (8) 1.56 (15) 1.67 (19)Adverbials of time 1.53 (11) 1.67 b (13) 1.41 a (10) 1.36 a (10)Many/much 1.56 (12) 1.72 b (15) 1.42 b (11) 1.36 a (9)Past continuous 1.56 (13) 1.48 a (10) 1.63 b (17) 1.67 b (20)Adverbials of place 1.59 (14) 1.68 b (14) 1.55 ab (13) 1.38 a (11) Negations (auxiliaries) 1.65 (15) 1.66 (12) 1.66 (18) 1.60 (16)Adverbials of place and time 1.67 (16) 1.82 b (17) 1.53 a (12) 1.53 a (15)Some/any 1.68 (17) 1.83 b (18) 1.56 a (14) 1.51 a (14)Quantifying phrases 1.74 (18) 1.77 (16) 1.75 (19) 1.65 (18)Genitive 1.78 (19) 1.94 b (21) 1.62 a (16) 1.62 a (17)Questions (auxiliaries) 1.81 (20) 1.90 b (20) 1.85 b (23) 1.43 a (13)Future: present simple 1.86 (21) 1.86 (19) 1.83 (21) 1.92 (24)Little/few 1.95 (22) 2.07 b (23) 1.77 a (20) 1.95 ab (25)Adverbs of frequency 1.97 (23) 2.13 b (25) 1.85 a (22) 1.76 a (21)Future: present continuous 1.98 (24) 1.96 (22) 2.03 (24) 1.95 (26)Future: to be going to 2.04 (25) 2.10 (24) 2.07 (25) 1.84 (22)Future: will/shall 2.12 (26) 2.22 b (27) 2.09 b (26) 1.85 a (23)Present perfect 2.16 (27) 2.19 (26) 2.11 (27) 2.18 (27); denite article C school etc. 2.29 (28) 2.32 (28) 2.20 (28) 2.36 (28)Past perfect 2.49 (29) 2.35 a (29) 2.55 b (29) 2.77 c (29)Future continuous 2.65 (30) 2.52 a (30) 2.72 b (30) 2.92 b (30)

Present perfect continuous 2.76 (31) 2.65a

(31) 2.80a

(31) 3.01 b

(31)abc Refer to homogenous subsets ( p < .05). For instance, for the adverbials of place, the postgraduates’ score a dif-fers signicantly from the undergraduates year 1 2 b ; the score of the undergraduates year 3 4ab does not differ signicantly from either the postgraduates or the undergraduates year 1 2.Between-group differences are statistically signicant at p < .05 (Kruskal Wallis ANOVA).

Note: The numbers between parentheses report the rankings. The scores are the arithmetic means calculated from the four answer categories respondents could choose from (1 D teach in year 1; 2 D teach in year 2; 3 Dteach in year 3; 4 D do not teach in lower secondary).


w


16/23

auxiliaries other than do scored highest on difculty of use. The nal category, determin-ers & quantiers, was more consistent again. Differences with the L1 were seen as themost important cause of difculty; either form or frequency had the lowest score.

DiscussionThe SLA literature offers various perspectives on grammatical difculty and complexity.Some researchers have focused on linguistic criteria such as formal and functional complexity and salience to dene difculty. Others emphasise that characteristics of pedagog-ical rules use of metalanguage and conceptual clarity, for instance play an importantrole in why some structures are considered difcult and others easy. Still other researchersturn to the learner to explain this difference; in this view, factors such as the learners’ L1,their developmental stage, and aptitude are the main predictors of grammatical difculty.In reality, however, it is probably fair to say that none of the three main categories our lit-erature review yielded are on their own responsible for the phenomenon in question.

Grammar feature difculty, pedagogical rule difculty, and learner characteristics arelikely to interact in a multiplicity of complex ways. A grammar feature that is formallycomplex for instance, because a relatively large number of criteria need to be applied to produce a correct form (Hulstijn & De Graaff, 1994 ) is likely to require a complex pedagogical rule, which in turn will increase the cognitive load it exerts on a learner,

Table 7. Correlation matrix (Spearman’s correlation coefcient).

Undergraduates Y3 Y4 Postgraduates

Undergraduates Y1 Y2 .939 [.846 .976] .899 [.778 .953]Undergraduates Y3 Y4 .948 [.869 .979]

p < .001 (two-tailed). Note: Bias corrected and accelerated bootstrap 95% condence intervals are reported in brackets.

