Introduction

Under-achievement in writing by significant numbers of students is widely reported internationally. In the United States the majority of students do not write well enough to meet grade level demands; by grade 4, this applies to two out of three students and, similarly, at grades 8 and 12 only 30 % of students perform at or above the “proficient level” (defined as solid academic performance) (National Center for Educational Statistics, 2012). Likewise, the United Kingdom’s Department for Education (2012) reported that amongst primary school students “writing is the subject with the worst performance compared with reading, maths and science” (p. 3). In New Zealand, the picture is similar. The most recent report of student achievement (2014) concluded that 70.6 % of all primary-age students (primary includes years 1–8 of schooling where students are aged 5–12/13 years) met or exceeded the year level National Standards expectations in writing, compared with 77 % that met or exceeded national expectations for achievement in reading and 75 % in mathematics (http://www.educationcounts.govt.nz/statistics/schooling/national-standards/National_Standards).

Writing is, arguably, still the neglected “R” (Alliance for Excellent Education, 2007; Biancarosa & Snow, 2004; Gilbert & Graham, 2010; National Commission on Writing in American Schools and Colleges, 2003), and the level of performance likely reflects this. A relatively small research base on writing instruction has grown from the 1970’s when open entry to tertiary education highlighted a lack of ability of students to write. It was at that point that educators began to recognise they had inadequate understanding and training to teach these students. The National Commission on Writing in American Schools and Colleges (2003) suggests, amongst other factors such as inadequate time given to writing and inadequate assessment, a lack of preparation of teachers to teach writing.

In reality, there is relatively little data on writing instruction generally in elementary or middle schools (Gilbert & Graham, 2010). There are a number of studies of exemplary teachers, commonly exemplary teachers of literacy-reading (e.g. Pressley, Gaskins, Solic, & Collins, 2006; Wray, Medwell, Fox, & Poulson, 2000). These have, in part, contributed to the identification, from systematic research, of effective practices in teaching writing (Graham & Perin, 2007). And, recourse to describing effective practices, in the absence of information regarding everyday practice in local classrooms, characterises the reports of governments such as the UK Department of Education’s (2012) very similar reports The research on writing and What is the research evidence on writing? Surveys reporting writing instruction at middle school and secondary school in the US present a relatively bleak picture in terms of a number of known effective practices (Applebee & Langer, 2011; Kiuhara, Graham, & Hawken, 2009; Lacina & Block, 2012). Similarly, the evaluation of the UK Every Child a Writer Project (Fisher, Myhill & Twist, 2011), in a snapshot of practice in 10 primary classrooms, showed variable practice, commenting in particular on the gap between aims and implementation and the lack of knowledge of writing of teachers in some classrooms.

Context of writing instruction in New Zealand

The general literacy achievement of New Zealand school students shows a high performance, low equity profile, unchanged since 2001. The New Zealand data from the Progress in International Reading Literacy Study (PIRLS) are characterised by high variability. Not only is the between school variability considerable but the within school variability is the highest level in the OECD. While there is solid average performance in reading with the very top students foremost internationally, there is large variation. The New Zealand school student population is increasingly linguistically and culturally diverse but particular groups, namely, indigenous Maori and also Pasifika (students from the various Pacific Island nations) are disproportionally represented in the group of lower achieving students, those who do not reach the PIRLS Intermediate International Benchmark (Chamberlain & Ministry of Education, 2013). Regarding writing, Ministry of Education data from National Standards reporting show that Maori and Pasifika students’ performance, on average, to be about 15 % below that of other students (http://www.educationcounts.govt.nz/statistics/schooling/national-standards/National_Standards). Raising literacy achievement and reducing disparity is a national strategic goal. A major policy thrust of successive New Zealand governments has been to enhance teaching practice in literacy for priority groups through quality professional learning, based on the belief that teachers have the largest single system-level impact on student achievement (Alton-Lee, 2003; Nye, Konstantopoulos, & Hedges, 2004).

The New Zealand schooling context is somewhat different to many countries. New Zealand is one educational jurisdiction approximately the size of Scotland, Norway or a small state like Vermont in the United States. Schools function autonomously with an elected Board of Trustees; these boards select their own staff, including the principal. Schools interpret and then apply the national curriculum to their context. There is no mandated testing and no national qualifications until the final 3 years of schooling (Years 11–13). Although schools until recently reported to the Ministry against the targets they themselves had set, they have been required, from 2012, to report annually performance of students in Years 1–8 of schooling against National Standards in reading, writing and mathematics (see Ministry of Education, 2009). This information is arrived at by means of an overall teacher judgement (OTJ) of each student. Schools are self-governing financially; they administer their own professional development funds so participation in any nationally offered professional learning is always voluntary as is the uptake of any nationally sanctioned assessment tools and resources.

These characteristics and the implications of them are significant in a consideration of classroom instruction; contexts shape classroom activities and student learning (Lipson, Mosenthal, Daniels, & Woodside-Jiron, 2000). The data for New Zealand from the most recent PIRLS report show a number of instances where reported practices or approaches in teaching reading differ markedly from practices of teachers in other countries (Chamberlain & Ministry of Education, 2013). And, with school autonomy and a broad national, to-be-locally contextualised curriculum, there is potential for considerable within country variation in how individual teachers and schools engage in the teaching of writing; such variability is likely to make it more difficult to generalise as to features that characterise this teaching. The aim of this article is to map the landscape of writing instruction in New Zealand. In an effort to provide insight into the nature of such instruction, the confluence of patterns that emerge from a consideration of several major sources is sought.

