Language knowledge and its application: a snapshot of Australian teachers' views.

Love, Kristina ; Macken-Horarik, Mary ; Horarik, Stefan 等

Introduction

The Language strand of the Australian Curriculum: English (ACARA, 2012) provides stretch for English teachers in a number of directions in terms of their subject knowledge. The first stretch is in their understanding of 'the structures and functions of word and sentence-level grammar and text patterns and the connections between them' (Commonwealth of Australia, 2009, p. 7). This knowledge of grammar as it operates within and between levels of language is further enhanced by a knowledge of how grammatical choices function in a range of contexts. Beverley Derewianka, one of the architects of the AC:E, identifies the model of language underpinning it as 'a unifying model of language in context which (has the potential to) bring together form and function, operating seamlessly from the level of discourse down to the phoneme' (Derewianka, 2012, p. 129). Furthermore, the sub-strands of the Language strand are organised to present language as systems of resources for making meaning. '(t)he choices we make from the language system are constrained by certain features in the context: the social purpose (genre), the field (subject matter), the tenor (roles and relationships), and the mode (channel of communication)' (Derewianka, 2012, p. 139). Knowledge of both the multidimensional and multi-strata features of language may represent significant reach for some teachers.

The second stretch for teachers, in a digital literacies environment, is the expansion of the knowledge base of English to integrate understandings of verbal grammar with new understandings of multimodal texts. One of the content descriptors for Language at Year 5 for example is for students to be able to 'Explain sequences of images in print texts and compare these to the ways hyperlinked digital texts are organised, explaining their effect on viewers' interpretations' (ACELA1511). Do Australian teachers have a confident grasp of the metalanguage required for this and related tasks in English? The third stretch is the systematic building of knowledge, requiring teachers to map knowledge in English progressively across the years of schooling as they develop 'a view across time' (Freebody, 2007). This requires that teachers have necessary knowledge not just to teach the curriculum for their year level but also to understand learning expectations for students in earlier years and to anticipate those in later years of English.

The AC:E thus presumes strong subject knowledge, which, when mobilised confidently, empowers teachers with resources for exploring the often complex and multimodal texts of 21st century English in order to foster cumulative knowledge from Foundation to Year 10. However, the expanded knowledge base required may have stretched teachers beyond their comfort zones. Do teachers' understandings converge with those of the curriculum? How productively are teachers exploiting its affordances? How confident do they feel to implement its requirements? Overall, how are teachers engaging with the expanded demands of English and the increased scope for grammar within this?

This paper addresses these questions as it reports on findings from a national survey of English teachers conducted as part of a large-scale project funded by the Australian Research Council from 2011-2014 (DP110104309). Following three years of case study research with 27 teachers and their students, we developed and disseminated a survey to investigate their views relative to those of English teachers nationwide and across all years of teaching from Foundation to Year 12. The aim of the national, cross sectorial survey was to provide empirical evidence about how ready teachers are to plan for, teach and assess language, even with the 'scope and sequence' of language embedded in the AC:E. The paper documents trends observed in quantitative analysis of teachers' beliefs about what is important in teaching language; levels of confidence in their own knowledge about language and image; and challenges faced in implementing knowledge of language and image in teaching. Such evidence is crucial if the profession is to continue to build a secure knowledge base for English, literacy and EAL/D teachers.

Characterising Linguistic Subject Knowledge (LSK) in English

Perhaps the first task facing researchers of teachers' knowledge of language is to define it and delimit its scope. The field is extensive and its various dimensions are captured by a wide range of terms. Terms such as 'grammatical knowledge' have been used by researchers in the UK (e.g. Andrews, 2005) to describe the capacity to identify word classes and linguistic features at the syntactic level, and to apply this in the construction of well-formed written sentences. Meta-analyses using this description in the mid-1990s concluded that teachers' grammatical knowledge is 'patchy and idiosyncratic' (Brumfit et al., 1996, p. 86). Others (e.g. Carter, 1990) use the term 'Knowledge about Language' to refer to broadly defined understandings of language and the abbreviation KAL has much currency in Australia. More recently, UK researchers use the term 'Linguistic Subject Knowledge' (Myhill et al., 2012) to describe not only the capacity to identify and apply word classes and syntactic features but 'the ability to explain grammatical concepts clearly and know when to draw attention to them' (Myhill et al., 2012, p. 142). Sangster et al. (2012) have broadened this definition even further to involve abilities to identify, discuss, analyse and evaluate literary devices.

Our study adopted the term 'Linguistic Subject Knowledge' (LSK) to capture multilevel and multidimensional knowledge of language that can be defined, explained and applied to the study of texts. These include verbal and multimodal texts that are increasingly prominent in the AC:E. In keeping with the view of language underpinning the AC:E, we construed LSK as a knowledge of different levels of language knowledge, attentive to form, function and meaning in a wide range of texts. A capacity to integrate grammar into the study and composition of texts is thus a key aspect of both the AC:E and the larger study out of which our survey developed. A key point here is that such a rich characterisation of LSK takes teachers and students far beyond a knowledge of grammatical terminology, often tested in prior studies (e.g. Andrews, 2005; Cajkler & Hislam, 2010; Kolln & Hancock, 2005) and found wanting. It includes necessary ability to identify and name grammatical structures but transcends this to incorporate teachers' ability to interpret and explain the contribution of grammatical structures to patterns of meaning in texts.

