Elizabeth S Allen

Increasingly, reading proficiency is being measured in terms of outcomes based data. This drive for accountability supports the growth of National Profiles, Literacy Benchmarks, and statewide assessments such as the Diagnostic Net, which record the numbers of students who fail to make the grade. In the face of widespread concerns about perceived low literacy standards, Governments have responded by providing increased funding for students experiencing difficulty, introducing specialized programs and personnel, and revamping syllabus documents. But are these responses based on an authentic assessment of the situation? Outcomes based data cannot adequately reflect the complex issues addressed in school contexts. Perhaps we should be asking further questions including: What is the nature of reading support? How is it planned and provided? What actually happens in support programs? How do students perceive the support they receive? This paper presents the rationale and methodology of current research investigating policies and practices in reading support across Years 1-3, in two Queensland schools. The research is designed to identify the critical elements contributing to success in these contexts recognized for their provision of highly effective reading support. It features an emphasis on qualitative methods of investigation, an exploration and analysis of teacher theoretical orientations and their effects on planning and practice, and comprehensive description and analysis of the actual experiences and perceptions of the students receiving various types of reading support. If we want to get answers that matter, we must ask the right questions.

It should be made clear at the outset that this paper represents research in progress. Its purpose is to raise an issue of importance in the field of literacy, and to outline research which might inform an investigation of the phenomenon. The value of presenting the paper at this stage of research lies in its focus on qualitative inquiry into successful practice in reading support. It uses a method aimed at gathering information which will supplement measures centred around standardised testing and outcomes based evaluation.

The research investigates planning and practice across Years 1-3 in two Queensland schools. A strong focus on student perceptions of support adds a further perspective which is not commonly sought. This paper provides an outline of the theoretical issues and the method used to investigate reading support across the early years of reading acquisition. It therefore illustrates an alternative way in which successful practice might be examined. Full results of this research will be available in 20000.

Millions of dollars have been expended in recent years in an attempt to improve literacy standards in Australia. This expenditure includes general education budget funding, reports on literacy standards, development and implementation of specific programs and assessment procedures, research grants to universities and individuals and school funding specifically targeted at literacy improvement (DEETYA, 1996, 1997; Department of Education, 1995a, 1995b; Education Queensland, 1998). Accountability for the expenditure of these funds, is clearly an important issue. Measures of accountability at the Federal and State levels are designed to focus on expenditure of funds and improvements in literacy levels. This information is then used to inform public discussion. Such measures, however, fail to investigate, nor have real impact on changing reading practices in classrooms (Wixson, Valencia, & Lipson, 1994).

The current drive for accountability means that increasingly, reading proficiency is being measured in terms of outcomes based data. It supports the growth of National Profiles, Literacy Benchmarks, and statewide assessments such as the Diagnostic Net, which record the numbers of students who fail to make the grade. But are these responses based on an authentic assessment of the situation and do they provide data that will inform measures to remedy or prevent the problem?

In Queensland, the current emphasis on outcomes in education, for example, the Year 2 Diagnostic Net levels, the Year 3 and 5 Benchmarks, and the widespread use of standardized reading tests in schools, means that there is a danger that children's progress will be measured only in terms of exit levels of achievement or time spent participating in particular programs. Programs may be assumed to be successful based solely on the results of standardised testing, comparisons with like schools results or previous year's scores. This may indicate whether improvement has occurred, and establish general trends, but does not fully describe children's experiences of reading support, nor identify elements of successful practice which might then be replicated.

Outcomes based data cannot adequately reflect the complex issues addressed in school contexts. In addition to assessing the end result of reading programs, we should be asking further questions. More informed questions might concern the nature of reading support currently provided, the ways in which programs are planned and implemented, an examination of the content and dynamics within the support situations and student perceptions of the support they receive. Answers to these questions could help to present a more comprehensive picture of the phenomenon by exploring the contexts in which statistics are generated. Statistics might provide an assessment of the magnitude of the problem, but a deeper understanding of the contexts and the practices occurring within them could identify key areas for improvement. This would have the potential to provide guidance for planning or re-evaluating support programs in an effort to provide improved reading support to students who experience difficulties.

