What's happening in schools for primary students with learning difficulties in literacy and numeracy? A national survey.

Rohl, Mary ; Milton, Marion

The purpose of the survey reported in this paper is to provide a national picture of the provision made in regular Australian school settings for students who have learning difficulties in literacy and numeracy. This survey formed part of a large multi-layered data set collected for the national study of learning difficulties Mapping the Territory: Primary Students with Learning Difficulties in Literacy and Numeracy (Louden et al. 2000) that was commissioned by the Commonwealth Government. (1) The impetus for this project was the Government's view that all students should be able to meet minimum educational standards measured in terms of national documents such as Benchmarks and Profiles, when appropriate strategies are used to meet their needs. However, as identified by studies such as the National School English Literacy Survey (1997 ACER) and One Hundred Children Go to School (Hill et al. 1998), there is a wide range of student ability and achievement in literacy and numeracy within Australian schools, even in the early years. As Greaves (1996) has pointed out, if students with learning difficulties are to achieve national standards in the areas of literacy and numeracy, more needs to be known about the ways in which systems, schools and classroom teachers across Australia address the specific needs of these students.

Elkins (2000) has indicated that there have been few broad-based national studies of learning difficulties in the Australian context, even though Australian interest in the field dates back to at least the middle of the last century. There has not been much knowledge, therefore, about how schools are trying to help children with learning difficulties achieve national standards. Further, there seems to be some confusion about the terms used to describe these children. In the United States the term `learning disability', which is sometimes qualified to `specific learning disability', has wide acceptance and is often tied to funding, which is not the case in most states of Australia. In the North American context `learning disabilities' are used to refer to a `heterogenous group of disorders' which are manifested by difficulty in the acquisition of literacy and/or numeracy. These difficulties are `intrinsic to the individual, presumed to be due to central nervous system dysfunction', and not the direct result of `other handicapping conditions' or `extrinsic influences (such as cultural differences, inappropriate or insufficient instruction)' (National Joint Committee on Learning Disabilities 2000). The term used by DETYA in the brief for this project was `learning disabilities', which has many similarities to the definition of `specific learning disability' in terms of exclusionary criteria, and was defined by DETYA in their tender document as: a generic term that refers to a heterogeneous group of students who have significant difficulties in the acquisition of literacy and numeracy and who are not covered in the Commonwealth's definition of a student/child with a disability ... Learning disability is believed to be a difficulty that is intrinsic to the individual and not a direct result of other conditions or influences (no page number given).

However, the Cadman report of 1976 (see Elkins 1983) took a non-categorical approach to definition in Australia and recommended the use of the broader term `learning difficulties'; this is also the term adopted by Snow, Burns and Griffin (1998) in their report of the United States National Research Council study into the prevention of reading difficulties in young children.

Snow, Burns and Griffin recommend the identification, as early as possible, of children who are at risk of developing learning difficulties and the implementation of early intervention programs where indicated. An assumption of their report was that `empirical work in the field had advanced sufficiently to allow substantial agreed-upon results and conclusions' (p. v) to be made in the area of prevention of learning difficulties. Accordingly, they specify from research features of programs that appear to lead to success in literacy for children in general and, in particular, to help children who are at risk for learning difficulties. Some Australian research has identified teaching strategies that are effective in producing improved outcomes in literacy and numeracy for students with learning difficulties (for example, Chan 1991, Cole & Chan 1990). Australian research also suggests that, to be effective, teaching strategies need to be embedded in literacy programs that have a whole-school focus (Hill 1997).

Thus, a whole-school focus, early identification and the use of appropriate programs and teaching strategies have been shown to be important in teaching children with learning difficulties. However, knowledge about such research findings may not have reached the majority of classroom teachers. A study by Milton and Rohl in 1998 suggested that teachers in the early school years perceived that they lacked knowledge about effective teaching strategies and diagnostic procedures with which to identify young children at risk of developing learning difficulties.

Closely related to teacher knowledge is the issue of teacher education. The research of Ferguson (1991) showed that the engagement of highly qualified staff was a most effective way of increasing school performance. Of course, schools' access to highly qualified staff and most other resources is dependent upon funding, although Chan and Dally (2000) in a review of research have identified resources that do not take up large amounts of school funds.