Having settled on the scope and definition of a term like LSK, as researchers we still know little about current levels of LSK among Australian teachers, whose diverse starting points may vary considerably across the country. Some teachers whose schooling and university education occurred in what Sangster and colleagues call a 'post-traditional grammar era' (Sangster, 2012, p. 2) may have little grammatical terminology of any sort and little explicit understanding of how language works. Others, who have acquired knowledge of sentence-level language structures rooted in Latinate grammars, may struggle with the functional terminology that underpins the AC:E. Still others with literary expertise developed during university study of English may find that they cannot integrate close study of language (including grammar) with interpretation of narratives, poetry or drama. Knowledge of literature and preferential selection of Literature majors in pre-service English courses coincides with a lack of LSK of a kind developed by Linguistics majors (Hudson & Walmsley, 2005). Such diverse starting points may make a common knowledge base much harder to establish.

Just as starting points vary in relation to LSK, so do views circulating in popular discourse about teachers' linguistic knowledge. We reported in earlier papers (Love et al., 2014; Macken-Horarik et al., 2011) on a consistent discourse of deficit in media reports on teachers' knowledge about grammar. Persistent criticisms of teachers' lack of LSK (framed in the rhetoric of 'Poor Grammar' and focussed on knowledge of grammatical and spelling conventions) and the ongoing Australian Teacher Quality debate (Louden, 2008) make it all the more urgent that questions about teachers' knowledge of different facets of language be addressed constructively. The expanded knowledge base and metalanguage of the AC:E offers the profession both opportunities and challenges. But we need to understand not just what teachers know about language, broadly conceived; we need to know how teachers apply understandings about language in planning, teaching and assessment.

Linguistic Pedagogic Subject Knowledge (LPSK)

Teachers' practice involves not only declarative knowledge, or 'knowledge about', but 'knowledge of', that is, knowledge 'activated when a need for it is encountered in action' (Scardamalia & Bereiter, 2006, p. 105). UK researchers like Myhill et al. (2012) caution that declarative knowledge about language, though necessary, is not sufficient for educational purposes and that 'teachers need to apply their LSK to published texts and to children's own writing, identifying significant linguistic features and being able to make connections for writers between a feature and its impact on a text or reader' (p. 162). They have coined the term 'Linguistic Pedagogical Subject Knowledge' (LPSK) to highlight the importance of turning 'knowledge about' language into linguistic 'know-how'. Myhill et al. (2013, p. 89) further argue that teachers with a secure LPSK are effective in three ways in the teaching of writing: firstly, they link linguistic features to a specific writing effect; secondly, they respond to students' writing sensitively, drawing out explicitly effective choices in the writing; and finally, they notice relevant aspects of reading texts to draw to students' attention. While Myhill et al. (2012, 2013) have focused largely on the impact of teacher LPSK on student writing, the AC:E offers resources for English teachers to make more or less explicit connections between language (in the extended sense referred to earlier) and the many other aspects of textual work, reading as well as writing. At stake here is teachers' ability to integrate grammar into study of the literary, persuasive and interpretive texts of English.

There has been much research exploring teacher preparedness and capacity to make these connections (Myhill et al., 2012), with findings that teacher beliefs about the importance of explicit teaching of language profoundly inform English curricula and classroom approaches. For example, in the UK, Brumfit, et al. (1996) found that teachers used text-based approaches to teaching language, rarely if ever taught grammar explicitly and believed that grammatical knowledge (understood as sentence level grammar) was only marginally significant to the development of students' overall linguistic ability. In a parallel way, Cameron (1997) found that UK teachers were hostile to approaches making the formal structure of language an object of analysis in its own right.

Research in the Australian context, where language knowledge expands beyond sentence to text level (genre) understandings and multimodal forms, presents a more complex portrait of teacher beliefs and practices regarding explicit teaching of language. Hammond and Macken-Horarik's (2001) survey of 126 Sydney Grades 3 and 5 teachers a year after the release of the New South Wales (NSW) primary English curriculum (NSW Board of Studies, 1998), indicated strong support for explicit and systematic teaching of literacy, contrasting strikingly with studies in the UK by Brumfit et al. (1996) and Cameron (1997) canvassed above. Despite strong commitment to explicit LSK, however, most NSW teachers expressed a lack of confidence in their own knowledge about language. For example, only 42% expressed confidence with rules of traditional grammar and 6% with functional grammar.

This disjunction between beliefs about the importance of LSK and confidence to teach it contrasts with Sangster's (2012) findings for Scottish pre-service teachers, whose confidence markedly outstripped knowledge, as measured in tests of grammatical knowledge. Importantly, Sangster et al. (2012, p. 19) concluded that this disparity between high levels of confidence and actual LSK could be a barrier to further understanding in preservice teachers. If teachers already feel confident with their LSK, this may mean that they do not recognise what they don't understand. It may be that access to specialised LSK drives 'easy confidence' levels down as teachers come to terms with the need for precision about workings of language at different levels of organisation.

Given the extent to which linguistic knowledge is incorporated systematically and explicitly in the AC:E, questions remain about what challenges teachers face as they build both their LSK and their LPSK in light of this new curriculum. Have levels of confidence risen since Hammond and Macken-Horarik's 2001 study? Does the Language strand of the AC:E continue to make 'substantial demands on teachers in terms of their subject matter knowledge and pedagogic knowledge' (Jones & Chen 2012, p. 149)? If so, what does this mean for implementation of the curriculum? Our survey offers a snapshot of the profession in 2014 and its uptake of the Language strand of the AC:E. It provides one picture of the current knowledge base of the profession and its capacity to deliver on the richly conceived model of LSK that the curriculum now asks Australian teachers to implement.

The survey: its design and scope

An anonymous online survey (http://www.une.edu. au/about-une/academic-schools/school-of-education/ research/arc-funded-projects#item0) was one component of the large-scale project described earlier that explored teachers' LSK in the context of changes in school English. This national survey sought to gain both quantitative and qualitative insights into key issues identified above, as part of a benchmarking of the views of those teachers participating in our study against those in the profession at large. The survey, designed by the researchers, was structured as a series of quantitative (Likert scale) questions and qualitative (open-ended) questions exploring teachers' views and experiences.