Table 8. Perceived causes of difculty ( N D 570).

Difculty

Use Form Rule L1 Frequency

Present perfect continuous 3.89 a 3.71 bc 3.73 bc 3.66 b 3.07 c

Past perfect 3.56 a 3.11 b 3.45 a 3.14 b 2.71 c

Present perfect 3.44 a 2.98 b 3.33 a 3.43 a 2.30 c

Adverbs of frequency 2.55 b 2.43 b 2.61 b 2.82 a 2.18 c

Questions (do support) 2.57 b 2.55 b 2.49 b 3.22 a 1.86 c

Questions (auxiliaries) 2.90 a 2.84 a 2.88 a 2.76 a 2.21 b

; denite article C school etc. 2.94 c 2.49 d 2.92 c 3.80 a 3.17 b

Little/few 2.91 b 2.51 c 2.72 c 3.21 a 2.54 c

Genitive 2.68 b 2.64 b 2.64 b 3.19 a 2.29 c

abc

Refer to homogenous subsets ( p < .05). For instance, for the present perfect continuous, difculty of usea

dif-fers signicantly from all other causes; L1 b differs signicantly from use a and frequency c ; form bc and rule bc onlydiffer signicantly from use a .Within-subject differences are statistically signicant at p < .001 (Friedman’s ANOVA).


w


17/23


18/23

of students’ growing professional condence and pride, or simply a shift in perspective more experienced student teachers no longer see themselves as learners who can talk about ineffective teachers without indirectly referring to themselves.

Moreover, if (unsatisfactory) teacher quality was mentioned, respondents linked it inalmost all cases to knowledge of grammatical features and not to the pedagogical arrange-ment category. In terms of our model then, the right subtriangle in particular the sideconnecting grammar feature and teacher quality seems be an area of focus in earlyteacher education, whereas it may be less important or even altogether absent at later stages. It is quite likely that undergraduates were actually referring to their own grammat-ical development when they raised the matter of teacher quality and grammar knowledge.Further qualitative analysis of the teacher category supports such a conclusion. Under-graduates indicated that for a teacher to be able to explain correctly a grammar structure,he or she needs to have a full understanding of it, which according to them is far from

Figure 3. Grammatical difculty in instructed settings.


w


19/23

obvious. As student teachers become more advanced, their insecurity regarding grammat-ical knowledge dwindles and nally disappears entirely.

Whereas the focus on teacher quality decreased as student teachers became moreexperienced, the opposite happened for the pedagogical arrangement category, whichunderscores student teachers’ development towards condent practitioners, and conrmsthe central position the latter category occupies in our model. It may also explain whyundergraduate students hardly ever link teacher quality with the pedagogical arrangementcategory. They are still struggling to come to grips with their own understanding of English grammar, as a result of which they are not yet fully focused on their future role aseducators, designing and implementing pedagogical arrangements.

Further evidence for changing cognitions in undergraduate and postgraduate studentswas found in their judgement of the degree of difculty of individual grammar features.In most cases rst- and second-year undergraduates considered the degree of difculty to be relatively high compared to their third- and fourth-year counterparts, who in turnscored higher than postgraduate students. These differences may be construed as students’ becoming more comfortable with English grammar as they progress through teacher edu-cation. Conversely, it may be interpreted as an undesirable effect: as students becomemore advanced they forfeit their learner perspective and underestimate the difculty of grammatical topics.

In a few cases, however, this pattern was reversed, notably for the tenses towards thedifcult end of the spectrum. Postgraduates judged the past perfect, for example, to bemore difcult than third- and fourth-year undergraduates did, whose score was still higher than rst- and second-year undergraduates. The subtleties of the English tense and aspectsystem are notoriously difcult for foreign language learners (Housen, 2002 ), which may provide an explanation for the reversed pattern. It is entirely conceivable that less advanced

student teachers who to an extent are still struggling with English grammar underesti-mate the challenging nature of these tenses, whose use and accompanying complex peda-gogical rules are considered major contributors to their overall difculty.