Thus, in view of research findings about learning difficulties and the Government's press for all children to meet national standards in literacy and numeracy, there was a need for a national study that addressed the area of learning difficulties. The issues addressed in the survey, at the school level and in relation to students with difficulties in literacy and numeracy, were: terminology, identification, prevalence, strategies and programs, evaluation, funding, staff training and school policy.

Methodology

The school principal, or other designated person, from 377 schools around Australia participated in this study. A questionnaire was designed in a `tick a box' format for ease of completion by school principals. It was piloted and modified where indicated by a trial group of school principals. The modified version of the questionnaire that was used in the study can be found in Appendix 1.

A database of 8,199 schools that contained primary school-aged children was obtained from the Australian Principals' Association Professional Development Council. For the purpose of this project, schools in the database that could be identified as `special schools' were deleted, as the focus was students with learning difficulties in mainstream primary schools. Questionnaires, with a covering letter and reply-paid envelope, were mailed to the principals of a random sample of 1,000 of the schools from the modified database. A reminder was mailed two weeks later.

Three hundred and ninety-two questionnaires were returned. Of these, 15 were not included in the analysis for one of the following reasons: the school did not contain students in the primary age range, the school had closed, or the school was a special school not identifiable as such from the database. Accordingly, the responses from 377 questionnaires were analysed. It can be seen in Figure 1 that our sample was generally representative of the whole database in terms of state, although Western Australia is somewhat over-represented, accounting for 17 per cent of our sample, but only 10 per cent of the database. The sample was also generally representative of school sector and school size.

[FIGURE 1 OMITTED]

Data analysis and presentation

Results are presented in the form of the percentages of schools in the survey sample that nominated each response to an item. The multiple response option has resulted in percentages adding to numbers greater than 100 for most items. Where items contained an `other, please specify' response category, all responses made by 10 or more respondents are reported. Comparisons were made between states and sectors for each item and are presented here where particularly noticeable differences occurred. While the proportions of respondents from the individual states in our sample matched well with the whole database, the size of our sample resulted in a small number of respondents from states with relatively small populations. Tasmania is represented in our sample by 11 schools, Australian Capital Territory by 8 schools and Northern Territory by 7 schools. In order to provide a comprehensive national picture of learning difficulties we thought it most important not to exclude these states from cross-state analyses. Nevertheless, in view of the small sample sizes, generalisations about these states should be treated with caution.

Results

Terminology

Figure 2 shows the most common terms used by schools in our survey. The most commonly used term was `learning difficulties', with the next most commonly used terms being `students at risk/students at educational risk' and `special needs/in need'. Only one in ten schools indicated that they used the term `learning disabilities'.

[FIGURE 2 OMITTED]

Despite our trialling of the questionnaire, some respondents apparently misread the question that asked about the terms used to identify students with learning difficulties, giving uninterpretable responses. One group (10%) focused on the word `identify' and responded with a particular assessment method, whilst another group (9%) confused the word `term' with school term and gave responses such as `Term 1' or `Term 4'.

In a state-by-state breakdown some differences in the use of terms were evident. In NSW, Queensland and SA, the most commonly used term was `learning difficulties', whereas in Victoria, WA, Tasmania and NT `students at risk' or `students at educational risk' were the most commonly used terms. ACT schools used a variety of terms.

Identification and prevalence

Respondents were asked to indicate the criteria used in their schools to identify students with difficulties and the times when identification took place. Some schools indicated that they used more than one criterion for identification of students. The most commonly cited criterion was `2 or more years behind' (49%), closely followed by `1 year behind' (45%), and the `bottom 10%' (25% of schools). Of the 24% of schools to nominate the `other' category, the most common response was `teacher judgement'.

Several items on the questionnaire related to the prevalence of literacy and numeracy difficulties. In one item respondents were asked to estimate the proportion of students in their schools who were experiencing difficulties in literacy and numeracy on a scale of `none', `less than 10%', `10-30%', `more than 30%'. They were also given the option of a `don't know' category. It can be seen in Figure 3 that literacy difficulties were perceived to be somewhat more prevalent than numeracy difficulties.