Part 1 sought a demographic profile of respondents (years of teaching, years of teaching English, school and class data and levels of teaching) and data about their students (state, system, region, profiles of student groupings and factors that affect their achievement levels in English). Part 2 sought information about aspects of language important in subject English and Part 3 explored respondents' views on specific aspects of language (such as text type and structure, types of sentences, grammatical components of sentences, cohesion, multimodality). Part 4 gauged teachers' levels of confidence in teaching specific aspects of language itemised in earlier questions and challenges faced with different aspects of language. Finally, Part 5 asked teachers to identify their professional learning needs, with an open ended question allowing them to elaborate on the kinds of support they required in teaching their students about language.

The on-line survey was disseminated by professional English and Literacy Associations in each state and was 'live' from mid-2014 to early 2015. Given that we did not offer any incentives to participate, the response rate is pleasing and indicates the importance of professional associations to studies like this one. The survey generated 373 viable responses across all states. Tables 1a-d below provide a picture of the scope of these responses. Response profiles may be somewhat skewed given investments in 'guild learning' by teachers active in their professional associations (Freebody, 2007). Such teachers typically invest in professional debates about matters such as LSK or LPSK in English. Even so, the data comes from a wide range of teachers across states, systems and regions and includes responses from cohorts working at all year-levels. The survey thus provides an indicative picture of national trends from professionally active teachers.

The quantitative data was collated and analysed by one of the authors with statistical expertise using SPSS software (Version 22). Our first analysis produced frequencies of all responses and a second analysis produced cross-tabulations to identify significant relationships between different variables.

Responses to the 4 open-ended questions asked in Parts 2-5 of the survey were collated into Excel spreadsheets and coded by the researchers through a process of grounded theorising (Glaser, 1978). In this respect, the first task was to develop categories as close as possible to those proffered by respondents themselves. Following this, emerging codes were related to key themes arising in contemporary research in language and literacy education (e.g. Luke & Freebody, 1999), in relational approaches to LSK (e.g. Myhill et al., 2012; Derewianka, 2012), and in professional knowledge (e.g. Hattie, 2009). In the remainder of this paper, we will use discursive comments purely to illustrate trends emerging in the quantitative data.

Survey findings

Overall, the survey revealed that Australian English and Literacy teachers believe that teaching students about language is important, approach this in highly contextualised ways in their pedagogy, express high levels of confidence in their own knowledge, but also require greater professional learning support on key aspects of LSK and LPSK than is currently available. Due to constraints of space in this paper, we focus only on three key aspects of the findings: what kinds of LSK teachers think is important; teachers' avowed levels of confidence in their LSK; and their current approaches to teaching of language (LPSK) in English.

Teachers' views on the value of knowledge about language in school English

Teachers reported on the importance of aspects of LSK in general and on aspects of LSK like text structure, sentence-level knowledge, cohesion and application of this knowledge (LPSK) in literacy. Teachers' views were identified quantitatively on a five-point Likert scale and their responses supplemented with comments on open-ended questions about what else was important in language. In order to capture broad findings about what teachers valued, we grouped their responses into related sets such as 'important'/'not important'. In this way, we captured trends in views on scales of relative importance. Table 2 provides frequencies and percentages of teachers' rating of language as 'important', 'very important' or 'extremely important'. The aspects of language identified in this table captures teachers' views on importance of text, sentence and word levels of language in addition to views on application of such LSK to composition and critical analysis of texts, including analysis of visual and multimodal texts.

The high percentage of teachers indicating the importance of student capacity to identify and name the structural stages of the texts of English (77.5%) is perhaps unsurprising given the focus on text types in Australian curriculum over the last 20 years. With insignificant variation across the states, it confirms at a national level, trends indicated by Hammond and Macken-Horarik in 2001 for NSW. In terms of variation by Grade level, a multiple response Crosstab analysis revealed that the percentage of teachers for whom identifying and naming the structural stages in various kinds of texts was important to extremely important gradually increased from 69.5% among Early Years teachers, to 80% among Mid and Upper Primary teachers, 90.8% among Early and Mid-Secondary teachers, and 90.1% among Senior Secondary teachers. The slightly smaller percentage of Early Years and Middle Primary respondents indicating the importance of this feature is of interest, given its significance in the AC:E, and warrants further investigation, despite the smaller numbers of respondents teaching at these levels.

Nonetheless, the findings overall represent a strong endorsement of the importance of genre-based approaches as characterised in the AC:E, focusing as they do on text structure and organisation. Open-ended responses to Question 3.5 (Overall, what are the most important aspects of language that your students need to learn about?) illustrated teachers' endorsement of genre knowledge as important, one Early Years teacher commenting 'In Foundation, students need to learn that text is used for different purposes and, as a result, has different features' and an early secondary teacher commenting 'Functional grammar would include a basic understanding of common classroom text structures and associated grammar, e.g. past tense and sequence markers in a recount'.

However, teachers' valuing of genre knowledge contrasts somewhat with their views on sentence-level knowledge. Looking at the second row of Table 2, we note that only 63.2% of teachers indicated the importance of students being able to name different types of sentences (e.g. simple, compound, complex) and explain grammatically how they are different. Crosstab analysis revealed little variation across states, but did indicate variation by year level taught, with a gradual increase in teachers' perceptions of the importance of naming different types of sentences as students move into the middle years of school (from 58.5% of Early Years, 69.5% of Mid and Upper Primary, to 70.1% of Early and Mid-Secondary teachers). Understanding grammar is seen as increasingly important as students move up the secondary years of English. A similar pattern emerges for the third item in Table 2, related to the perceived importance of knowledge of grammatical components of sentences. Sentence-level LSK is considered important by teachers at all levels of schooling across the states, but overall, less so than genre knowledge. These trends are consistent with Hammond and Macken-Horarik's 2001 findings on grammatical knowledge among NSW primary teachers.