Important sources include the new research study presented here (“A survey of teacher practice in writing classrooms”) and other purposively selected New Zealand research (“Writing instruction through the lens of existing material”). However, the argument is that there are also indications of fore-grounded practices in instruction from the mediating layer of official documents, tools and resources provided by the Ministry and designed to support practice, as these instantiate messages about desired pedagogy in writing. The New Zealand Curriculum (2007) was fully implemented by 2010 and then schools were required to set out in their individual charters, priorities and targets for accelerating student achievement. As noted, from 2012, primary schools have been required to report annually against National Standards (Years 1–8). Accordingly, the first source of evidence considered to inform the description of instructional practice in writing is official Ministry documents, namely, the New Zealand Curriculum and the National Standards. The second is nationally sanctioned tools (in the sense that Norman (1988) used the term) and resources; the third is a meta-review of the small body of existing research that specifically focuses on writing instruction in New Zealand primary classrooms. Then, the final source is a report of the findings of a recent survey of the practices of writing teachers in New Zealand primary schools, designed to provide a wider picture than exists in current research literature and to allow, combined with other sources, the picture of writing instruction to be built.

Writing instruction through the lens of existing material

Official documents

The national curriculum (New Zealand Curriculum, Ministry of Education, 2007) is intended as a broad, guiding document to “set the direction for student learning” (p. 6) and schools are urged to adapt and implement the curriculum to meet the needs of their local context. English is one of the eight learning areas (English, the arts, health and physical education, learning languages, mathematics and statistics, science, social science, and technology). Writing is nominally within all learning areas as students learn “how to communicate knowledge and ideas in appropriate ways” (p. 16) within each learning area, but the skills for writing largely are described within the English learning area. The English curriculum focuses on meaning making within two groupings, broadly receptive and productive: listening, reading and viewing, and speaking, writing and presenting. For each grouping, processes and strategies are identified and, using these, outcomes are described for students to achieve. A broad aim is specified for each of four areas: purposes and audiences; ideas; language features, and structure and for each aim a small number of indicators for different levels of the curriculum are specified.

Importantly, in line with the curriculum policy of adapting the national curriculum to local needs, the New Zealand Curriculum positions teaching as involving a process of inquiry and teachers as inquiring practitioners (p. 35). Teachers are seen as responsible for their own learning and for inquiring into student learning and making appropriate adjustments to practice to better meet the needs of students. This is also consistent with the National Standards policy which states that the decision about the level of achievement of each student in relation to standards will be based on overall teacher judgment (OTJ).

The desired levels of performance in writing at various curriculum levels (Years 1–10) are described in the New Zealand Curriculum Reading and Writing Standards, (Ministry of Education, 2009). The standards are not standards for writing as a subject but are based on an analysis of curriculum documents in the eight learning areas that make up the national curriculum and describe the level of writing required to meet curriculum demands in each area. The emphasis is on writing (and reading) in the service of learning. The process of devising standards did not draw on evidence of students’ performance levels in writing in the curriculum areas nor did the analysis draw explicitly on any theory of writing development. The standards represent aspirational goals in two senses: they are the standards that students need to achieve to be on track to succeed in the National Certificate for Educational Achievement (Level 2) at Years 11 and 12 and, in devising the standards, there was a clear sense that they were in advance of what available normative data would suggest.

National resources and tools

With a non-prescriptive approach to how the curriculum might be implemented and with an emphasis on teaching as inquiry, it is necessary to provide teachers with supports or scaffolds, in the form of professional development and also in the form of quality tools and resources that instantiate a sound theory of the task of writing and research informed exemplars of effective practice. High quality professional learning is seen as essential, given the widespread acceptance of the view that teachers are the single most important influence on student performance. Professional development projects are offered nationally (e.g. Literacy Professional Development Project, see Parr, Timperley, Reddish, Jesson, Adams, 2007b; Timperley, Parr & Meissel, 2010) and research and development collaborations exist between universities and groups of schools (e.g. Jesson, McNaughton & Wilson, 2015). These research and development projects work with schools and teachers to build capacity, specifically evaluative capacity to support teacher inquiry and, as part of addressing student need, to build content knowledge and knowledge of effective practice. Such emphasis is also linked to a theory of development that includes the notion of differential profiles of students at same overall level (Clay (1998) talked of different paths to common outcomes) as well as culturally and linguistically diverse classrooms which require differentiation of instruction and ongoing inquiry into impact of pedagogical action/practice on student learning.

In writing pedagogy and assessment in the New Zealand context, there are, arguably, key tools that provide important messages to teachers about writing pedagogy. These are extensively used in professional learning. The first is a tool to provide diagnostic assessment of writing that also allows normative comparisons, namely, e-asTTle: Writing (revised) (NZCER and Ministry of Education, 2012). As noted, this is not a mandated assessment tool, but one of a range that schools might choose to use. In its original instantiation (Ministry of Education & the University of Auckland, 2004), an important design feature of this tool in terms of messages about writing and writing instruction, was that writing was described as serving major communicative purposes (relevant to the curriculum and context of schooling). To illustrate the features associated with different purposes for writing, detailed scoring rubrics specified, for seven different dimensions of writing (audience, content, structure, language resources, grammar, spelling and punctuation), the criteria commonly associated with each purpose at various curriculum levels of primary schooling. The design of the rubrics thus reinforced the idea that development may be patterned differently across students whose writing might reflect the same overall level. Annotated exemplars illustrated scoring of criteria and different curriculum levels. From this original development, teachers reported learning a great deal about how language works to achieve different communicative purposes in writing from working with the tool (Parr, Glasswell, Aikman, 2007a). The revision retains the diagnostic focus and this, together with the fact that teachers can select when to utilise the assessment tool, reinforce the idea of assessment as an integral, ongoing part of learning and teaching, rather than for accountability purposes. Commonly, results from the use of asTTle form part of a teacher’s overall judgement of whether a student has met national standards.