The problem of reliance on outcomes based data is also compounded by the largely fragmented approach of research into reading and reading difficulties. Much of the current research about reading focuses on specific issues of gender, socio-cultural aspects, teaching strategies and practices, assessment of literacy levels, methods for addressing literacy problems, specific reading programs and their efficacy, and specific reading difficulties (Christie, 1990; Freebody, Ludwig, & Gunn, 1995; Luke, 1992). Such studies inform particular aspects of reading, and factors affecting the learning and teaching of reading, but they have not explored how reading support is delivered, what students experience as reading support, or the cumulative effects of various support programs which the student might experience. They do not provide a holistic view of reading support in the early years. The Benchmarks are underpinned by the belief that children should attain mastery of fundamental literacy skills by the end of Year 3 (Commonwealth of Australia, 1997). If the process of attainment of these levels is to be understood, then more informed questions need to be asked to assess current practices across the years of formal schooling to that point.

Schools need to be accountable to their sources of funding, but current collective measures of accountability lack the specificity to inform practice. More helpful information might be obtained from a study of specific contexts. Schools are ultimately accountable to their communities and this puts the focus firmly on schools and classrooms at the local level (Wiggins, 1993). If the real locus of control for accountability is at this level, then research into actual practices, in day-to-day interactions in individual schools, is where useful information can be generated.

The aim of providing reading support is to improve children's reading ability, therefore, it follows that information about practice should be sought from schools where highly successful practice has been identified. An examination of the services in such schools could identify the key elements of effective practice operating in those particular contexts. Regardless of the unique ways in which the identified schools combine these elements to address the needs of their communities, it is highly likely that certain common, critical elements could be identified. Standards, scores, and levels of achievement are not enough. Inquiry must go further and examine the processes through which these standards, scores, and levels were generated. Questions must be asked of actual day to day practice in authentic settings.

A research project currently in progress has been designed to ask questions about exactly what happens when students receive reading support, at the actual point of policy implementation. This will provide an accurate, in-depth picture of practice which will supplement and illuminate information currently available from outcomes based data.

This study will address these questions by providing comprehensive coverage of the phenomenon across the critical years 1-3, to allow examination of elements and interactions among elements that contribute to successful provision of reading support.

Evaluation of school policy, planning, and practices, together with student experiences and perceptions, will help educators to gain a more comprehensive picture of how schools service clients requiring support for reading difficulties and what counts as reading support within programs offered. This will better inform policy making, and provide for more effective planning of support services, to maximize student reading achievement.

A description of the research project will now follow. Research questions will be outlined and a concise review of relevant theoretical issues will be presented in order to indicate areas where questions can profitably be answered.

The research questions indicate that research will be centered on three broad areas of interest. It will be concerned with ways in which schools interpret the nature of reading and reading difficulties, the service delivery models through which the support is provided, and the interface between these factors, as manifested in student experience of that support. All three areas will be influenced by various factors unique to individual school contexts.

The research questions will address three levels of implementation of reading support: school, classroom, and student.

What are the key elements associated with the delivery of effective support services to students in Years 1-3 who are experiencing reading difficulty?

1. How do schools develop policy for providing reading support?

2. What support services are offered at each year level from 1 - 3?

3. What are the relationships among elements of policy and practice, which influence provision of effective reading support?

Class level

4. What are the characteristics of the support provided?

5. How is reading support practiced in each of the support programs?

6. What theoretical views about reading are reflected in the observed practices?

7. What are the students' experiences and perceptions of the reading support

provided?

Overarching all considerations of reading support in schools is the question of what theoretical orientation to literacy (and therefore reading) has been adopted in those contexts. Reading is a complex task, and it has been established that what counts as reading is culturally defined and changes over time (Anstey & Bull, 1996; Christie, 1990; Freebody et al., 1995; Meek, 1992). The current Queensland English syllabus is a blend of several models of literacy and its theoretical underpinnings are explicitly acknowledged. The English syllabus is supported by aspects of Skills approaches, Growth, development, process, and whole language approaches, Cultural heritage approaches, Functional linguistic approaches, Genre based approaches, and Critical literacy approaches (Department of Education, 1994b). These are underpinned by perspectives related to social justice policies and effective learning and teaching principles (Department of Education, 1994b). School English policy documents are meant to reflect this interrelated broad base.