Moving to the fourth row of Table 2, we note a higher percentage (68.4%) of teachers overall indicating the importance of students being able to describe and use different aspects of cohesion (e.g. referring words, word associations, ellipsis). Crosstab analysis reveals very little variation across year levels but somewhat greater variation across states on this dimension of language. Teachers in different states accord high but varying levels of importance to student capacity to describe and use cohesion in texts, ranging from New South Wales (82.9%), Queensland (82.0%), Australian Capital Territory (76.9%), Western Australia (69.6%) Victoria (69.6%), South Australia (56.5%) and Tasmania (50%). Lower numbers of respondents from Tasmania and ACT, however, make us cautious about generalising from these findings.

Overall, it appears that text and sentence-level LSK, as captured in the first four rows of Table 2, is generally considered important by Australian teachers across all states and at all levels of schooling. Australian teachers appear to accord significantly more importance to knowledge about structural and linguistic features compared with their counterparts in the UK (e.g. Brumfit et al., 1996; Myhill et al., 2012). However, the smaller percentage of teachers identifying the importance of sentence level grammatical knowledge, when compared with other aspects of language requires further investigation. Minor variations across states on these aspects of LSK may also have historical explanations worth exploring.

Responses to the open-ended question (Q3.5: What are the most important aspects of language that students need to know about?) complexify this national portrait and suggest further areas of enquiry into teachers' perceptions of LSK. There was considerable variety in the detail and content of these responses and initial analysis suggests a varied range of understandings of what sentence-level LSK means for the profession at different year levels. To illustrate, English as Additional Language/Dialect (EAL/D) teachers more often expressed strong views about the importance of LSK and identified types of sentences, syntactic features or cohesion as the most important aspect of language, response #66 being typical: 'The group has a range of levels from close-to-beginner in English, to almost ready to move to mainstream English. The former needs close attention to basic grammar; the latter on complex sentence forms and text-wide features such as reference and ellipsis.' Further analysis of these discursive comments will be undertaken, along with cross-tabulations with statistical data, to identify more fully the range of teacher understandings of sentence-level LSK at different year levels and states.

The last three rows in Table 2 refer to teachers' perceptions of the importance of student capacity to apply knowledge about language in composing texts, critically analysing language choices and understanding how visual and multimodal texts work. A high percentage of teachers identified application of linguistic knowledge (LPSK) to composition (86.6%) and critical analysis (84.5%) as important. Such figures appear to contrast with international findings such as those of Cajkler and Hislam (2010) in the United Kingdom, Kolln and Hancock (2005) in the United States, which identified gaps in English teachers' preparedness to integrate grammar in mainstream English classes. However, the impression of higher levels of LPSK among Australian teachers again becomes complex when we examine our respondents' open-ended responses to Question 3.5, in which they specified the underpinning body of knowledge to be applied. Many comments here remained at a general, non-linguistic level, response #134 being typical: 'Using language to articulate opinion, analysis and creative output with confidence in all aspects of their lives.' Only a small number of responses specified the linguistic resources important for students to apply, response # 261 being representative: 'Whilst knowledge of grammar, text types etc is important, it is more than just knowing the structures or what they are. Knowing that a verb is a 'doing word' isn't of much use unless you know how to use it. I believe it is also important for students to know the how and why of grammar and structures and how use of these enhances their use of language'. Such views which emphasise the importance of rhetorical understandings to effective language use, align strongly with the intentions of the AC:E, and their relative infrequency is cause for some concern and further analysis.

Visual and multimodal analysis is strikingly important to the majority of respondents. The high percentage (83.7%) of teachers who consider this as important stands in contrast with much lower figures (43%) reported in response to a similar question by Hammond and Macken-Horarik (2001). This contrast is perhaps unsurprising given the increased focus on multimodality in English in the last 15 years (e.g. Callow, 2013). Even so, few teachers provided specific comments about this aspect in worded responses to open-ended questions, suggesting that while enthusiasm may be high for multimodal text analysis, the detail on this in English is slight.

In summary, our analysis from Part 2 of the survey indicates that, despite some variation across year levels, an overwhelming majority of respondents believe that students need to know about language at text, sentence and word levels and apply this in composing and analysing texts. Tools for analysing visual and multimodal texts are regarded as highly important. More variable attitudes towards the importance of sentence level understandings emerge from both the statistical and discursive responses. Furthermore, these vary considerably in levels of specificity and technicality. Further studies are needed to reconcile recognition of the importance of an aspect of subject knowledge with teachers' ability to offer precise, linguistically-informed explanations of that knowledge.

Teachers' reports on their approaches to teaching of language in English

The next cluster of survey questions sought to identify teachers' approaches to teaching of language their linguistic pedagogic subject knowledge (LPSK). In the absence of direct observations of teacher practice, these measures provide a rich national picture of how teachers integrate knowledge about language into teaching of English.

Question 3 asked teachers to indicate how they taught about each of the above aspects of language on a 5 point scale: 0: 'Not at all', 1: 'Indirectly', 2: 'At the point of need', 3: 'As one key aspect of the writing process' and 4: 'As an aspect of language as system'. A majority of teachers (over 69%), reported that they taught each of four aspects of language (text structure, cohesion, sentence type and sentence structure) either as 'One part of the writing process', or 'Explicitly and systematically'. Cross tab analysis indicated that a high number of teachers across all year levels and states value explicit and systematic teaching, most prominently of text structure, followed by cohesion, sentence type and sentence structure in that order. Cross tab analysis further indicates that it is largely in the middle to upper primary school curriculum where the most explicit focus on sentence level knowledge (sentence type and structure) is to be found.

