In press: F. Christie (ed.) Curriculum Issues (Australian Curriculum Studies Association), 1998

Abstract

Australia has been an international leader in curriculum development and reform in language and literacy education since the 1970s. Successive innovations have included: whole language, process writing, genre approaches, critical literacy, and media and cultural studies. Each of these has generated innovative assessment practices, many of which have been implemented in other countries. The advent of national literacy benchmarks marks a move towards an assessment-driven approach to curriculum. This article situates the development of literacy benchmarks against a background of curriculum change in literacy. It makes the case that the risk is that the substantial innovations, achievements and expertise in Australian literacy education may be subordinated to a test-driven, basic skills approach to literacy learning.

Introduction

Will the implementation of national standards, and with this, large scale testing systems improve students’ literacy? At the heart of MCEETYA and DEETYA’s calls for national student performance benchmarks are the beliefs that

(1) articulated and consistent national standards and achievement criteria will lead to increased government and school accountability to key ‘stakeholders’ – parents, communities and governments - and that

(2) a national system of coordinated state-level standardised achievement tests (and perhaps eventually a single national test) guided by these benchmarks will enable the improvement of the quality of teaching and student achievement in literacy.

Few curriculum developers and educational researchers would disagree with the need for interstate curricular coherence, public accountability and systems of assessment. Our view is that the descriptions of literacy curriculum benchmarks to date are, at best, problematic in their capacity to encompass the broad range of curriculum reforms undertaken by Australian educators. At the same time, there is little international evidence that national or state-wide testing systems per se will lead to improvements in instructional effectiveness, innovation or student achievement. Our case here is that the translation of the former into the latter – benchmarks into testing – puts at risk innovations in instruction, curriculum and assessment built up by Australian teachers, researchers and curriculum developers over the past three decades.

Assessment practices have histories

The techniques and instruments of educational assessment have traditions and uses, histories and abuses. The development and implementation of mass scale standardised achievement testing is part of a long historical legacy in the United States, having begun in the 1910s with the development and implementation of literacy tests for World War I recruitment. Within two decades of the introduction of testing in the 1920s, virtually all American children were tested. Standardised, norm-referenced achievement tests were a lynch pin in what was sold to administrators and communities as "factory" or "assembly-line" approaches to instruction, with test scores used for sorting and categorising children, assessing teachers and school districts’ effectiveness, assessing and developing curriculum, and public accountability (Callahan, 1962). In this regard, what Matthew Arnold once cynically termed "payment by results" – decision making about funding, about the disposition and support of particular communities of students, about the efficacy of particular schools and teachers - is hardly a new concept (Luke & DeCastell, 1985). Nor is it surprising that almost all American educational debates - from arguments over class size and school district funding levels, charter schools and vouchers, to debates between phonics and whole language advocates - are fought by ideological opponents armed to the teeth with contending standardised test data. For the last century the ‘test’ has become the arbiter of US educational policy and practice. And at this key juncture in the development of Australian educational policy, it is worth asking why and how the American educational system has persistent and deep-seated literacy problems despite repeated years of testing.

The traditions of educational assessment in Australia have followed a very different historical pathway. They are continuing to evolve from the traditional British Examination system – a secondary curriculum-based, connoisseur model of assessment that exists today in modified form in some states – towards integrated systems of student work profiling with across-school moderation systems. In the primary school system, the initial use of standardised tests by Fred Schonell and colleagues in the 1950s focussed on the individual diagnosis of students’ learning problems for specialised interventions. In the 1980s and 1990s, successive state and federal governments supported the development of national, discipline and profession-specific competency scales for post-secondary and adult education, and the development and implementation of diverse qualitative and quantitative approaches to student assessment for school-aged children. The Australian educational system thus has hybridised and adopted a broad range of traditional connoisseur (e.g., examinations), norm-referenced (e.g., diagnostic tests), qualitative (e.g., student profiles, observational scales), text analytic (work portfolios), and criterion-referenced (e.g., competency-based) approaches to assessment.