Notwithstanding the differences between groups regarding individual grammaticalstructures, undergraduates and postgraduates showed remarkable similarities in the man-ner in which they ranked the 31 grammatical structures that were included in our study.Postgraduate students, who on average had more than 10 years of teaching experience,and undergraduates, who are in the process of transforming from learners to teachers of English, were similar in their intuitions about the relative difculty of grammatical fea-tures. These ndings are in line with earlier research by Schefer ( 2011 ), who compared teacher intuitions about difculty and student errors. He, too, found a strong correlation,from which we may infer that learners’ and teachers’ intuitions are strong predictors for learners’ performance.

Conclusion

In this study we have examined teacher cognitions about grammatical difculty in order to investigate whether such cognitions can contribute to what is already known about thetopic from SLA literature. The result is a triangular model of grammatical difculty thatencompasses several dimensions, represented as subtriangles. The left subtriangle corre-sponds to what is found in most SLA literature: a focus on mostly grammar-item speciccauses of difculty, which emphasise linguistic or rule complexity interacting with a lim-ited scope of learner characteristics such as L1 and developmental stage. The right sub-triangle represents the main focus of beginning or inexperienced (student) teachers, who


w


20/23

are still struggling themselves to grasp the intricacies of the language. Finally, the bottomtriangle focuses on how more experienced teachers view grammatical difculty: as theinterplay between teachers and learners with a strong emphasis on the role of pedagogicalarrangements. The model in its entirety will serve as a framework guiding our futureresearch into the beliefs on form-focused instruction that (student) teachers hold during dif-ferent stages in their educational and professional lives. Specically, it will be our starting point for exploring whether the relationship between grammatical difculty and type of instruction as studied by numerous SLA researchers is also reected in teacher cognitions.

Acknowledgements

We wish to thank Gerda Geerdink for her contributions. Also, we owe our gratitude to two anony-mous reviewers, whose comments helped us improve this article.

Notes1. The 2 in L2 is used here to refer to any other language than the rst.2. In some cases, researchers use the terms difculty and complexity interchangeably. In this study,

however, difculty is used as a holistic concept that includes more specic notions such as complexity of form, complexity of use, rule complexity, etc.

3. In essence, the Jonckheere Terpstra test is the same as the Kruskal Wallis test. The onlydifference is that it includes information about whether there is a meaningful order of medians(Field, 2013 ; Jonckheere, 1954 ; Terpstra, 1952 ).

Notes on contributors

Johan Graus is a senior lecturer in linguistics and second language acquisition, and he is also thecoordinator for the MEd programme Teacher Education in English. His research interests areteacher cognitions and grammatical awareness.

Peter-Arno Coppen is a professor of linguistics with a special interest in formal linguistics, seman-tics, computational linguistics, and grammar education.

ReferencesAyoun, D. (2004). The effectiveness of written recasts in the second language acquisition of aspec-

tual distinctions in French: A follow-up study. Modern Language Journal, 88 , 31 55. doi:10.1111/j.0026-7902.2004.00217.x

Bardovi-Harlig, K. (1987). Markedness and salience in second language acquisition. Language Learning, 37 , 385 407. doi: 10.1111/j.1467-1770.1987.tb00577.x

Bardovi-Harlig, K. (2000). Tense and aspect in second language acquisition: Form, meaning and use . Malden, MA: Blackwell.

Bialystok, E. (1979). Explicit and implicit judgments of L2 grammaticality. Language Learning, 29 ,81 103. doi: 10.1111/j.1467-1770.1979.tb01053.x

Borg, S. (2006). Teacher cognition and language education: Research and practice . London:Continuum.

Butler, Y. (2002). Second language learners’ theories on the use of English articles: An analysis of the metalinguistic knowledge used by Japanese students in acquiring the English article system.Studies in Second Language Acquisition, 24 , 451 480. doi: 10.1017/s0272263102003042

Carroll, J., & Sapon, S. (1959). The modern language aptitude test . San Antonio, TX: Psychological

Measurement.Collins, L., Tromovich, P., White, J., Cardoso, W., & Horst, M. (2009). Some input on the easy/difcult grammar question: An empirical study. The Modern Language Journal, 93 (3),336 353. doi: 10.1111/j.1540-4781.2009.00894.x


w

http://dx.doi.org/10.1111/j.0026-7902.2004.00217.xhttp://dx.doi.org/10.1111/j.1467-1770.1987.tb00577.xhttp://dx.doi.org/10.1111/j.1467-1770.1979.tb01053.xhttp://dx.doi.org/10.1017/s0272263102003042http://dx.doi.org/10.1111/j.1540-4781.2009.00894.xhttp://dx.doi.org/10.1111/j.1540-4781.2009.00894.xhttp://dx.doi.org/10.1017/s0272263102003042http://dx.doi.org/10.1111/j.1467-1770.1979.tb01053.xhttp://dx.doi.org/10.1111/j.1467-1770.1987.tb00577.xhttp://dx.doi.org/10.1111/j.0026-7902.2004.00217.x