Overall, our cross tabulations present an important profile of teachers' integration of key aspects of verbal language into teaching. As a generalisation, it appears that grammatical knowledge is taught more explicitly in upper primary and junior secondary years of school, and that specialised text structuring, rather than 'language features', is the focus of senior secondary English. This profile can be elaborated by examining teachers' views on the integration of language into pedagogy. Once again, though responses were initially coded on a five-point Likert scale, we grouped these into two sets ('not important to somewhat important' and 'important to extremely important'). Table 3 summarises responses to this question about the importance of integrating grammar into pedagogy.

It is clear immediately that integration is a prized pedagogic goal in English. The percentage of teachers attributing high importance to integration of grammar into literature study (82.6%) contrasts with relatively low percentages in Hammond and Macken-Horarik's (2001) study, which reported 45% on a similar measure with Grade 3 and 5 teachers. This may be because integration of grammar has become increasingly important in the intervening 15 years or because it is more likely to be valued in secondary English, which focuses on literature. The responses on the open-ended question reveal details about particular aspects of language that teachers integrate. Response #54 made by a Year 7 teacher reports integrating specific features of language structure and function in a range of text production and text analysis activities: 'We discuss language choices when reading texts, when we are creating jointly constructed texts, during whole class novel reading, guided reading lessons, and when teachable moments arise.' In contrast, Response #224 made by a Year 8 teacher reports practices where the language focus is deliberately diffuse and grammatical terminology eschewed: 'I expose students to rich literature and close study of rich literature as a whole class. I avoid over labeling and over complicating of grammatical concepts'. Again, it appears that teachers bring highly varied views on what counts as relevant 'grammatical concepts' to relate to literary texts.

Moving to the second row of Table 3, the relatively high percentage of teachers attributing importance to explicit teaching of grammar in texts (72.6%) likewise suggests that teachers are integrating language with the Literacy strand of state and national curricula. Crosstab analysis indicated little variation across states and year levels on this question. Open-ended responses to Question 4 reinforce the importance of a contextualised approach to language learning, the comment by a Year 9 teacher (#291) being typical:

'I do not teach my students about grammar that is irrelevant to what I'm teaching e.g. one off lessons about adverbials from a text book. I do however teach it explicitly but as long as it is connected to the writing I am asking my students to do at the time. At our school we are focusing on modelling, joint construction and 'Reading to Learn' strategies to improve student outcomes. Grammar has become a whole school focus and shown significant improvement on students reading and writing abilities. I have found that teaching the grammar through sentence patterns and small achievable 'chunks' for the students stops that fear of the unknown.'

A clear majority of teachers reported valuing a contextualised approach to grammar, confirming insights generated in other studies of embedded grammar teaching in English (e.g. Fearn & Farnan, 2007; Kolln & Hancock, 2005; Myhill et al., 2012).

However, an emphasis on contextualised approaches to grammatical instruction, while dominant, was not exclusive. A significant minority (29.2%) also value grammatical knowledge for its own sake as they reveal that 'Giving students decontextualised grammar exercises e.g. from grammar resources or textbooks' is 'important, very or extremely important'. This finding requires careful interrogation, since it raises the question of what LSK is considered as foundational as an antecedent to application in work on texts. Building knowledge about grammar is a pre-requisite, one would think, to effective contextualisation of grammatical instruction.

The final line of Table 3 reinforces the high importance that teachers place on students knowing about multimodal texts (consistent with Table 2), and foregrounds teachers' perceptions of the importance of the integration of language into the study of multimodal texts. The 80.2% identifying this as important ('important' --'extremely important') is a significant contrast with Hammond and Macken-Horarik's (2001) study that reported only 43% on a similar measure. The differences are not surprising, given the vast research and professional resource development in this arena over the last decade.

These figures also raise questions about which systems of language teachers apply in analysis of multimodal texts like picture books or websites. Interestingly, no teachers emphasised multimodal work in their discursive responses to Question 4 about how they approached language teaching. This suggests that teachers do not yet perceive this as work on language. Only five responses to the open ended question asking teachers to list challenges about language specifically mentioned 'understanding visual grammar' or 'understanding multimodal meaning' as concerns. Technical analysis of multimodality is relatively new to English (Unsworth, 2013) and while indicated as important, few teachers had the meta-language to identify particular challenges in this field.

In summary, our statistical analysis of Part 3 indicates that a clear majority of Australian teachers attribute high importance to the integration of language into work on texts, including multimodal texts. Yet again, our discursive data provides both an elaboration of and counterpoint to this portrait, suggesting that teachers bring different views on what counts as relevant grammatical concepts to work on texts. Australian teachers clearly have strong views on what is important to understand and teach in relation to language and image. The question of their levels of confidence to do so immediately presents itself.

Teachers' levels of linguistic confidence and challenges they face

Question 4 of the survey identified respondents' levels of confidence in teaching specific aspects of language covered in earlier questions. Respondents again had the opportunity to elaborate in a subsequent open-ended question on challenges they faced with particular aspects of language. Table 4 below summarises percentages of responses of teachers who indicated in Question 4 that they were confident-extremely confident to teach language.