A further tool is provided to help view progress in writing, the English Literacy Learning Progressions or LLP (Ministry of Education, 2010). It describes and illustrates writing behaviours and features of writing characteristic of particular levels. These descriptions cue teachers as to what to notice as students write, as well as what to attend to in the pieces produced. In a study shortly after the LLP were introduced, teachers in a national professional development project in literacy reported finding the tool useful in providing a bigger picture of development in writing; they reported learning in relation to expectations, pedagogical aspects, and the links between writing and reading (Parr, 2011). Actions resulting from such learning were reportedly taken with respect to setting goals or targets; in terms of using the LLP to build profiles of students but also using LLP to reflect on aspects of practice (Parr, 2011). A new tool, the Progress and Consistency Tool (PACT), designed to support teachers to make dependable judgements about students’ progress and achievement in relation to the standards, is now added to available resources. It contains clearly annotated examples of writing and each has been allocated an appropriate standard level with justification.

While the assessment tools and the progression descriptions are largely focussed on identifying student outcomes, actions and behaviours, more general advice to inform pedagogy is provided in the form of research-based literacy handbooks for teachers, limited numbers of which are provided to schools free of charge [Effective Literacy Practice Years 1–4, (2006) and Effective Literacy Practice Years 5–8, (2007)]. The handbooks are widely used in professional development. They synthesise international research in literacy drawing from quality-assured published sources; they are non-prescriptive, presenting a balanced view of different theoretical emphases, for example in the teaching of reading. In the book covering the early years, approaches to writing are described in terms of language experience activities (drawing on the work of past exemplary New Zealand educators such as Sylvia Ashton Warner) which make explicit the links between spoken and written language. Described are practices such as shared writing and a variant, interactive writing, which are supportive instructional settings involving the teacher and a group of students- often the whole class- in the construction of a common text. Such may involve teacher modelling of the process and questioning or musing to encourage student participation; guided writing (practices similar to guided reading) where the student progressively takes control of the writing process, and independent writing. Another section dealing with creating texts describes the writing process and writing strategies, focusing on what learners do. The book for teachers of older writers contains the same messages about writing as a purposeful social communicative activity. Theoretically the handbooks draw on cognitive, socio-cognitive and socio-cultural views of writing but with an emphasis on the latter. The handbook for teachers of older primary writers describes how teachers make use of instructional strategies (the “tools of effective practice”; the “deliberate acts of teaching that focus on learning in order to meet a particular purpose” (p. 11)), within a range of approaches and a balanced programme. In both books the reciprocal nature of reading and writing is stressed as is the use and creation of a range of texts.

Overall, the official documents and national tools and resources establish broad parameters and provide guidance. They are intended to be interpreted and applied in a way consistent with the local instructional context and to support teacher inquiry into, and adjustment of, practice. In devising the curriculum and national tools, widespread consultation is undertaken with all levels of the education profession (professional associations, unions, teacher educators, researchers and those in schools). Generally, teacher groups provide the greatest input, with trial draft versions implemented and feedback sought to inform subsequent iterations. The approach is consistent with views of teachers as reflective practitioners, adaptive professionals able to engage in ongoing learning about and adjustment of their practice.

Previous research on writing in primary schools in New Zealand

Further information about practices in teaching writing in New Zealand at primary school level is drawn from the limited research studies available. Here we present a meta-review of selected studies, summarised in tabular form. In order to select studies for inclusion, we specified criteria, namely, that the research:

  • focus on writing instruction (not literacy more generally where reading is almost always the focus, as in studies internationally).

  • relate to the year levels encompassed by primary schooling in NZ (Years 1–8 of schooling, ages 5–12/13).

  • describe empirical work or is a scholarly review of instruction in NZ

  • is published in journals or in official technical reports or doctoral theses available in the public domain.

  • is recent (2000 on)

Searches were made of general databases (ERIC, Psych Lit) and of New Zealand specific databases (NZCER Journals Online, Index New Zealand), using principally the keywords “writing instruction”, “writing pedagogy” “elementary” or “primary school” and “New Zealand”. Resulting articles were read to ascertain whether they met the criteria above. The selected research was considered and coded in terms of several variables (which comprise the columns of Table 1). These allow the research to be situated; the overall design and key method details to be described; the focus in terms of writing instruction to be identified and the findings to be summarised. Each author took responsibility to review and code half of the studies and this analysis was cross-checked by the other author. The variables used to summarise were: the context in which it was conducted, that is whether it was an intervention, part of a professional development project or a description of practice to be viewed in relation to other features (like student learning or the nature of PD) and, if an intervention, whether it was primarily aimed at students or teachers or whether an interactive relationship was envisaged; the overall design or approach of the study; the foci within writing practice; details of the sample- its size and nature; the measures and outcomes measured, the main findings and the implications about pedagogy in New Zealand. In total 13 studies were selected and reviewed. The compilation of reviewed studies is presented in Table 1.

Table 1 Summary of studies of writing instruction in New Zealand
Full size table

A number of recurrent foci emerge from the combined studies. Glasswell’s (2000), and Glasswell, Parr and McNaughton (2003a, b) descriptions of effective teachers and Jesson and Cockle’s (2014) descriptions of teachers pre-intervention identify some commonly recognisable classroom routines and structures: whole class modelling, including the reading and exploring of text often as part of shared writing, followed by independent or guided writing for small groups, with in-task teacher support of writers through feedback within writing conferences. Within these dominant structures, the difficulty of enacting pedagogy in ways that best promote equitable outcomes for learners is also identified. Glasswell (2000), Glasswell et al. (2003a, b) identifies interaction patterns, even in the most effective teachers’ classes, that constrain teachers’ attempts to provide high quality instruction for every writer. Similarly, Jesson and Cockle (2014) identify the typical lesson pattern as one which constrains teachers’ and students’ ability to incorporate diverse learners’ existing textual and communicative repertoires of expertise.