However a number of researchers have indicated that this theoretical base does not necessarily carry over into teacher planning and practice because teachers tend to mix different approaches (Freebody et al., 1995; Lo Bianco & Freebody, 1997; Smith, 1994). This means that it is likely that a range of theoretical positions might be operating across classrooms and across Years 1-3. This has implications for the differing definitions of reading and reading difficulty that might be held, and also for the different ways in which support might be organized. As this largely determines students' opportunities for support, and cumulative effects and experiences of support, questions should therefore be asked to determine theoretical orientations.

Several broad models pertaining to reading have been identified as significant (Anstey & Bull, 1996; Rude & Oehlkers, 1984). These could be categorized as Information Transfer Models, Interactive Models, and Critical Literacy Perspectives. Reflecting these orientations, reading could be seen as (a) a series of hierarchical skills incorporating visual and auditory discrimination, phonics, structural analysis of words, and word recognition, allowing children to find the correct meaning vested in the text (Bransford & Franks 1974), (b) as an active participation in constructing the author's meaning through experience of the predictable ways in which texts are organized, prior knowledge and experience of life, and conscious knowledge and application of strategic skills (Cambourne 1988; Smith 1978) or (c) as a socio-cultural practice requiring knowledge of text structure, and the ability to make judgements about ways in which texts reflect the dominant culture's ideologies (Christie, 1990; Luke, 1993).

Following on from this, reading support may be approached in the following ways depending on theoretical orientation: (a) Information transfer: placing poor readers into ability groups, withdrawing them for remedial sessions with specialist teachers, giving them more practice with carefully graded texts, and providing explicit, direct teaching incorporating word attack and phonic skills as well as comprehension exercises (Freebody et al., 1995; Rude & Oehlkers, 1984) (b) Interactive models: providing support in the context of the classroom at an individual developmental level, providing additional opportunities to learn the features of the genre through continued exposure to predictable texts, and monitoring progress through observation, analysis of miscues and strategy use and comprehension skills (Jennings, 1996), and (c) Critical literacy models: focusing initially on prevention, early identification of difficulties, in class modes of support combined with withdrawal for intensive sessions, direct teaching and explicit modeling of the selected genres and intervention for students from groups known to be at risk, that is, from low socio-economic backgrounds, of ethnic origin, masculine gender, or from isolated communities (DEETYA, 1997; Freebody et al., 1995; Hill, Comber, Louden, Rivalland, & Reid, 1998).

From the above outline of views of reading and the type of support likely, several broad definitions of reading difficulty can be identified. These include: (a) Deficiency definitions characterized by a belief that student difficulties are a result of an inability to master mechanical phonological skills, sound and symbol matching and blending, decoding, and comprehension. (b) Discrepancy definitions characterized by a mismatch between reading expertise and a reader's optimal intellectual capacity or oral language development level (Gunning, 1998) and (c) Functional definitions characterized by poor achievement, usually in test results or according to tasks that others in the same context are able to do (Friend & Bursuck, 1999; Gunning, 1998).

Views of reading difficulty reflected in support practices should be compatible with the views of reading espoused in school policy and classroom planning documents. Gunning (1998) however, has pointed out that discrepancy definitions are necessarily used in schools because of the funding attached to programs using a functional definition and that students with recognized learning difficulties, are usually identified for support primarily on the basis of the discrepancy definitions (Gunning, 1998).

Although students are identified in these various ways, it has been found that reading support reflecting deficiency definitions is still prevalent (Anstey & Bull, 1996; Cambourne, 1995). Models of reading have changed over time, and deficiency definitions of reading difficulty are not supported by the current views of reading reflected in the Queensland English syllabus, yet practices in reading support have not reflected these changes.

It is therefore important to ask questions about the match between theoretical orientations of reading, and reading support as represented in planning and revealed in practice.