Taken together, these approaches serve crucial educational functions: the regulation of entry into tertiary study, the provision of school and classroom-based feedback for teacher decisions on curriculum and instruction, and state-level feedback on the achievement of particular groups and sectors of the system. The prototypical Queensland system of secondary student assessment blends student portfolio work, classroom and school-based assessment, state-wide moderation criteria and practices, and school-level standardised testing score reporting (ASAT) as a norming mechanism for comparing overall school results (Maxwell, 1997). It is at present being studied closely by educational authorities in Asia and the UK. Teacher-based early literacy profiling developed in Western Australia, Queensland and other states, and variations of the New Zealand Reading Recovery early literacy assessment procedures are currently being adopted and implemented in the US and UK (Luke, Land, van Kraayenoord & Elkins, 1997). Note here that many US school districts considering these hybrid Australian methods are seeking alternatives to large scale standardised achievement testing.

It would appear, then, that recent claims that Australian schools do not have systematic grids and ‘safety nets’ for literacy are not altogether accurate. While several states do not have comprehensive American-style standardised achievement testing systems, all of the state systems have mixed system-level approaches to assessment. For now, we want to step aside from the public and political debate around mass testing. Our questions here are: Can the benchmarks accommodate and coexist with these distinctive Australian approaches to classroom-based assessment? Does the ‘rush to testing’ that appears to stand behind their development place at risk the affiliated curriculum innovations in literacy? We begin with a brief review of curriculum and assessment innovation, then turning to look at how the proposed national benchmarks for literacy stand in relation to these developments.

Australian literacy curriculum and assessment

The diverse literacy education practices and models used in Australian schools have been extensively documented (Christie et al. 1991; van Kraayenoord & Paris, 1994; Freebody, Gunn & Ludwig, 1997). The development of, for instance, process writing, whole language, genre and critical literacy models have generated ongoing and, at times, heated debates over appropriate methods and curriculum. It is not our intention here to rehearse these debates over method yet again. Our aim is to describe - especially for those readers of Curriculum Issues who are not literacy specialists - what each approach has contributed to how Australian teachers assess children’s literacy in their classrooms.

Whole Language: With a basis in American psycholinguistics and Deweyian progressive pedagogy, whole language approaches had their most extensive implementation in the 1980s and early 1990s (see articles in Turbill & Cambourne, 1997). Key elements of that project included a focus on meaning-centred and child-centred instruction, and the learning of linguistic cueing systems through a variety of "naturally-occurring" texts. The affiliated approach to assessment is described in the metaphor of "kidwatching" (Goodman, 1978): principled observation of student behaviour and social interaction based on psycholinguistic, sociolinguistic and ethnographic models of language use in everyday life. Systematic teacher observations include: anecdotal records of students' participation and interactions; running records and miscue analysis of students' oral reading; and the face-to-face assessment of comprehension achievement through teacher analyses of retelling and/or answers to questions. The emphasis has been upon assessment using whole texts, preferably material used in classroom literature-based programs, rather than the decontextualised text excerpts that are used in many machine-scored reading tests. In terms of assessment, then, whole language models emphasise: (1) teacher expertise and classroom-based knowledge of language acquisition as a prerequisite for the systematic observations of students; (2) the immediate diagnostic and developmental uses of teacher-based assessment as feedback on the appropriateness and effectiveness of particular models of curriculum and instruction; (3) the need to situate a particular behaviour or practice (e.g., invented spelling) as part of a larger formative developmental process; and (4) the need to assess behaviours and practices with authentic and ‘whole’ texts within the larger contexts integrated projects of work and study.