A dominant theme of both intervention studies and other studies of effective teachers is investigation of strategies to promote more effective assessment for learning processes and feedback for students about their writing. Most of these studies have been conducted within the broad parameters of working with teachers to enhance their practice through professional learning. The emphasis on assessment for learning sits readily with the tradition of teaching the individual student and the focus on promoting, from the earliest years, independent learning behaviours, reflected most notably in the works of Marie Clay (1977, 2010). The strong relationship between the ability of a teacher to give high quality feedback, defined in terms of features designed to promote self regulatory behaviours (providing information on performance, that is extent to which student has met learning goals for the task; what the criteria or requirements are for a quality performance; what the student needs to do to progress and how the student might go about improving) and students’ progress in writing was demonstrated statistically in a study by Parr and Timperley (2010). For Ward and Dix (2004) and Dix and Cawkwell (2011), in their small-scale, qualitative studies, peer response groups, supported by teacher demonstration, were seen to offer opportunities for students to give and receive more effective guidance from each other. Similarly, multi-method studies have identified the value in relation to student understanding of clear alignment between learning goals and their instantiation in terms of quality writing, lesson activities and feedback for students (Parr & Limbrick, 2010; Timperley & Parr, 2009). In the study by Gadd (2014) of a small group of exemplary teachers, what differentiated those most successful in fostering accelerated progress in writing was the ability to draw students into participating in their own learning through practices such as student input as to the nature of writing tasks; through co-constructing learning goals and criteria for success and through supporting students to self monitor and self regulate. These practices yielded the greatest variation in performance even amongst exemplary teachers.

A minor strand of research identifies links in teacher practice between reading and writing (Jesson, 2010; Jesson, McNaughton, & Parr, 2011; Parr & McNaughton, 2014) as avenues for building student textual knowledge and also writing strategies. Ways in which teachers make such links between texts within writing lessons, as described by Jesson et al. (2011), are by encouraging identification of the authors’ craft through text analysis, creation of shared texts as tools (e.g. signs, checklists, charts), the use of multiple texts as models and provision of opportunities to discuss texts in small groups.

Amongst the small number of studies of writing instruction in New Zealand classrooms, a number have been conducted within professional learning projects, most of which operate broadly within a research and development tradition, to investigate teacher learning and its translation to practice. Evaluation of practice is routinely embedded in the building of inquiry skills in teachers. Other studies have been specifically designed to identify effective practices, specifically from the practices of those teachers whose students make accelerated progress in writing. This focus reflects, in part, an effort to identify what works for particular groups of students. A consideration of the focus dimension of the analysis shows the foregrounding of practices related to assessment for learning and to differentiation. This would seem to be consistent with a policy context of adaptation of curriculum and pedagogy to meet the needs of students in the local context.

A survey of teacher practice in writing classrooms

In the introduction, the point was made that, internationally, a limited amount is known about writing instruction in local classrooms. This survey was designed to address the broad question “What are major characteristics of practice of New Zealand primary teachers?” and, given that these may vary with year levels, an additional question asks “and do patterns of practice differ across year levels?”

Method

Design of study

An on-line survey was employed to endeavour to gather data nationally from a stratified sample of teachers of students in Years 1–8. The survey drew items from Cutler and Graham’s survey (2008), while other questions were designed to capture particular New Zealand characteristics of writing instruction. For example the emphasis in curriculum and assessment tools on the social communicative purposes for writing meant that we asked separately about this aspect of instruction rather than confounding it, as in other surveys, with forms of writing. Given national policy that judgements of achievement be based on overall teacher judgement, we also emphasised acts related to ongoing, assessment for learning. Appropriate institutional ethical approval was obtained for the study.

Participants

Selection of participants was at a school level as there is no centralised database of teachers working in schools available in New Zealand (The Ministry of Education funds individual schools through an operational grant and teachers’ salaries are paid through a private data-handling agency). The sampling frame was all 2106 state funded primary schools; stratified first by sorting into urban (main, minor and secondary urban) and rural categories and determining proportions in each. Urban schools were further stratified into the three types of primary schools: full primary (40 %), contributing (46 %) and intermediates (14 %). Then, a national sample consisting of about 15 % of schools receiving state funding (n = 316) was selected. The response rate was low at around 13 % of schools (41 schools, 118 teachers representing around 3000 students). A check of the resulting sample showed it to be approximately similar in proportions of school types with a mix of urban and rural schools. While it has to be acknowledged that this small sample cannot be viewed as representative, the argument advanced earlier was that there is likely to be considerable variability in practice, given a broad curriculum which is to be adapted to local conditions and given the large variability in student performance both across and within schools. This was the rationale for employing the three sources of evidence of which the survey is one.

Survey instrument

The instrument was divided into sections that sought information about the teacher and his/her beliefs, including confidence about teaching aspects of writing, and about classroom practice. Questions (1–10) in the first section obtained demographic and background information about teacher and class, including a rating on a six point scale of how well prepared the respondent feels, from pre-service and in-service training, to teach writing. The questions about class included how many students, how many English as another language (EAL) students and the estimated writing achievement of the current class in terms of proportions in each of five categories, similar in format to those specified by National Standards (well below, slightly below, at, slightly above and well above). Question 11 asked about use of commercial resources to teach any aspect of writing, to name the resource(s) and to rate its importance in the overall writing programme on a 4 point scale.