21/23

De Graaff, R. (1997). The eXperanto experiment: Effects of explicit instruction on second languageacquisition. Studies in Second Language Acquisition, 19 , 249 297. doi: 10.1017/s0272263197002064

De Graaff, R., & Housen, A. (2009). Investigating the effects and effectiveness of L2 instruction. InM. Long & C. Doughty (Eds.), The handbook of language teaching (pp. 726 755). Malden,

MA: Wiley-Blackwell.DeKeyser, R. (1995). Learning second language grammar rules: An experiment with a miniaturelinguistic system. Studies in Second Language Acquisition, 17 , 379 410. doi: 10.1017/s027226310001425x

DeKeyser, R. (1998). Beyond focus on form: Cognitive perspectives on learning and practicing sec-ond language grammar. In C. Doughty & J. Williams (Eds.), Focus on form in classroom sec-ond language acquisition (pp. 42 63). Cambridge: Cambridge University Press.

DeKeyser, R. (2005). What makes learning second-language grammar difcult? A review of issues. Language Learning, 55 , 1 25. doi: 10.1111/j.0023-8333.2005.00294.x

Dietz, G. (2002). On rule complexity: A structural approach. EUROSLA Yearbook, 2 , 263 286.doi: 10.1075/eurosla.2.16die

Dulay, H., Burt, M., & Krashen, S. (1982). Language two . Oxford: Oxford University Press.Ellis, N. (2002). Frequency effects in language processing: A review with implications for theories

of implicit and explicit language acquisition. Studies in Second Language Acquisition, 24 (2),297 339. doi: 10.1017/s0272263102002024

Ellis, N. (2003). Constructions, chunking and connectionism. In C. Doughty & M. Long (Eds.), Thehandbook of second language acquisition (pp. 63 103). Oxford: Blackwell.

Ellis, R. (1990). Instructed second language acquisition . Oxford: Basil Blackwell.Ellis, R. (2006). Modelling learning difculty and second language prociency: The differential

contributions of implicit and explicit knowledge. Applied Linguistics, 27 (3), 431 463.doi: 10.1093/applin/aml022

Ellis, R. (2008). The study of second language acquisition (2nd ed.). Oxford: Oxford UniversityPress.

Ellis, R. (2009). Investigating learning difculty in terms of implicit and explicit knowledge. In R.Ellis, S. Loewen, C. Elder, R. Erlam, J. Philp, & H. Reinders (Eds.), Implicit and explicit knowl-

edge in second language learning, testing and teaching (pp. 143 167). Bristol: MultilingualMatters.Field, A. (2013). Discovering statistics using IBM SPSS statistics (4th ed.). London: Sage.Goldschneider, J., & DeKeyser, R. (2001). Explaining the ‘natural order of L2 morpheme acquis-

ition’ in English: A meta-analysis of multiple determinants. Language Learning, 51 (1), 1 50.doi: 10.1111/1467-9922.00147

Goldschneider, J., & DeKeyser, R. (2005). Explaining ‘the natural order of L2 morpheme acquis-ition’ in English: A meta-analysis of multiple determinants. Language Learning, 55 (S1),27 77. doi: 10.1111/j.0023-8333.2005.00295.x

Green, P., & Hecht, K. (1992). Implicit and explicit grammar: An empirical study. Applied Linguistics, 13 , 168 184. doi: 10.1093/applin/13.2.168

Hammerly, H. (1982). Synthesis in second language teaching . North Burnaby: Second LanguagePublications.

Housen, A. (2002). The development of tense-aspect in English as a second language and the vari-able inuence of inherent aspect. In R. Salaberry & Y. Shirai (Eds.), The L2 acquisition of tense-aspect morphology (pp. 155 198). Amsterdam: John Benjamins.

Housen, A., Pierrard, M., & Van Daele, S. (2005). Structure complexity and the efcacy of explicitgrammar instruction. In A. Housen & M. Pierrard (Eds.), Investigations in instructed second language acquisition (pp. 235 270). Berlin: Mouton de Gruyter.