Process writing: Based on American curriculum experiments that redefined primary, secondary and adult learners as real, practicing ‘authors’, these approaches conceived of writing as a social and rhetorical "process" involving conceptualisation, composing, editing and revision, and presentation (Graves, 1983). The assessment strategy of process writing also was based on the systematic observation by teachers of student progress, particularly as part of peer and teacher-led "conferences" during the editing and revising phases of the writing cycle. Because of its basis in rhetorical models of composition (Coe, 1983), process writing looked for the effectiveness of student "voice" in communicating to particular audiences - often real audiences - rather than grammatical or technical correctness per se. Process writing thus offered three significant insights for literacy assessment: (1) that student texts were statements of individual and cultural identity and "voice"; (2) that spelling, grammar, punctuation and other technical matters can only be evaluated in terms of the integrated social processes and products of actual student writing, and are not proper objects of assessment in and of themselves; (3) that student behaviours and participation in the social processes of composition are assessable aspects of student writing; and, (4) that texts should be assessed in terms of effectiveness for real world functions, audiences and purposes – rather than in terms of decontextualised criteria of correctness.

Reading Recovery: Developed and implemented in New Zealand by Clay (1983) and colleagues, Reading Recovery is a comprehensive system for the identification of at risk readers in Years 1 and 2. It involves face-to-face reading assessment using a set of diagnostic tasks that assess alphabetic knowledge, phonemic awareness and so forth, and lead to intensive programs of systematic one-to-one instruction. The approach has been implemented in parts of Victoria, Queensland and NSW, in the UK and in several US states. It has had a significant impact on mainstream teachers’ approach to early literacy assessment, with many of its tasks adapted for widespread use in state-level early literacy profiling. The overall influences of Reading Recovery on Australian literacy assessment have included: (1) an emphasis on assessment and early intervention in the first two years of schooling; and (2) the use of face-to-face diagnostic tasks that assess early "print knowledge" and "reading behaviour" rather than standardised tests of decontextualised skills.

Genre Approaches: Based on the Systemic Functional Linguistics, genre approaches view literacy as a social and linguistic practice, with written and spoken language shaped into conventional text forms to achieve particular social purposes (e.g., Christie, 1996; Halliday & Martin, 1995). Genre approaches aim for the development of an awareness and identification of the language features characteristic of particular conventional genres. This involves explicit instruction in a grammatical metalanguage that enables students to consciously control how they reshape texts for intellectual, institutional and occupational purposes. As in other approaches, students participate in "authoring cycles" that involve analysis of audience, drafting, and revision. Yet genre approaches differ in their key focus on the enhancement of specific technical lexical, syntactic and semantic features of students’ writing. Key assessment practices, then, integrate evaluation of effectiveness for particular audiences and context with more specific issues of technical lexical and syntactic features of the students’ writing. Major contributions of genre approaches to assessment practices thus include: (1) assessment of students’ proficiency with diverse text types, particularly non-fiction and expository; and (2) criterion-based assessment of students' mastery of particular technical aspects of language in the contexts of their relevant genres. In this way, genre approaches reframed many of the assessment approaches initiated by whole language and process writing, providing teachers and students with (3) a theoretically consistent and precise metalanguage for talking about the linguistic features of texts, and with ways of understanding the relations between social contexts and linguistic features.

Critical literacy and cultural studies: Critical literacy approaches attempt to teach students how to critique, deconstruct and reconstruct a range of canonical and popular texts and their cultural contexts (Muspratt, Luke & Freebody, 1998). This may involve use of the textual metalanguages advocated in genre approaches, combined with those derived from literary, semiotic and cultural studies. As in the aforementioned models, there is an focus on how texts work, but this is accompanied by explicit instructional attention to the social conditions and contexts for the production and interpretation of texts. In this way, critical literacy approaches use text analysis and study as a springboard towards the analysis of social institutions, cultural formations, and new media of information (Buckingham, 1998), arguing that such analyses are the "new basics" required by new workplaces, technologies and institutions in postindustrial cultures. Curriculum development in critical literacy, cultural studies and media analysis thus has had several significant influences on assessment practices: (1) a further broadening of the range of assessable texts to include popular cultural, media and everyday community texts; (2) a focus on the assessment of critical responses to and alternative reconstructions of texts and contexts; (3) the recognition that new "multiliteracies" involve integrated uses of print and visual, traditional and media literacies.