Subsequent questions asked about actual practices. Question 12 asked which functions or purposes for writing (e.g. writing to persuade, to describe, classify, organise and report information; to compare and contrast; to explain etc.) the teacher intended to cover in most depth in the current year (select 4 from a list of 10 plus the option of “other”). For those purposes selected, the forms of writing commonly associated with that purpose appeared and teachers were asked to indicate which forms the students would likely engage with (e.g. correspondence, newspaper articles, posters/signs). Question 13 involved rating the importance in the teacher’s practice (on a 5 point scale) of nine practices, the majority related to assessment. The next section dealt with teaching approaches or ‘moves’ and, in question 14, teachers were asked to rate the extent to which they emphasised 14 teaching moves that characterise the teaching of writing (e.g. deconstructing or discussing a text, teaching to build vocabulary, teaching strategies for spelling unknown words, goal setting with students or communicating learning intentions). The next section (Question 15) asked about efficacy; teachers rated (on a 6 point scale) the extent to which they felt confident in terms of the same 14 actions they rated for emphasis in the previous question. In addition, five items asked about their own confidence as a writer; their knowledge of out of school literacy practices; their confidence in identifying student learning needs in writing; their knowledge of features of text in relation to purpose and their confidence in teaching a range of writing types and purposes.

Teachers were asked in the final section to estimate time: total writing time in a week, including planning, revising and editing/publishing (Q. 16); what amount of time was spent writing in specific writing time and what in other areas of the curriculum. Then (Q. 17) they were asked to estimate time spent on 13 teaching moves, virtually the same moves asked about in question 14. To enable comparisons, given the variable amount of time teachers may spend on writing, an average week was given as 150 min (this was an informed estimate of the average time primary teachers would spend on writing in a week).

Procedure

Emails were sent to the principals of the selected sample of schools (a requirement of the ethical procedures of our institution not to approach teachers directly and also to seek approval from their institution). Principals were asked, if they approved, to forward the questionnaires and participant information sheets to their classroom teachers. The information sheet explained who we were and that we were conducting the survey to find out about the teaching of writing in primary schools. The teachers could then decide whether to participate. Participation was anonymous; only the school was identified. We have no way of knowing how many principals simply did not forward the invitation to their teachers or how many teachers received it and declined to participate. A reminder email was sent to the schools after a fortnight.

Analysis

The first set of data describes the teachers and their students and for this descriptive statistics are used. To investigate any differences in practices by year level grouping, a one way analysis of variance examined differences in time reportedly given to writing across year groups while multivariate analyses of variance tested whether teachers responded differently by year grouping to items within questions dealing with instructional practices. Post-hoc comparisons (Bonferroni) were used to identify where any differences lay. Four groupings of year levels were identified (1–2, 3–4, 5–6 and 7–8). Those teachers who taught a range of levels that did not fit these categories (e.g. taught Years 1–8 in a composite class) were excluded from these particular analyses (there were seven instances of such groupings).

Findings

Demographics, beliefs about preparation and class details

The teachers, as a group, were relatively new to teaching (mean years of service = 3.55 SD = 2.4, range less than a year to 11 years), in comparison to the most recent PIRLS data (Chamberlain & Ministry of Education, 2013) where the average years of teaching of the Year 5 teachers involved was 11 years. The sample’s qualifications were predominantly a Bachelor of Education or Diploma (54 and 15 % respectively, the same as the 69 % from the PIRLS data holding such qualifications). Almost all reported participating in further in-service learning in writing. Asked about building their preparation to teach writing, overwhelmingly, in-service learning was valued with 62 % reporting in-service professional development had prepared them well or extensively. Conversely, around 70 % rated their pre-service preparation regarding teaching writing as non-existent, or minimally or somewhat helpful.

The teachers were asked what level they were currently teaching. This is because, in NZ, primary teachers do not necessarily specialise in a year level and may teach different levels at various times. The picture is even more complicated as it is common in smaller schools to have composite classes. While these mostly involve two year levels in a common configuration, some may involve more; at the extreme in very small schools a single teacher teaches Years 1–8. The current teaching level of respondents indicated they came from all levels but with around 60 % from years 1–4. The average class size, as would be expected increased with level, averaging around 28 at Year 8, while composite classes were smaller on average, with 19 students (again as might be expected given such characterise smaller, rural schools). There was considerable variability in the number of students in class who reportedly had English as another language. This ranged from zero in around a quarter of the classrooms to 28; 70 % of classes had fewer than six such students although 10 % of classes reported more than 18 EAL students, the latter likely a function of the concentration of immigrants in particular regions. The reported achievement of students against the appropriate national standard quite closely mirrored the national picture with more students reportedly achieving the standard in the early years with only 33 % reported as below or well below, compared to Years 7 and 8 where 58 % were seen to be achieving below standard.

Confidence to teach writing

On average, teachers were confident in the teaching practices nominated (although for each item the full range of the scale was employed: 1 = strongly not confident to 6 = extremely confident). Teachers were reportedly most confident about providing feedback to writers (M = 5.18, SD = 0.98) but they were also they were confident of their ability to monitor using rove and assist (this term describes teacher actions that include moving around the class as they write, noticing instances where a comment or question or prompt might move the writer forward and also responding to students who seek assistance) (M = 5.03, SD = 0.99); of their ability to model writing (M = 5.01, SD = 0.91) and of their ability to identify learning needs (M = 5.0, SD = 0.91). Teachers were least confident of their knowledge of the out-of-school writing practices of students (M = 3.97, SD = 1.09) which was the item the lowest rated by a reasonable margin. All other practices were rated moderately (means between 4.5 and 4.99) regarding confidence to implement. Interestingly, teachers were confident that they wrote well (M = 4.94, SD = 1.00).