Hulstijn, J.H., & De Graaff, R. (1994). Under what conditions does explicit knowledge of a second language facilitate the acquisition of implicit knowledge? A research proposal. AILA Review,11 , 97 112.

Izquierdo, J., & Collins, L. (2008). The facilitative role of L1 inuence in tense aspect marking: Acomparison of Hispanophone and Anglophone learners of French. Modern Language Journal,92, 350 368. doi: 10.1111/j.1540-4781.2008.00751.x

James, C., & Garrett, P. (1992). The scope of awareness. In C. James & P. Garrett (Eds.), Languageawareness in the classroom (pp. 3 20). London: Longman.


w

http://dx.doi.org/10.1017/s027226319%3C?re3,j?%3E7002064http://dx.doi.org/10.1017/s027226319%3C?re3,j?%3E7002064http://dx.doi.org/10.1017/s027226310001425xhttp://dx.doi.org/10.1017/s027226310001425xhttp://dx.doi.org/10.1111/j.0023-8333.2005.00294.xhttp://dx.doi.org/10.1075/eurosla.2.16diehttp://dx.doi.org/10.1017/s0272263102002024http://dx.doi.org/10.1093/applin/aml022http://dx.doi.org/10.1111/1467-9922.00147http://dx.doi.org/10.1111/j.0023-8333.2005.00295.xhttp://dx.doi.org/10.1093/applin/13.2.168http://dx.doi.org/10.1111/j.1540-4781.2008.00751.xhttp://dx.doi.org/10.1111/j.1540-4781.2008.00751.xhttp://dx.doi.org/10.1093/applin/13.2.168http://dx.doi.org/10.1111/j.0023-8333.2005.00295.xhttp://dx.doi.org/10.1111/1467-9922.00147http://dx.doi.org/10.1093/applin/aml022http://dx.doi.org/10.1017/s0272263102002024http://dx.doi.org/10.1075/eurosla.2.16diehttp://dx.doi.org/10.1111/j.0023-8333.2005.00294.xhttp://dx.doi.org/10.1017/s027226310001425xhttp://dx.doi.org/10.1017/s027226310001425xhttp://dx.doi.org/10.1017/s027226319%3C?re3,j?%3E7002064http://dx.doi.org/10.1017/s027226319%3C?re3,j?%3E7002064


22/23

Jarvis, S. (2002). Topic continuity in L2 English article use. Studies in Second Language Acquisi-tion, 24 , 387 418. doi: 10.1017/s0272263102003029

Jonckheere, A. (1954). A distribution-free k -sample test against ordered alternatives. Biometrika,41, 133 145. doi: 10.1093/biomet/41.1-2.133

Kansanen, P. (1999). Teaching as teaching studying learning interaction. Scandinavian Journal

of Educational Research, 43 (1), 81 89.Krashen, S. (1981). Second language acquisition and second language learning . Oxford: Pergamon.Krashen, S. (1982). Principles and practices in second language acquisition . Oxford: Pergamon.Lado, R. (1957). Linguistics across cultures: Applied linguistics for language teachers . Ann Arbor:

University of Michigan.Larsen-Freeman, D., & Long, M. (1991). An introduction to second language acquisition research .

London: Longman.Leeman, J. (2003). Recasts and second language development: Beyond negative evidence. Studies

in Second Language Acquisition, 25 , 37 63. doi: 10.1017/s0272263103000020Long, M. (2007). Recasts in SLA: The story so far. In M. Long (Ed.), Problems in SLA

(pp. 75 116). Mahwah, NJ: Lawrence Erlbaum.Luk, Z., & Shirai, Y. (2009). Is the acquisition order of grammatical morphemes impervious to L1

knowledge? Evidence from the acquisition of plural -s, articles, and possessive ‘s. Language Learning, 59 (4), 721 754. doi: 10.1111/j.1467-9922.2009.00524.x

Lyster, R. (2007). Complementary roles for input and output enhancement in form-focused instruc-tion. In C. Gascoigne (Ed.), Assessing the impact of input enhancement in second language edu-cation: Evolution in theory, research and practice (pp. 129 151). Stillwater, OK: New Forums.