These paradigms of literacy education in Australia have raised serious challenges to conventional norm-referenced standardised testing paradigms. In fact, it is arguable if these kinds of curricular reform could have taken place in ‘test-driven’ educational environments such as the US or UK. The problem is simple: many of the ‘new basics’ described have never been accommodated by pencil and paper mass testing systems. The historical bases and operational assumptions of standardised achievement testing reflect behaviourist models of literacy, with reading and writing conceived of as discrete and transportable behaviours or skills that can be assessed in the contexts of pencil and paper stimulus/response tasks. The curriculum innovations we have described here place on the table a range of key considerations for those designing assessment systems and instruments. These include concerns about:

(1) context: assessment itself is not neutral and constructs a social context for literate behaviour. Accordingly its tasks and items should approximate as much as possible real and diverse conditions of use and practice;

(2) systematic observation: single-shot pencil and paper assessment in itself is not as useful for flexible adaptation and adjustment of curriculum and instruction as teacher-based observation of literacy practices and events;

(3) pedagogic process: observation of instructional interaction (not just performance outcomes) is a crucial part of formative and developmental assessment;

(4) textual diversity: assessment should encompass a broad array of text types rather than simply school-based or textbook style texts;

(5) descriptive metalanguage: assessment requires an analytic metalanguage that has some technical and theoretical power beyond commonsense namings of skills and behaviours; and,

(6) new textual practices and integrated modalities: assessment should focus on ‘new basics’, including aspects of critical literacy, visual and media literacy, and on-line literacy and how particular literate practices and events may require and value new combinations of these.

The current situation

Many of the aforementioned assessment practices are commonly used in Australian primary and secondary classrooms (van Kraayenoord, 1996). Beginning in the late 1980s, a number of policy reorientations influenced the direction of Australian educational evaluation: specifically, economic rationalist approaches to government and public administration that stressed accountability; and the adaptation of a human capital model as evidenced in the "clever country" educational slogan (Lingard, Knight & Porter, 1993). Both of these policy reorientations stressed the generation of quantitative performance indicators as evidence of educational output and success. They provided the sociopolitical context for the development of such testing systems as the NSW Basic Skills Test in Literacy and Numeracy in 1989, and, in the mid 1990s, the Year 6 test in Queensland Schools. The introduction of A Statement on English for Australian Schools (Curriculum Corporation, 1994a) and frameworks for reporting outcomes of the curriculum such as English - A Profile for Australian Schools (Curriculum Corporation, 1994b), signalled a further incremental move towards universal reporting for accountability purposes.

In 1996, the current government’s National Literacy and Numeracy Plan called for the development of national benchmarks that would act as descriptions of minimum standards, and "comprehensive assessment of all students by teachers as early as possible in the first years of schooling", as well as Years 5, 7 and 9 by 1998 (MCEETYA, 1997). The current benchmarking project was initiated in 1997 with extremely tight timelines for drafting, consultation, and evaluation. That process is still underway. In the year-long process of national consultation, a host of issues have been raised about the proposed Benchmarks. These include questions about:

· Performance description that is reduced to behavioural skills and subskills;

· A minimal competency approach to performance description;

· Limited verification by recent literacy-education theory and research;

· Disintegration of modalities of reading and writing, but also spelling etc. into discrete domains;

· Standardised version of English and language development that fails to acknowledge the cultural and linguistic diversity of school age populations ;

· Omission of critical literacy and other key areas of curriculum innovation;

· Proposed and potential uses and abuses of the data generated by resultant testing and assessment systems;

· The cost effectiveness of large scale census testing, when sample testing and teacher-based profiling may provide more valuable and useful local data.

Anxieties over the uses and abuses of the benchmarks have been fuelled by the climate of ‘crisis’ and media representations of literacy failure, problematic teaching methods, and ineffective schools. In 1997 there were ongoing political disputes between state and Commonwealth governments over when and how the Benchmarks might be tied to state testing, and over whether test results might be used as a grounds for determining levels of Commonwealth funding to states (Comber et al. in press). This has contributed to a belief among many educators that benchmarks are a prelude to national testing, the tying of public funding of schools to testing, and, in the long run, teacher appraisal via test results. Our concern here, however, is with the potential adverse and counter-innovative effects on literacy teaching that the imposition of the benchmarks might lead to. In what follows, we take a critical look at one of the draft benchmarks, asking whether and how it might accommodate the curriculum directions noted above.