Overall, a high level of confidence (72.7%) is reported across the board in teacher LSK (last row of Table 4), with only marginal variation across Year levels. The very high percentage (85.5%) of teachers expressing confidence in their knowledge of, and capacity to teach text types and structures (see row 1, Table 4) is in keeping with teachers' reports of the importance of this knowledge (see row 1, Table 2) and with Hammond and Macken-Horarik's 2001 findings that the largest area of teacher confidence was in teaching text types (69%). High levels of confidence were also reported for those aspects of LSK to do with cohesive resources (72.9%) and sentence structure (76.1%), indicating a stark contrast with low measures of confidence on these aspects reported by Hammond and Macken-Horarik (4% and 42% respectively). While NSW teachers in 2001 claimed it was important to teach children about sentence structure and cohesion, most felt under-confident to do so.

Of course, teacher confidence in their knowledge about language does not necessarily match competence (e.g. Harper & Rennie, 2009; Louden et al., 2005). The study by Sangster et al. (2012) is a salient reminder that avowed levels of confidence can markedly outstrip demonstrated levels of competence, particularly in areas dependent on technical knowledge such as sentence grammar. The potential for a similar disjunction between actual and reported confidence in LSK in our survey increases when we turn to responses to the open-ended question asking which areas of language teachers found particularly challenging. Here, over 20% of respondents reported finding sentence/clause structure difficult, with specific areas identified such as the following: 'Classifying parts of speech--functional or traditional'; 'Analysing grammar confidently and accurately'; 'Teaching tense'; 'Teaching gerunds; modal verbs' and 'Teaching clauses and sentences'.

A high percentage of teachers (76.5%) indicated confidence in their ability to teach about the grammar of visual and multimodal texts, consistent with findings on the importance of multimodal analysis in Table 2. Again, these figures raise questions about meta-languages teachers deploy as they explore contribution of words and images to multimodal meaning. They may be confident for example with relatively commonsense terms when describing the elements of visual texts, or the relationship between the visual and verbal elements in a multimodal text. Or they may use a precise technical metalanguage. But the lack of responses to the open-ended question on this issue suggests a lack of specialist knowledge about multimodality. The few teachers who did mention 'visual grammar' or 'multimodal meaning' as concerns seemed to be seeking a more grammatical framework for analysing multimodal texts, but it is possible that many others were simply more comfortable with less technical frameworks.

Overall, while teachers expressed high levels of confidence in their capacity to teach about the four aspects of LSK identified in Table 4, the qualitative data suggests that teachers are limited in their capacity to identify, name and explain grammatical features, particularly those which operate at sentence level and in multimodal texts. This implies that teachers assume a confident stance on general points such as those identified in Likert scale questions but are far less confident when it comes to technical specifics. This view is reinforced in responses to the open-ended question in Part 5 of the survey, where teachers were asked to identify their professional learning needs. Here, while testifying to a contextualised orientation to the teaching of language, many comments highlight a need for more professional support in terms of both LSK (e.g. 'PD on parts of speech, sentence construction and cohesion') and PLSK (e.g. 'In-service training for integrating grammar into meaningful investigation of a range of texts') that is somewhat at odds with their professed levels of confidence.

Conclusion

At the outset of this paper, we enquired about the extent to which Australian teachers were engaging with the expanded demands of English and its knowledge base, especially the increased scope for language within the AC:E. Positing a multidimensional model of knowledge which involves both general and specific expertise about language at text, sentence and word levels of organisation has enabled us to 'net in' different dimensions of teachers' understandings. This includes the ability to name, define, explain and relate grammatical and other linguistic choices to literary and multimodal meaning. But for teachers who have to implement what they know about language in diverse classroom settings, it is the application of knowledge--or LPSK that counts. We have drawn on these two dimensions of linguistic knowledge in providing a broad portrait of teacher perceptions on key aspects of language now highlighted in national curriculum demands.

Firstly, the vast majority of respondents to our survey claimed that language knowledge is important to English. While knowledge of text structure is highly valued, so too are more complex aspects of LSK like cohesion. The profession gives extremely high levels of importance to multimodal knowledge across all levels and systems of schooling. This national stance is a crucial prerequisite for any meaningful adoption of an English curriculum with Language (including multimodality) as one of three interrelated strands. However, the significantly smaller proportion of teachers reporting the importance of knowledge of sentences and their grammatical components invites further research into what is understood as LSK and how much teachers actually know about each aspect of LSK.

Secondly, our findings have revealed a highly contextualised approach to teaching of language. The three strands of the AC:E are organised on the assumption that Grammar is related to patterns of meaning in texts and to contexts of usage. The sense of a contextualised orientation to teaching of language emerging from the Likert scale responses is reinforced in the extensive responses to open-ended question 3.5 (What are the most important aspects of language that students need to know about?), which reflect a deep engagement with language as a resource for meaning-making, rather than as a set of prescriptive rules. Some small variation in LPSK across states and year levels warrants further investigation. So too does the minority of teachers who stressed the importance of accessing grammatical knowledge in text books and other decontextualised resources, signalling as they do a need for more specialised understandings of grammar within LSK. Further studies are needed in this area, especially at a time when large-scale, mixed methods studies point convincingly to the benefits of contextualised approaches to the teaching of grammar (Jones et al., 2013; Myhill et al., 2012). Just how do we ensure that all English teachers build coherent and secure expertise in language and are well prepared to teach this effectively? Part of the response may be to induct English teachers into more systematic (even technical) understandings about linguistic and other modes of communication. Addressing this challenge is crucial in light of findings that 'teachers whose grammatical pedagogical content knowledge was more secure fostered greater learning about writing and the repertoire of possibilities open to writers' (Myhill et al., 2013, p. 89).