Time and place of writing instruction

There were three questions about time spent writing. One question asked for an estimate of minutes students spend on writing in an average week. The mean was 297.24 min (4.95 h) but the variability was relatively high (SD = 179.19 min). There was an effect of year level (F = 4.66, p < .01); the time spent increased with year level (244–413 min) with years 7–8 significantly higher than Years 1–2 (F = 4.67, p < .01). The time spent writing in specific writing sessions within a week averaged 170.53 min (SD = 87.21) and this figure was reasonably similar across year levels; there was no effect of year level. However, as a proportion of total writing time, progressively, writing instruction was reported to occur at a time not specifically designated. In junior classes (Years 1–2), the majority (64 %) of writing happened in writing time whereas only 43 % occurred here in Years 7–8. Minutes spent writing in other curriculum areas across a week averaged 126.7 but the variability was considerable (SD = 126.04). The time spent increased with year level from 88 min at Year 1–2 to 234 min in years 7–8. There was an effect of year (F = 7.35, p < .01) with Years 7–8 spending more time writing in other curriculum areas than any other grouping.

Resources

When asked about commercial resources, the majority of respondents (56 %) named none. There was a reasonable variety named by those who did consider they used a resource. Some named commercial sets like PM Writing (http://cengage.co.nz/primary/browse-series/pm/pm-writing) (n = 10) or First Steps (http://learningstair.co.nz) (n = 2), produced in NZ and Australia, respectively, while two teachers nominated “teaching primary writing” by Calkins (it is unclear exactly which resource by Calkins is referred to- see www.heinemann.com). Other resources they named largely dealt with specific aspects of writing like a locally produced resource for spelling or a phonics or vocabulary resource (each of these types named by only one teacher). Greater numbers, however, named professional readings, including the Ministry produced handbooks mentioned above (n = 15) or Ministry of Education on-line resources and tools like the Learning Progressions or assessment tools like e-asTTle: Writing (n = 8) which would not technically be considered commercial resources. For those who reported using resources, the junior school teachers (1–3) largely considered them (an even split) moderately important or important. By the end of primary the resources used were considered of greater importance by those employing them.

Range of writing purposes covered in current year

Ten major functions of, or purposes for, writing that are curriculum referenced were listed in this question: writing about cause and effect/explain; writing to compare/contrast; writing to creative or express/narrative; writing to analyse or critique; writing to demonstrate learning in a content area; writing to describe/classify/organise and report information; writing to persuade; writing as personal response to material read or discussed; writing to recount, and writing to instruct. Teachers were asked to indicate which four of these purposes they would most likely cover in more depth in the current year. Table 2 shows the percentage of teachers who identified each purpose as one of the four most important that they would cover in the current year. The bold indicates the three most often chosen at each year level.

Table 2 Percentage of teachers selecting writing purpose as an emphasis by year groupings
Full size table

The most nominated across all year groups were creative narrative and writing to recount, then to report. There were age-related trends: by the end of primary recount had declined and reports and other purposes aligned with content area writing, like writing to explain and to persuade had increased. There was a reasonable degree of consistency in nominations at each year group level. At Years 1–2, a wide range of purposes was nominated across the respondents. The most popular choices, however, were recount and narrative/creative writing. The third most nominated was writing to describe, classify, organise in order to report. At Years 3–4 narrating and recounting remain foremost followed by instructing. By Years 5–6, writing to recount and to narrate predominate, however, writing to persuade appears. Writing to explain cause and effect also features. At Years 7–8 the most frequently taught purpose was again writing to narrate, followed by writing to report, writing to persuade and writing to demonstrate learning in content areas. Predictably, given student level, purposes that barely featured in nominations were to compare and contrast and writing to analyse.

When a respondent selected a purpose, common examples of the possible form of that function appeared on screen and the teacher again selected up to four that/he would use. So, if narration were a purpose selected, then the following options appeared: comic strips, correspondence (email, letters etc.), legends/fables/fairy tales etc., plays, poems, stories, wall stories and “other—please provide details”. Over all, in writing to recount, the major forms were telling a personal narrative, retelling and a diary entry or blog. Writing to narrate mainly took the form of stories, legends and fables, and poems; personal response was in the form of letters or reviews while classifying and organising information to report involved note taking, lists, picture captions and profiles. The function of writing to persuade, not nominated much till Years 5–6, took the form principally of letters, speeches, advertisements and posters while writing to demonstrate learning involved topic reports, summaries and worksheets at higher year levels and wall stories and lists as well as topic reports at more junior levels.

Instructional practices

In this section teachers were asked about emphasis and also about proportion of time spent on teaching actions. They rated (on a scale of 1–5 where 1 = to a very great extent and 5 = to a very small extent) a series of teaching moves (14) in terms of their emphasis in their writing instruction. These included deliberate teaching moves, namely, deconstructing a text, modelling the writing process, building vocabulary, teaching sentence combining or grammar, teaching punctuation, teaching spelling through rules and patterns, teaching spelling strategies, goal-setting or communicating learning intentions, providing feedback, monitoring through ‘rove and assist’, using small group approaches to writing (such as interactive, guided or shared writing), teaching planning strategies, facilitating content generation (brainstorming/reading material) and facilitating Language Experience activities. The mean ratings, by year level are shown in Table 3.