Can the benchmarks work?

The following sample is taken from the February 1998 draft Benchmark statements, the latest available as this article goes to press.

Insert Figure 1 about here

Benchmark: Writing – Year 3

Students compose several pieces of writing that, on the whole, make sense to the reader and show a basic understanding of the writing task.

The pieces of writing contain a few related ideas relevant to the task and subject matter. The ideas are usually briefly expressed.

The pieces of writing show evidence of some organisation of the subject matter (eg a simple beginning, middle and end in a story). However, they may also include some irrelevant details and may be confusing.

In these pieces of writing, students use:

· simple sentences and some more complex sentences (eg using joining words like and, but, then, because)

· phrases and words to locate events in place and time, and to specify the way in which something is done (eg in the park, on the lid, on Monday, today, by bus)

· vocabulary appropriate to the subject matter of the writing, including some technical words

· capital letters and full stops (used correctly most of the time).

Figure 1: Benchmark: Writing – Year 3

source: Draft Literacy Benchmarks, Curriculum Corporation, 1998

As it is described here, early writing for Year 3 students involves a "task", directed to an audience ("reader"), and "subject matter". Mastery or knowledge of text structure is here stated in terms of the "organisation of the subject matter". The desired textual features are a beginning mastery of conjunctions and some cohesive ties, the use of prepositional phrases for deictic references to spatial and temporal contexts ("locate events in space and time"), some technical vocabulary relevant to the subject, and basic punctuation and orthographic conventions to mark sentence beginnings and endings.

To its credit, the description opens the possibility for a diversity of texts and could apply equally to any one of "several pieces" of expository, literary, functional community texts commonly written across a range of the approaches to literacy teaching. There also is an indirect acknowledgment of the need for assessing student writing in relation to the construction of meaning in a specific context, with mention of the need to "make sense" to an audience.

However, the technical language here for describing textual features falls somewhat short of accommodating current best practices. The concept of text structure is underdeveloped and vague, viewed in terms of subject-matter organisation and exemplified by reference to narrative. This leads to affiliated problems in the detailing of text features. Specifically, the actual use of (and absence/presence of) conjunctions, deictics, and indeed basic punctuation depends on the genre (i.e., audience, purpose, task, text structure) in question. That is, if the student is composing a typical functional text featured in many Grade 3 classrooms (e.g., advertisement, label), the use of deictics might not be relevant or the punctuation conventions might vary considerably. One of the tasks for early writers is, indeed, the recognition of the variability of text features according to context. Part of the problem is that the actual metalanguage has been reduced to a commonsense naming of skills and behaviours. Admittedly, the benchmarks are supposed to be readable by all stakeholders, including parents, and non-specialists, however, the reduction of that metalanguage is, we would argue, indicative of some serious difficulties in describing the relationship between text features and context.

As it is constructed, this benchmark cannot effectively address the issue about the relationship of social context to linguistic features, and more generally, genre. This may be because it is caught in what we would term a contextual/universal contradiction, attempting to achieve two contradictory tasks. On the one hand, it is trying to maintain the openness of the benchmark to a diversity of possible tasks, genres, and contexts, a curricular goal not incompatible with most current practice. On the other, it is trying to specify universal mastery of particular textual features as a developmental goal that may or may not be manifest in the text chosen by the teacher, classroom, student or, indeed, test in question. That is, the ‘appropriate skills in context’ approach of much current curricula is in direct conflict with the ‘universal basic skills irregardless of context’ approach that is part of the benchmarks’ stated brief.