Monstrul, S., & Slabakova, R. (2003). Competence similarities between native and near-nativespeakers: An investigation of the preterite imperfect contrast in Spanish. Studies in Second Language Acquisition, 25 , 351 398. doi: 10.1017/s0272263103000159

Norris, J., & Ortega, L. (2000). Effectiveness of L2 instruction: A research synthesis and quantita-tive meta-analysis. Language Learning, 50 (3), 417 528. doi: 10.1111/0023-8333.00136

Odlin, T. (1989). Language transfer . Cambridge: Cambridge University Press.Odlin, T. (2003). Cross-linguistic inuence. In C. Doughty & M. Long (Eds.), The handbook of sec-

ond language acquisition (pp. 436 486). Oxford: Blackwell.

Pavesi, M. (1986). Markedness, discoursal modes, and relative clause formation in a formal and informal context. Studies in Second Language Acquisition, 8 , 38 55. doi: 10.1017/s0272263100005829

Phipps, S., & Borg, S. (2007). Exploring the relationship between teachers’ beliefs and their class-room practice. The Teacher Trainer, 21 (3), 17 19.

Pienemann, M. (1989). Is language teachable? Psycholinguistic experiments and hypothesis. Applied Linguistics, 10 , 217 244. doi: 10.1093/applin/10.1.52

Pienemann, M. (2003). Language processing capacity. In C. Doughty & M. Long (Eds.), Handbook of second language acquisition (pp. 679 714). Oxford: Blackwell.

Pienemann, M., Johnson, M., & Brindley, G. (1988). Constructing an acquisition-based procedurefor assessing second language acquisition. Studies in Second Language Acquisition, 10 ,217 243. doi: 10.1017/s0272263100007324

Ravid, D. (1995). Language change in child and adult Hebrew . New York, NY: Oxford UniversityPress.

Reber, A. (1989). Implicit learning and tacit knowledge. Journal of Experimental Psychology:General, 118 , 219 235. doi: 10.1093/acprof:oso/9780195106589.001.0001

Robinson, P. (1996a). Consciousness, rules, and instructed second language acquisition . Frankfurt:Peter Lang.

Robinson, P. (1996b). Learning simple and complex second language rules under implicit, inciden-tal, rule-search, and instructed conditions. Studies in Second Language Acquisition, 18 , 27 67.doi: 10.1017/s0272263100014674

Robinson, P. (2005). Aptitude and second language acquisition. Annual Review of Applied Linguistics, 25 , 46 73. doi: 10.1017/s0267190505000036

Sawyer, M., & Ranta, L. (2001). Aptitude, individual differences, and instruction design. In P. Rob-inson (Ed.), Cognition and second language instruction (pp. 319 353). Cambridge: CambridgeUniversity Press.

Schefer, P. (2011). Rule difculty: Teachers’ intuitions and learners’ performance. Language Awareness, 30 (3), 221 237. doi: 10.1080/09658416.2011.570349


w

http://dx.doi.org/10.1017/s0272263102003029http://dx.doi.org/10.1093/biomet/41.1-2.133http://dx.doi.org/10.1017/s0272263103000020http://dx.doi.org/10.1111/j.1467-9922.2009.00524.xhttp://dx.doi.org/10.1017/s0272263103000159http://dx.doi.org/10.1111/0023-8333.00136http://dx.doi.org/10.1017/s0272263100005829http://dx.doi.org/10.1017/s0272263100005829http://dx.doi.org/10.1093/applin/10.1.52http://dx.doi.org/10.1017/s0272263100007324http://dx.doi.org/10.1093/acprof:oso/9780195106589.001.0001http://dx.doi.org/10.1017/s0272263100014674http://dx.doi.org/10.1017/s0267190505000036http://dx.doi.org/10.1080/09658416.2011.570349http://dx.doi.org/10.1080/09658416.2011.570349http://dx.doi.org/10.1017/s0267190505000036http://dx.doi.org/10.1017/s0272263100014674http://dx.doi.org/10.1093/acprof:oso/9780195106589.001.0001http://dx.doi.org/10.1017/s0272263100007324http://dx.doi.org/10.1093/applin/10.1.52http://dx.doi.org/10.1017/s0272263100005829http://dx.doi.org/10.1017/s0272263100005829http://dx.doi.org/10.1111/0023-8333.00136http://dx.doi.org/10.1017/s0272263103000159http://dx.doi.org/10.1111/j.1467-9922.2009.00524.xhttp://dx.doi.org/10.1017/s0272263103000020http://dx.doi.org/10.1093/biomet/41.1-2.133http://dx.doi.org/10.1017/s0272263102003029


23/23

Schmidt, R. (1990). The role of consciousness in second language learning. Applied Linguistics, 11 ,129 158. doi: 10.1093/applin/11.2.129

Schmidt, R. (1994). Deconstructing consciousness in search of useful denitions for applied linguis-tics. AILA Review, 11 , 11 26.