Table 3 Ratings of emphasis placed on instructional practices by year level
Full size table

Overwhelmingly, teachers reported placing most emphasis on providing feedback (M = 1.44, SD = 0.53). In total, 98 % of teachers reported that they emphasised this to a great or very great extent. Modelling the writing process was strongly emphasised (M = 1.63, SD = 0.64) with 91 % selecting to a great or very great extent. Other moves emphasised were goal setting with students (M = 1.83, SD = 0.79) and using small group approaches such as interactive, guided and shared writing (M = 1.95, SD = 0.92). Also rated close to 2 (to a great extent) were monitoring through rove and assist and vocabulary building (75 and 74 %, respectively, chose to a great or very great extent). The least emphasis was placed on teaching spelling through rules and patterns and teaching strategies for spelling unknown words. The greatest difference between year levels was seen in the extent to which teachers placed emphasis on teaching strategies to spell unknown words (F = 5.88, p < .001). Whereas 73 % of teachers from Year 1–2 emphasised this to a great or very great extent, in other year levels, this received less emphasis.

Reported time spent on each of these moves reflected an increasing focus on small group approaches as year levels increase. Overall, the teaching moves which participants reported spending the greatest amount of time involved in (in minutes) across a week were deconstructing texts with small groups (M = 50.53, SD = 48.80); using small group instructional approaches such as shared, guided and interactive writing (M = 51.98, SD = 47.87); roving and assisting students as they write (M = 46.23, SD = 45.79); modelling writing processes (M = 45.23, SD = 45.66); and feedback to students (M = 43.73, SD = 37.94). The teaching moves that reportedly were allocated the least estimated amount of time were goal setting (logical as it does not take long in a lesson or a goal may span several lessons), teaching punctuation and teaching spelling rules, teaching sentence combining or grammar and brainstorming for content generation. Each of these, on average, was reported to consume between 20 and 25 min across a week. There were differences between year levels, however, with the predominance of small group approaches not apparent in Years 1–2. Monitoring using rove and assist was reported as the activity taking the most teacher time for both Years 1–2 and Years 3–4, emphasising the teaching of the individual student at these levels.

Importance of specific dimensions of practice

These dimensions were providing written feedback to students, providing oral feedback to students, engaging in ongoing interaction with students about their writing, facilitating students sharing of writing, facilitating peer response to writing, facilitating publishing of student work, incorporating cultural and linguistic diversity and diagnostic writing assessment. Ratings of these dimensions (from 1 = very important to 5 = not very important), suggested, unsurprisingly, that all dimensions were considered important. Highest ratings were given to providing oral feedback to students (cumulative percentage of very important and important = 99.2 %; M = 1.14, SD = 0.38) and engaging in ongoing interaction with students about their writing (cumulative percentage of very important and important = 98.4 %; M = 1.22, SD = 0.46). The lowest rated item concerned incorporating cultural and linguistic diversity (M = 2.26, with 66.7 % regarding this as important or very important). A similarly lower rated practice was providing written feedback to students (M = 2.20, SD = 1.07). The relativity and difference across year levels provides more information. Tests of between subject effects showed providing written feedback (F = 4.66, p < .01) and two others: facilitating sharing of writing (F = 2.73, p < .05) and facilitating peer responses to student writing (F = 2.20, p < .05) to be differentially rated across year groups. Multiple comparisons showed Years 1–2 teachers, understandably, rated providing written feedback lower, consistent with the abilities of their students. The patterns for the other two items are less clear. There were no significant differences shown in multiple comparisons across groups for facilitating sharing. Regarding peer response, Years 7–8 teachers considered it more important than Years 5–6 teachers.

Discussion of survey findings

We discuss the survey findings in terms of their relationship to other New Zealand research and briefly comment, where appropriate, on the relationship of the survey to other international research such as the US data from a similar number of teachers.

Our teachers did not feel well-prepared to teach writing from their pre-service training. It may well be that this is in marked contrast to the time and training they have in the teaching of reading but there is little other than anecdotal evidence to support this. While the US National Commission on Writing (2003) comments on a lack of preparation, data from a representative sample of Grades 1–3 teachers in Cutler and Graham’s (2008) survey suggests that 70 % of these teachers feel adequately prepared or better through their teacher education programme to teach writing. In contrast to this, our teachers’ confidence in writing instruction was reportedly a result of their in-service learning, likely reflecting increasing emphasis on writing within large scale professional development projects (Parr et al. 2007a, b). The Literacy Professional Development Project achieved large effect size gains in writing (over normative expectations) and this was replicated over three different cohorts of schools (see Meissel, 2014; Timperley, Parr & Meissel, 2010), suggesting teachers’ views that in-service professional learning is significant in their writing practice, is well founded.

Time reportedly spent on writing represents opportunity to engage with writing, through instruction and by practicing the craft. Time spent on writing instruction, or the lack of it, is something commented on by The National Commission on Writing in American Schools and Colleges (2003). But, as Mullis, Martin, Minnich, Drucker and Ragan (2012) state, in reference to the PIRLS data, it is often difficult to examine the direct effect of instructional time on achievement as the quality of the curriculum and of the instruction is important. The only other data available, the PIRLS data, suggests that, in New Zealand at Year 5, about 37 % of total instructional time is spent on language related activities, namely, reading, writing, speaking and other language-related skills. Of this, about a quarter of the time is devoted to reading, ranking NZ fourth in terms of number of hours spent teaching reading. This appears to be consistent with the 4.9 h a week hours spent in writing reported in the survey, bearing in mind that the teaching of writing is often interwoven with reading and, in later primary particularly, accomplished also within teaching in the content areas. By contrast, students in the US reportedly spend relatively little time writing. In Grades 1–3, considerable variability was noted in the time spent per week writing: 0–380 min with a median of 105 min (Cutler & Graham, 2008). In middle and high school only about 7.7 % of class time is spent writing in any extended way (Applebee & Langer, 2011). However, the notion of time spent writing is problematic as the time can be higher depending on the definition of writing. It appears that US students actually do little composing; writing without composing is common in writing activities especially by secondary level (Kiuhara, Graham & Hawken, 2009).