Additionally, there is the matter of "subject matter". In its attempt to accommodate diverse texts and genres, the benchmark attempts to maintain the possibility of a range of relevant content. So far, so good. However, what goes unremarked is the student’s relation to the subject matter. There is no comment here on the intellectual relationship between the student and the knowledge in question other than the metaphor of "mastery". That is, there is no concern here with the student’s capacity to position her/himself in relation to the knowledge and the text – through agreement, disagreement, critique – all key to various models of critical literacy and critical thinking in primary education.

This benchmark is avowedly a statement of minimal competency. But by our standards, the benchmark stops short of addressing many of the six challenges for assessment we noted above. To its credit, it attempts to open out the possibility for students to be assessed at their writing of a diversity of texts (4) in relation to context (1) variables of task, audience and relevant content. However, the limitations of its descriptive metalanguage (5) lead to an inability to connect context with text, a limited model of text structure, and, relatedly, a failure to see the text/context specificity of the use of the linguistic features (conjunctions, prepositions, punctuation) sought. As a result, the systematic observation (2) enabled by this benchmark is flawed and contradictory. At the same time, there is little indication of the need to assess or observe participation in the pedagogic processes (3) of writing, nor to observe for new textual practices and integrated modalities (6), specifically critical literacy.

The rush to testing

That there are significant technical, theoretical and, indeed, practical limitations to the benchmarks should not be surprising. The task facing the authors, Ministers and government is indeed difficult and complex: as a guide for assessment nationally, they must meet scientific criteria of theoretical coherence and validity; as a guide for teachers, curriculum developers and others, they must constructively guide and contribute to the various professional practices and knowledges we have discussed here; as a means for accountability to parents and communities, they must be accessible, clear and of visible local relevance and value; and, as government educational policy, they must both capture and speak to a diversity of states, constituencies and interests.

In this context, the benchmarks inevitably will be hybrids, composites and, indeed, compromises. The most problematic issue, however, is not the benchmarks per se but how they become reifications of standards, taken not as the artefacts of the various trade-offs and compromises noted above but as real phenomena - conglomerations of preferred literate skills, traits, attributes, behaviours, practices that are seen to actually exist and serve some purpose in the world - to be translated into single-shot standardised achievement tests. In an environment of administrative devolution, school-based management and increasing accountability via measurable performance outcomes, any such tests would quickly become the dominant criteria specifications for what should be learned at different year levels (in effect, bypassing existing state curriculum statements).

In such a scenario, the risk is that the very limitations of the benchmarks become major liabilities for the educational system - that the descriptive minima will become prescriptive maxima. The scenario we would envision here is that teachers who at present might be observing Year 3 writers for a range of language competences and practices (e.g., competence in the construction of community texts, using language for problem-solving, higher order thinking, use of critical reading as a resource for writing, emergent bilingualism) might find themselves testing for and teaching for proficiency with conjunctions and prepositions. Such is the case in many jurisdictiond in the US, where ‘what can be tested’ has become the major driving force in determining what gets counted as literacy in classrooms and communities. Where this is the case, benchmarks and affiliated assessment practices could narrow and limit the capacity of teachers and school communities to adjust to the challenges of new workplaces, new technologies, new literate practices and institutions.

It is axiomatic in curriculum theory that what counts as knowledge, practice, skill and competence is driven by assessment systems (Bernstein, 1990). This is increasingly the case in an educational and public sector environment where ‘performativity or disappearance’ is the order of the day: where the institutional imperative is to produce visible evidence of economic efficacy. We do not object to assessment. It is, we have argued here, an essential component of every major approach to literacy education in Australia. Nor do we object to the accountability of schools to systematically demonstrate student educational outcomes. This is a particularly significant matter in the context of declining state funding across the postindustrial West and the increasing educational and economic problems for the most ‘at risk’ communities in Australia.

Yet it would be a great tragedy if 20 years of innovation in curriculum, instruction and assessment were discarded in favour of national benchmarks and standardised testing.

Note

The authors are listed in alphabetical order.

References

Bernstein, B. (1990). Class, Codes and Control, Vol. IV: The Structuring of Pedagogic Discourse. London: Routledge.