Service delivery models have been described as: "a plan for bringing together students, teachers, instructional methods and materials . . . . to foster effective teaching and learning" (Hallahan, Kauffman, & Lloyd 1996, p.437).

There is wide consensus that classroom teachers are expected to provide reading support as part of regular curriculum delivery (Ashman & Elkins, 1998; Department of Education, 1994a; Friend & Bursuck, 1999). This might be termed a classroom modification mode. Ideally, it is the support system students encounter first, and which remains at the core of all support received (Department of Education, 1994a, 1995b). Other models might include collaborative consultation and co-teaching models where learning support teachers take varying degrees of responsibility for whole class or small group teaching depending on their allocated roles. Hallahan et al. (1996) and van Kraayenoord and Paris (1994) stated that most schools use a range of models, combining these with a variety of instructional approaches and grouping strategies. It is also possible that certain delivery models may be more appropriate for use in years 1-3. Few questions have been asked concerning the ways in which schools decide approaches or combine approaches to provision of reading support. This is clearly a significant issue, considering that a number of other avenues of support may also be accessible and schools generally attempt to plan a delivery model which maximize input from all support systems at their disposal.

Additional sources of support might include Reading Recovery, Support-a-Reader programs, teacher aides, parents and a range of specialist teachers (Education Queensland, 1998). Questions leading to documentation of the unique way in which schools combine such sources and models of support effectively, would add to current knowledge.

While theoretical models and service delivery models provide part of the picture of reading support, it is the interface of these areas with the student experiencing difficulty, which represents school policy in practice. Initially, the setting where this takes place is the classroom. This is where a number of critical questions can be answered

Questions about student experience of reading support encompass at least two facets, namely what teachers plan and deliver as support, and how students experience and perceive that support. Therefore it will be important to examine aspects of both facets. Classroom teachers have the primary responsibility for providing support for all children experiencing difficulty in reading, and are often the first to notice reading problems (Department of Education, 1994a, 1995b; Forlin, 1997; Friend & Bursuck, 1999). They make decisions about needs and levels of support, assign students to particular support programs, and refer students to learning support teachers and other specialist personnel. The individual classroom teacher is therefore at the heart of the support process, and the initial reference point for examination of student experience and perceptions.

School policy reflects current views as set down in the syllabus. However, McLaughlan (as cited in Jennings, 1996, p. 174) stated that "What is actually delivered or provided under the aegis of a policy, depends finally on the individual at the end of the line." Jennings (1996, p. 15) supported this, stating that implementation of policy "is mediated through the individual classroom practitioner's beliefs and experience." Teachers' theoretical orientations directly affect their beliefs and teaching practices (Anstey & Bull, 1996; Freebody et al., 1995; Hiebert & Taylor, 1994).

However, research has shown that questions should not concern which particular theories or beliefs are held by the classroom teacher, but whether the various elements fit together as a coherent and complementary whole, to provide the support program for the student with reading difficulties. General issues to be included in research have been identified as (a) teacher definitions of reading, (b) teaching methods, (c) content and materials, (d) classroom interactions, (d) modes of support provision, and (e) use of time. A brief summary of the issues involved will indicate where questions should be asked.

(a) Definitions of reading

Meek (1992) has pointed out that the ways in which teachers define reading might not be the ways in which students understand it. It has been found that the learner reader's view of what counts as reading depends on the value assigned to particular kinds of reading and the purposes of reading they experience (Hunter-Carsch, 1995). Children experiencing reading difficulties might be confused as to what reading is, thereby creating a barrier to learning (Meek, 1992). This problem might be exacerbated because of the range of support situations and personnel in schools, and the different definitions of reading which might be encountered.

This study will examine planning and practice to identify how reading is defined in each reading support situation. Subsequent analysis of this data will establish how the cumulative effects of different definitions of reading, might impact on student experience and perceptions of reading support.