Thirdly, teachers nationally professed generally high levels of confidence in their own LSK, particularly in relation to the structure of different text types, but also in relation to sentence structure and cohesive resources. In contrast to much lower levels of confidence revealed in earlier Australian studies (e.g. Hammond & Macken-Horarik, 2001; Harper & Rennie, 2009), our respondents avow that they already know what they need to know about language. Even so, some disjunction appeared between discursive and Likert-scale data, calling into question again what teachers understand by LSK. Through cross-tabulating the Likert scale responses to Question 4.5 (identifying confidence about LSK) with coded discursive responses to Q4.6 (identifying challenges), we will further explore this relationship between avowed levels of confidence and remaining challenges.

International research points to the need to narrow the gap between teachers' professed confidence and actual expertise (Gordon, 2005; Myhill et al., 2012; Sangster et al., 2012; Williams, 2005), both about language as a body of knowledge and about principles underpinning effective implementation of this in classrooms. Direct evaluations of teacher LSK, designed in keeping with the expanded view of language in the AC:E may clarify the extent of this disjunction between the avowed and the actual and may help identify the linguistic knowledge which '... confidently communicated by the teacher offers the potential of increasing students' language repertoires and thus expanding their meaning-making resources' (Coffin, 2010 p. 4).

Though dealt with only briefly in this paper, Part 5 of our survey also revealed rich data about teachers' views of the professional support needed to support them in building their students' language knowledge in coherent and cumulative ways as required by the AC:E. Overall, comments here were somewhat at odds with teachers' professed confidence in aspects of LSK and LPSK. An urgent need was expressed for effective teacher professional learning across more technical aspects of LSK and PLSK, with a particularly strong need for further support in cumulative building of knowledge (e.g. 'I'd like to have examples of how over a term expert teachers bring all the parts of the English syllabus into a cohesive program' #230).

Addressing these needs is crucial if the profession is to exploit the gains afforded by the consistent, rhetorically-oriented metalanguage of the AC:E, to continue to build a secure knowledge base for English, literacy and EAL/D teachers and to avoid the retreat to a narrowly focused, prescriptive view of grammar. We clearly have a strong profession, ready and willing to build on the (relatively) secure foundations of knowledge and know-how revealed in our national survey.

Kristina Love

Australian Catholic University

Mary Macken-Horarik and Stefan Horarik

University of New England

Note

(1.) Note that some percentages do not add up to 100% because some respondents did not provide particular information in their responses. For example, four respondents did not indicate the Australian state they worked in.

References

Andrews, R. (2005). Knowledge about the teaching of (sentence) grammar: The state of play. English Teaching: Practice and Critique, 4 (3), 69-76.

Australian Curriculum Assessment and Reporting Authority (ACARA) (2012). Australian Curriculum: English. Version 3.0. Sydney: ACARA. Retrieved from http://www.australiancurriculum.edu.au/English/Curriculum/F-10

Brumfit, C., Mitchel, R. & Hooper, R. (1996). Grammar, language and classroom practice. In M. Hughes (Ed.), Teaching and learning in changing times, (pp. 173-201). Oxford, Blackwell.

Cameron, D. (1997). Sparing the rod: what teachers need to know about grammar. Changing English: Studies in Reading and Culture, 4 (2), 229-239.

Cajkler, W. & Hislam, J. (2010). Trainee teachers' grammatical knowledge: The tension between public expectations and individual competence. Language Awareness, 11 (3), 161-177.

Callow, J. (2013). The shape of text to come: How image and text work. Sydney, Australia: Primary English Teachers Association Australia (PETAA).

Carter, R. Ed. (1990). Knowledge about Language. London: Hodder and Stoughton.

Commonwealth of Australia, (2009). The shape of the Australian Curriculum. National Curriculum Board: ACT, Australia.

Coffin, C. (2010). Language support in EAL contexts: Why systemic functional linguistics? Editorial Special Issue of NALDIC Quarterly (pp. 2-5), Reading, UK.

Derewianka, B. (2012). Knowledge about language in the Australian Curriculum: English In Australian Journal of Language and Literacy, 35 (2), 127-146.

Fearn, L. & Farnan, N. (2007). When is a verb? Using functional grammar to teach writing. Journal of Basic Writing, 26 (1), 63-87.

Freebody, P. (2007). Literacy Education in School: Research perspectives from the past for the future. Australian Council for Education Research. Camberwell, Victoria.

Glaser, B.G. (1978). Theoretical sensitivity: Advances in the methodology of grounded theory. Mill Valley, CA: Sociology Press.

Gordon, E. (2005). Grammar in New Zealand schools: Two case studies. English Teaching: Practice and Critique, 4: 48-68.

Hammond, J. & Macken-Horarik, M. (2001). Teachers' voices, teachers' practices. Australian Journal of Language and Literacy, 24 (2), 112-132.

Harper, H. & Rennie, J. (2009). 'I had to go out and get myself a book on grammar': A study of pre-service teachers' knowledge about language. Australian Journal of Language and Literacy, 33 (1), 22-37.

Hattie, J. (2009). Visible Learning: A synthesis of over 800 meta-analyses relating to achievement. London and New York, Routledge.

Hudson D. & Walmsley, J. (2005). The English Patient: English grammar and teaching in the Twentieth Century. Journal of Linguistics, 41 (3), 593-622.

Jones, P. & Chen, H. (2012). Teachers' knowledge about language: Issues of pedagogy and expertise. Australian Journal of Language and Literacy, 35 (2), 147-169.

Jones, S., Myhill, D. & Bailey, T. (2013). Grammar for writing? An investigation into the effect of Contextualised Grammar Teaching on Student Writing. Reading and Writing, 26(8), 1241-1263.

Kolln, M. & Hancock, C. (2005). The story of English grammar in United States schools. English Teaching: Practice and Critique, 4 (3), 11-31.

Louden, W., Rohl, M., Gore, J., Greaves, D., McIntosh, A., Wright, R., Siemon, D. & House, H. (2005). Prepared to teach: An investigation into the preparation of teachers to teach literacy and numeracy. Canberra, ACT.