Schmidt, R. (2001). Attention. In P. Robinson (Ed.), Cognition and second language instruction

(pp. 3 32). Cambridge: Cambridge University Press.Shirai, Y. (2004). A multi-factor account for form-meaning connections in the acquisition of tense aspect morphology. In B. VanPatten, J. Williams, S. Rott, & M. Overstreet (Eds.), Form meaning connections in second language acquisition (pp. 97 122). Mahwah, NJ: Law-rence Erlbaum.

Shiu, L. (2011). EFL learners’ perceptions of grammatical difculty in relation to second language prociency, performance, and knowledge (Doctoral dissertation). University of Toronto,Toronto, Canada.

Skehan, P. (1998). A cognitive approach to language learning . Oxford: Oxford University Press.Skehan, P. (2002). Theorizing and updating aptitude. In P. Robinson (Ed.), Individual differences

and instructed language learning (pp. 69 93). Philadelphia, PA: John Benjamins.Spada, N., & Tomita, Y. (2010). Interactions between type of instruction and type of language

feature: A meta-analysis. Language Learning, 60 (2), 263 308. doi: 10.1111/j.1467-9922.2010.00562.x

Swain, M. (2005). The output hypothesis: Theory and research. In E. Hinkel (Ed.), Handbook of research in second language teaching and learning (pp. 471 481). Mahwah, NJ: LawrenceErlbaum.

Taraban, R. (2004). Drawing learners’ attention to syntactic context aids gender-like categoryinduction. Journal of Memory and Language, 51 , 202 216. doi: 10.1016/j.jml.2004.03.005

Terpstra, T. (1952). The asymptotic normality and consistency of Kendall’s test against trend, whenties are present in one ranking. Indagationes Mathematicae, 14 , 327 333.

VanPatten, B. (2007). Some thoughts on the future of research on input enhancement. In C.Gascoigne (Ed.), Assessing the impact of input enhancement in second language education: Evolution in theory, research and practice (pp. 169 189). Stillwater, OK: New Forum.

White, J. (1998). Getting learners’ attention: A typographical input enhancement study. In C.

Doughty & J. Williams (Eds.), Focus-on-form in classroom second language acquisition(pp. 85 113). Cambridge: Cambridge University Press.White, J., Munoz, C., & Collins, L. (2007). The his/her challenge: Making progress in a ‘regular’ L2

programme. Language Awareness, 16 (4), 278 299. doi: 10.2167/la419.0White, L. (1991). Adverb placement in second language acquisition: Some effects of positive and

negative evidence in the classroom. Second Language Research, 7 , 133 161. doi: 10.1177/026765839100700205

Zobl, H. (1985). Grammars in search of input and intake. In S. Gass & C. Madden (Eds.), Input in second language acquisition (pp. 32 44). Rowley, MA: Newbury House.


w
http://dx.doi.org/10.1093/applin/11.2.129http://dx.doi.org/10.1111/j.1467-9922.%3C?re3,j?%3E2010.00562.xhttp://dx.doi.org/10.1111/j.1467-9922.%3C?re3,j?%3E2010.00562.xhttp://dx.doi.org/10.1016/j.jml.2004.03.005http://dx.doi.org/10.2167/la419.0http://dx.doi.org/10.1177/026765839100700205http://dx.doi.org/10.1177/026765839100700205http://dx.doi.org/10.1177/026765839100700205http://dx.doi.org/10.1177/026765839100700205http://dx.doi.org/10.2167/la419.0http://dx.doi.org/10.1016/j.jml.2004.03.005http://dx.doi.org/10.1111/j.1467-9922.%3C?re3,j?%3E2010.00562.xhttp://dx.doi.org/10.1111/j.1467-9922.%3C?re3,j?%3E2010.00562.xhttp://dx.doi.org/10.1093/applin/11.2.129

defining grammatical difficulty a student teacher perspective

Documents