The data from the survey of New Zealand teachers support the notion of writing taking place increasingly in a time not specifically set aside for writing. Although writing is largely described within the subject English in the New Zealand Curriculum, the National Standards and the Literacy Learning Progressions describe the competencies students need in writing to succeed in all curriculum areas. The survey responses reflect a predominance of recount and narrative writing in a time specified as writing time in junior classes, gradually moving toward non-fiction purposes, and also gradually moving to more time spent writing in other curriculum areas. Writing to demonstrate learning also increases as age levels increase, with reports, summaries and worksheets starting to figure more prominently in Years 7–8.

Regarding how teachers report they organise for instruction in writing, it is interesting to note that while teachers placed considerable emphasis on small group approaches in writing, and reportedly spent over 150 min in a week on deconstructing text, modelling, or using shared, guided or interactive writing techniques with small groups, they are only moderately confident in using these small group approaches. Small group approaches are prevalent in New Zealand for reading. Indeed, the PIRLS reading data suggest that New Zealand teachers report rarely teaching reading to a whole class group (only 12 % respond always or almost always, compared to an international average of 83 %). Our data indicate that the shift toward small group approaches begins for writing in Years 3–4, but only from Years 5–6 on does the time spent in small group approaches begin to surpass the time spend in supporting individuals (rove and assist). It seems likely that this pattern reflects increasing student independence in writing.

Similarly, other patterns reported such as the low use of resources and also the relatively minor emphasis on teaching basic skills suggest that, as in reading, the instructional practices of New Zealand teachers in writing may differ from those reported elsewhere, for example by Cutler and Graham (2008). In the case where modelling is reported to be strongly emphasised (and this is not necessarily the case elsewhere as the work of Puranik, Al Otaiba, Sidler, and Greulich (2014) suggest) it may be that NZ teachers generally think of modelling as including using text as a model, reading, deconstructing and reflecting on the text in relation to the current writing. Other New Zealand studies (e.g. Gadd, 2014; Glasswell, 2000) show modelling in the broad sense to be a key practice and one linked to reading (Parr & McNaughton, 2014).

Considerable emphasis was reportedly given to practices linked to assessment for learning, for example making clear the purpose and aims of lessons and the quality of writing aimed for; providing feedback to bridge the gap between current level and desired level of performance and including the students as active agents in their own learning through, for example, the setting of individual goals for learning in writing and the self monitoring of progress towards those goals. The teachers in our survey reported greatest confidence in providing feedback, identifying student needs and providing in-task support while students write (rove and assist). These data confirm the reported dominant practices in writing classes from both studies of those nominated as effective teachers (Glasswell, 2000, Glasswell et al. 2003a, b) as well as general teaching (Jesson & Cockle, 2014).

Such practices, particularly feedback, figure prominently, both in our survey and in our review of New Zealand research. The results from the survey may provide some evidence that large scale professional development foci, for example Assess to Learn (Poskitt & Taylor, 2008) and the Literacy Professional Development Project (Parr et al. 2007a, b; Timperley et al. 2010) are having an impact on teachers’ confidence in these areas, and also on teachers’ emphasis on such practices.

General discussion

Given the unique nature of the New Zealand educational system, classroom practice is likely both to be constructed, and to vary, in ways unlike that of more centralised systems. There are indications of differing emphases in writing instruction that may be a function of the context. The description presented draws from the survey of classroom teachers, with commonalities sought from policy documents, tools and research studies of classroom practice. Such a synthesis provides opportunity to investigate the extent to which policy initiatives and research-to-date align with the reported everyday classroom practice; how aspects of the context might shape classroom practice. In particular, two key policy concerns warrant consideration in the light of these data about writing instruction. The first is the influence of the framing of teaching as inquiry within the New Zealand Curriculum and the second is how current practices relate to the major policy goal of reducing inequity in literacy achievement.

The inquiry framing of the curriculum is consistent with a longstanding New Zealand practice (emphasised by eminent New Zealand educators like Sylvia Ashton-Warner and Marie Clay) of teaching to the needs of the individual. Knowing the needs of the individual student guides teaching and synergies with the assessment for learning movement are clear. Recent increased emphasis on outcomes for priority learners and in terms of meeting national standards has subtly shifted this identification of need into priorities for instruction and powerful practices in instruction. The results are evident in the focus on formative assessment in policy documents; in the focus on evidence-informed inquiry in professional learning; in descriptions of effective practice; in terms of foci for interventions; in existing New Zealand research and in the tenor of the responses of teachers in the survey.

The evidence of the influence of the other major policy plank, equitable outcomes, is less clear. In the survey, teachers were less confident in their knowledge of out of school writing practices and gave a relatively low rating to the importance of incorporating cultural and linguistic diversity. It is likely that they are weak in identifying and utilising the existing expertise of diverse learners and recent research regarding Pasifika learners and literacy would support this (Si’ilata, 2014). However, as Si’ilata showed, they respond to support to make their practice more culturally responsive. Given the patterning of achievement in writing (and literacy more generally) this suggests the need to nuance teacher inquiry to investigate not just students’ identified needs based on assessment data but to seek more explicitly to identify and utilise students’ diverse areas of expertise. In terms of inquiry into their own practices there is also a need to investigate (as for example, Glasswell (2000), Glasswell et al. (2003a, b) did) the proximal processes that produce differential effects for different groups of learners but in ways that might support equity of outcomes for non-dominant groups. The conclusion from our mapping of the landscape of instruction in New Zealand classrooms is that research and research and development endeavours actively pursue these two avenues.