Louden, B. (2008). 101 Damnations: the persistence of criticism and the absence of evidence about Teacher Education in Australia. Teachers and Teaching: theory and practice. 14 (4), 357-368.

Love, K., Macken-Horarik, M., Sandiford, C. & Unsworth, L. (2014). From 'bored witless' to 'rhetorical nous': Teacher orientation to knowledge about language and strengthening student persuasive writing. In English in Australia 49 (3), 43-56.

Luke, A. & Freebody, P. (1999). Further notes on the four resources model. In Reading online http://readingonline. org/research/lukefreebody.html (retrieved on 8/6/2015).

Macken-Horarik, M., Love, K. & Unsworth, L. (2011). A grammatics 'good enough' for school English in the 21st century: Four challenges in realising the potential. Australian Journal of Language and Literacy, 34 (1), 9-21.

Myhill, D. (2005). Ways of knowing: Writing with grammar in mind. English Teaching: Practice and Critique, 4 (3), 77-96.

Myhill, D., Jones, S., Lines, H. & Watson, A. (2012). Re-thinking grammar: the impact of embedded grammar teaching on students' writing and students' metalinguistic understanding. Research Papers in Education, 27(20), 139-166.

Myhill, D., Jones, S., Watson, A. & Lines, H. (2013). Playful explicitness with grammar: a pedagogy for writing, Literacy, 47 (2), 103-111.

NSW Board of Studies (1998). English K-6 North Sydney. Board of Studies NSW.

Sangster, P., Anderson, C. & O'Hara, P. (2012). Perceived and actual levels of knowledge about language amongst primary and secondary student teachers: do they know what they think they know? Language Awareness, 22 (4), 1-27.

Scardamalia, M. & Bereiter, C. (2006). Knowledge building: Theory, pedagogy, and technology. In K. Sawyer (Ed.), Cambridge Handbook of the Learning Sciences (pp. 97-118). New York: Cambridge University Press.

Unsworth, L. (2013). Re-configuring image-language relations and interpretive possibilities in picture books as animated movies: A site for developing multi-modal literacy pedagogy. lha do Desterro: A journal of English language, literature in English and cultural studies, 64, 15-47.

Williams, G. (2005). Grammatics in schools. In R. Hasan, C. Matthiessen & J. Webster (Eds.), Continuing discourse on language: A functional perspective (Vol 1 pp. 281-310). London: Equinox.

Kristina Love is Professor of Literacy Education at Australian Catholic University. As an educational linguist with a particular focus on systemic functional linguistics, she has researched and produced numerous resources related to teachers' knowledge about language and literacy in subject English and across the school disciplines.

Mary Macken-Horarik is Associate Professor of English and Mulitliteracies Education in the School of Education at the University of New England. She worked firstly as a teacher of school English but has been teaching pre-service English teachers about language since 1996. She has a special interest in systemic functional linguistics and recently completed a large Discovery project on grammar and praxis for English teachers with Kristina Love, Carmel Sandiford and Len Unsworth.

Stefan Horarik is a Research Fellow in the SiMERR National Centre at the University of New England, Armidale. He has a doctorate in social anthropology and over the past 16 years he worked on research projects in experimental psychology, sociology, anthropology and education. Table 1a. Responses by states School systems Percentages Numbers New South Wales 33.0% 123 Queensland 18.2% 68 Victoria 16.9% 63 Western Australia 13.4% 50 ACT 7.2% 27 South Australia 6.2% 23 Tasmania 4.0% 15 Table 1b. Responses by systems School systems Percentages Numbers State Schools 53.9% 201 Independent Schools 28.2% 105 Catholic Schools 15.5% 58 Table 1c. Responses by regions School regions Percentages Numbers Metropolitan Schools 63.5% 237 Regional Schools 29.5% 110 Remote Schools 3.5% 13 Table 1d. Responses by school levels School levels Percentages Numbers Early years (F-2) 16.1 60 Mid-Upper Primary (3-6) 18.5% 69 Early-Mid Secondary (7-10) 38.3% 143 Senior Secondary (11-12) 26.3% 98 Table 2. Responses rating importance of different aspects of language Aspect of language Importance Important--Extremely Important Percentages Numbers Identify and name the structural stages 77.5% 289 in various kinds of texts (e.g. narrative, exposition and text response) Name different types of sentences 63.2% 236 (e.g. simple, compound, complex) and explain grammatically how they are different Describe the grammatical components 62.2% 232 of sentences (e.g. subject, finite verb etc.) Describe and use the different aspects 68.4% 255 of cohesion in texts (e.g. referring words, word associations, ellipsis) Apply knowledge about language 86.6% 323 in composing texts Learn to analyse language choices 84.5% 315 in texts critically Understand how visual and multimodal 83.7% 312 texts work Table 3. The importance of integrating aspects of grammar into pedagogy. Importance Important-Extremely Important Aspect of grammar Percentages Numbers Integrating grammar into the 82.6% study 82.6% 308 of literature (e.g. exploring language choices in fiction or poetry Teaching students explicitly 72.6% about 72.6% 271 grammar in texts (e.g. highlighting grammatical choices in texts) Giving students grammar 29.2% exercises 29.2% 109 (e.g. from grammar resources or textbooks) Integrating grammar into 80.2% study of 80.2% 299 multimodal texts (e.g. exploring words and images in picture books or websites) Table 4. Levels of confidence in teaching knowledge about language. Aspect of language Confidence Important-Extremely Confident Percentages Number Text types and their structure 85.5% 319 Cohesive resources 72.9% 272 Sentence structure 76.1% 284 Grammar of visual and multimodal texts 76.5% 285 Overall confidence 72.7% 271