(b) Teaching methods

Teachers may tend to select teaching methods on the basis of familiarity rather than by choosing from a wide range, therefore discriminating against some students with different learning styles (Brady, 1995). Brady (1995) and Pinnell, Lyons, Deford, Bryck and Seltzner (1994) found that teachers concentrated on "safe" methods of support for poor readers, focusing on isolated decoding skills and systematic teaching of phonics. Explicit teaching in practical and meaningful interactions has been implicated in successful reading acquisition (Department of Education, 1995b; Smith, 1994). Freebody et al. (1995) however, found that in some classrooms, the learning focus was rarely made explicit, lessons were neither real life nor life like, and non-specific vocabulary was used in discussion.

(c) Content and materials

Teacher labeling may result in a narrow selection of content for students with reading difficulties, favoring discrete sets of skills of reading (Freebody et al., 1995). This has clear implications for provision of suitable curriculum for children experiencing difficulty, as content, teaching methods, and evaluation need to be tailored to individual needs. However, research has indicated that such processes do not always occur (Meek, 1992). This issue is also important because of the variety of support sites, programs and personnel encountered by students with reading difficulties. Research has shown that without coordination across sites, support can lack cohesion and result in a fragmented curriculum (Christie, 1990). This indicates an important area of inquiry.

(d) Classroom interactions

Several researchers have identified the importance of the quality of student and teacher classroom interactions, in ensuring successful support for students with reading difficulties (Allington, 1994; Smith, 1994).

Research has found that classroom interactions are tightly bound up with pedagogical procedures unrelated to useful reading skills and knowledge (Freebody et al., 1995; Meek, 1992). Students need to be fluent in pedagogical routines concerning roles, responsibilities, sequences of talk interchanges, and conventions of classroom behavior, in addition to knowledge of the reading process (Freebody et al., 1995; Meek, 1992). Patterns of teacher interaction with students experiencing reading difficulty, typically show more interruptions, phonological cues, and lack of specific feedback, when compared to teacher interaction with able readers (Freebody et al., 1995). This reinforces the idea of reading as a set of skills, perpetuates low expectations and results in less time actually spent reading (Hiebert & Taylor, 1994). This is a concern as research has shown that the amount of time spent reading is directly related to achievement (Smith, 1994). Freebody et al. (1995) said that researchers seldom questioned the ways in which teacher practices affected individual students. This study will examine interactions between support personnel and students, to address these issues.

(e) Modes of support provision

Concerns about the relative benefits of withdrawal, or in-class delivery of support, are well documented. Withdrawal programs offers opportunities for intensive support, but some difficulties identified include: removal of responsibility from classroom teachers, reinforcement of the idea of reading as a separate subject, little transfer of learning to other contexts, concentration on discrete skills of reading, lack of collaboration between support staff and classroom teachers, fragmentation of curriculum because support does not match the class program, presentation of different views of reading, and loss of student self esteem because of the stigma associated with withdrawal (Ashman & Elkins, 1998; Hiebert & Taylor, 1994; van Kraayenoord & Elkins, 1998). In-class support is intended to minimize such concerns, but a number of studies indicate that this does not necessarily happen. Problems identified include inadequate consultation, and confusion of teacher roles (Ashman & Elkins, 1998). Ashman and Elkins (1998) state that whole school policies and extensive collaboration are necessary to achieve effective support provision, but little research has addressed this question of how schools effectively manage such diverse models, modes, and influences. Nor have the effects of the total experience of all support received, been documented from the student's perspective.

(e) Use of time

Allington (1994) found that only 10% of the school day was spent in reading and writing. Hiebert and Taylor (1994) stated that programs were often evaluated on the basis of time allocated to them. Time allocated, however, is not necessarily the amount of time available for instruction. The amount of engaged time spent in reading instruction has been identified as the critical variable in the development of reading proficiency (Pinnell et al., 1994). Engaged time represents that part of instructional time when students are actively involved in relevant, meaningful tasks. Within engaged time, the success rate experienced by the student is thought to be critical (Clay, 1996; Gunning, 1998). Estimates of necessary success rates for students experiencing difficulty range between 70%-100% (Clay, 1996). It will be relevant, therefore, to evaluate how much engaged time is spent in reading instruction in the various support settings, and if students are experiencing high rates of success.