Breen, M., Barratt-Pugh, C., Derewianka, B., House, H., Hudson, C., Lumley, T., & Rohl, M. (1997). Profiling ESL Children. Vols 1-3. Canberra: Department of Employment, Education, Training and Youth Affairs.

Buckingham, D. (ed) (1998). Teaching Popular Culture. London: UCL Press.

Callahan, R.B. (1962) Education and the Cult of Efficiency. Chicago: University of Chicago Press.

Christie, F. (1995) The role of a functional grammar in the development of a critical literacy. In G. Bull & M. Anstey (eds), The Literacy Lexicon (pp. 46-57). Sydney: Prentice Hall.

Christie, F., Devlin, B., Freebody, P., Luke, A., Martin, J.R., Threadgold, T., & Walton, C. (1991). Teaching English Literacy: A Project of National Significance. Canberra: Department of Employment, Education and Training.

Clay, M. (1983). Reading: The Patterning of a Complex Behaviour. Auckland: Heinemann.

Coe, R.M. (1983). Form and Substance: An Advanced Rhetoric. New York: John Wiley.

Comber, B., Green, B., Lingard, R. & Luke, A. (in press). Literacy and public education: A question of ‘crisis’? In A. Reid (ed) Public Schooling: Public Good or Safety Net. Canberra: Australian Curriculum Studies Association.

Curriculum Corporation. (1993a). A Statement on English for Australian Schools. Carlton, VIC: Curriculum Corporation.

Curriculum Corporation. (1993b). English - A Profile for Australian Schools. Carlton, VIC: Curriculum Corporation.

Curriculum Corporation. (1998, 23 Feb. Draft). Draft Literacy Benchmarks, Writing - Year 3. Carlton, VIC: Curriculum Corporation.

Freebody, P., Ludwig, C. & Gunn, S. 1995. Everyday Literacy Practices in and out of Schools in Low Socio-economic Urban Communities. Vol. 1. Caberra: Department of Employment Education and Training.

Goodman, Y. (1978). Kidwatching: An alternative to testing. National Elementary School Principal, 57, 41-45.

Graves, D. (1983). Writing: Teachers and Children at Work. Portsmouth, NH: Heinemann.

Halliday, M.A.K. & Martin, J.R. (1995) Writing Science. London: Taylor & Francis.

Lingard, R., Knight, J. & Porter, P. (eds) (1993) Schooling Reform in Hard Times. London: Falmer Press.

Luke, A. (1994). The Social Construction of Literacy in the Classroom. Melbourne: Macmillan.

Luke, A. & Decastell, S.C. (1985) Educational crises and the rhetoric of reform: The Arnold/Huxley debate reconsidered. Access: Contemporary Themes in Educational Inquiry, 4(2), 22-50.

Luke, A., Land, R., van Kraayenoord, C. E, & Elkins, J. (1997). The Year Two Diagnostic Net: A Critical Analysis. Brisbane: Queensland School Curriculum Council.

Maxwell, G.S. (1997) A Systems Analysis of Selection for Tertiary Education: Queensland as a Case Study. Unpublished PhD dissertation, Graduate School of Education, University of Queensland.

Ministerial Council on Education, Employment, Training and Youth Affairs (MCEETYA). (1997). Media Release. Eighth (Special) Meeting, December 11.

Muspratt, S., Luke, A. & Freebody, P. (eds) (1998) Constructing Critical Literacies. Sydney: Allen & Unwin.

Turbill, J. & Cambourne, B. (eds) (1997). Whole issue. Australian Journal of Language and Literacy 20 (2).

van Kraayenoord, C.E. (1996). Literacy assessment. In G. Bull & M. Anstey (Eds.), The Literacy Lexicon (pp. 237-247). Sydney: Prentice Hall.

Van Kraayenoord, C.E. & Paris, S.G. (1994). Literacy instruction in Australian primary schools. The Reading Teacher, 48(3), 218-228.

Wertsch, J.V., Del Rio, P., & Alvarez, A. (eds) (1995). Sociocultural Studies of Mind. New York: Cambridge University Press.