Conclusion

Little investigation has taken place into the ways in which schools draw from the range of theoretical models, instructional approaches, and service models, to formulate policy for reading support. The match between teachers' practices and articulated school reading policy has not been fully explored. Elements of classroom teaching practice and reading support have not been analyzed to determine if they represent theoretically compatible paradigms of reading, nor have these analyses been made in other reading support settings. Little research has been done on students' experiences and perceptions of reading support. In particular, the cumulative and combined effects of teaching practice on students, across all reading support sites attended, has not been fully documented. Therefore, these are the areas where questions should be asked in order to inform and improve practice.

Use of qualitative methodology will allow examination of these day-to-day interactions, real life occurrences, unpredictable events arising out of the support settings, and serendipitous findings, which may or may not be significant. An accurate picture of what the child ultimately experiences as reading support can be constructed, if descriptive, interpretive methodology is used

A Case study design was chosen because of the need for rich, detailed information from authentic, contemporary settings, where the meaningful characteristics of the phenomena under study must be preserved (Patton, 1990; Yin, 1994). Two schools will be selected for case studies of effective reading support provision. Within each case study of the school, a series of student case studies will then take place to ascertain student experience and perceptions of reading support.

Data will be collected through interviews, observations, and document analyses (Stake, 1994; Yin, 1994). Analysis of these multiple sources of data can provide a means of validation of results and also indicate where further clarification is necessary. A large amount of relevant data can be collected from multiple sources in a relatively short time, with a minimum of disruption to school routines, while allowing different perspectives to emerge.

The research questions show that data will need to be collected at three levels: the school level, the classroom level, and the student level. Each level will be described in turn.

Initial group interviews with the principal and other personnel involved in policy making, will be guided by the broad research questions to allow procedures to be explained in the school's own terms (Guba as cited in Patton, 1990). Documents relevant to the particular school situation will be collected.

All teachers from Years 1-3, and support staff will be interviewed, and relevant English overviews, lesson plans and reading support timetables will be collected for analysis.

This study, draws on the multiple case study design developed by Yin, Bateman and Moore, (as cited in Yin, 1994). The case study design allows small samples of children to be drawn from a variety of support programs, thereby giving access to a comprehensive and rich body of data for analysis.

A sample of students will be followed to every support program attended over a continuous period of a week in order to gauge the cumulative experience. This will provide a vignette of real life experience over time. Numbers chosen for the sample will depend on number of support programs offered, numbers of students receiving support, and time constraints imposed because of human limitations. For each student case study, observations will be made, lessons will be audiotaped, and lesson plans from any sites will be collected for analysis. Informal student interviews will be conducted over the period of a week, to gain insights into student perceptions of reading support.

Questions of validity have been addressed by using triangulation of multiple cases and data sources, including interviews, documents and observations, and by using multiple methodologies and strategies, including case studies, content and discourse analyses and through collection of quantitative data where relevant (Patton, 1990; Silverman, 1993; Stake, 1994).

Content analysis will be used wherever relevant information is explicit in texts and can be isolated and transferred directly from existing documents. However, discourse analysis will be used for answering questions about views of reading, teacher orientations and beliefs, student perceptions, and information which is implicit in language and practice. This will reveal the ideological stance behind the actions and utterances during reading support. This is important because these texts will represent the current discourse of reading operating for that context, and therefore, for those persons within it (Fairclough, 1992).

Other analyses will also be used, for example, quantitative analysis of time spent in reading lessons, and engaged time within time allocated for reading support. Analytical tools will be finalized depending on the type of data which emerges as the study progresses. Use of QSR*NUDIST (Qualitative Solutions and Research, 1994) will facilitate the management and analysis of the large body of data collected in interviews, lesson tapes and observations.

Outcomes based data cannot authentically reflect what is happening in schools. While it does represent an end point evaluation of program effectiveness, and reveals collective trends and standards, it does not generate information which has practical and significant effects on planning or improving provision of reading support to the children who need it. It does not provide guidelines which promote change. A more comprehensive review of effective practice is needed to address the issues surrounding reading difficulty. This research will ask the questions which identify key elements affecting support provision, through comprehensive analysis of effective practice, represented by complex interactions in real life school contexts. If we want to get answers that matter, we must ask the right questions.