[FIGURE 3 OMITTED]

Another questionnaire item that addressed prevalence, this time without distinguishing between literacy and numeracy, occurred early in the survey, where respondents were asked to estimate the number of students with learning difficulties in their school. The proportion of students said to be experiencing difficulties, as a percentage of the total number of primary students in the school, ranged from 0-100%, with a mean of 16% (n = 301; not all schools provided information in an interpretable form). This mean proportion of 16% falls within the 10-30% category nominated by most schools for literacy and numeracy difficulties, but suggests a lower prevalence of these difficulties than could be determined from the category question alone.

A comparison between states indicated some differences in the proportions of children estimated to be experiencing difficulty. The proportion of respondents who estimated less than 10% of children in their schools with literacy difficulties varied from 0% in the NT to 31% in WA. The proportion of schools estimating that 10-30% of their students had difficulty with literacy ranged from 57% in NT to 90% in the ACT, while the proportion of schools estimating literacy difficulties at above 30% ranged from 0% in the ACT and Tasmania to 43% in the NT. With regard to students having difficulty in numeracy, the proportion of schools estimating less than 10% of their students as having difficulty in this area varied from 0% in the NT to 46% in WA. For those schools estimating that 10-30% of their students had difficulty in numeracy, the proportion varied from 46% in WA to 88% in the ACT. Finally, for those schools estimating that over 30% of their students had difficulty in numeracy, the range was the same as for literacy, with the proportion varying from 0% in Tasmania and the ACT to 43% in the NT.

Schools were also asked to indicate the prevalence of a range of conditions that had been identified by a specialist. The conditions reported as being most commonly identified by a specialist were speech/ language problems and ADD/ADHD, followed by specific learning disabilities, behavioural disorders, perceptual motor problems, otitis media and dyslexia (see Figure 4). Other conditions reported by more than 10 respondents were Autism/Asperger's syndrome, intellectual impairment including Down syndrome, and physical impairment.

[FIGURE 4 OMITTED]

A few differences were noted between states. The NT had the highest proportion of respondents indicating that speech/language problems (100%) and otitis media (57%) had been diagnosed in students in their schools. The ACT had the largest proportion of schools indicating diagnosis of perceptual motor problems (89%) and dyslexia (56%), whilst the state with the lowest proportion of schools indicating diagnosis of perceptual motor problems was SA (23%). Queensland had the lowest proportion of schools indicating diagnosis of dyslexia (6%).

There was an interesting pattern of results across states for behavioural disorders and Attention Deficit Disorder/Attention Deficit Hyperactivity Disorder (ADD/ADHD). Whilst only 57% of NT schools indicated that ADD/ADHD had been diagnosed, 86% indicated diagnosis of behavioural disorders. WA schools indicated an average level of ADD/ADHD diagnosis, but the lowest level of diagnosis of behavioural disorders (27%), whereas a high proportion of ACT schools (89%) indicated diagnosis of both ADD/ADHD and behavioural problems.

Sector responses reflected those of the whole sample, except that independent schools were more likely to indicate diagnosis of ADD/ADHD (90%) and more Catholic schools (85%) and government schools (79%) indicated diagnosis of speech/language problems than did independent schools (66%).

Assessment

The schools indicated that they assessed various aspects of literacy and many assessed numeracy. Not so many schools indicated that they assessed oral language or IQ (see Figure 5). More than three-quarters of schools indicated that they assessed reading comprehension, spelling and word identification, with just under three-quarters indicating that they assessed phonological awareness and numeracy. Oral language and IQ were said to be assessed by around one-third of schools.

[FIGURE 5 OMITTED]

A sector level comparison showed differences in the proportions of schools indicating IQ assessment, with 52% of independent schools indicating this type of assessment, compared to 37% of Catholic and 31% of government schools. Similarly, more independent schools (86%) indicated they assessed numeracy compared to Catholic schools (70%) and government schools (69%) and again, more independent schools indicated they assessed oral language (48%) than Catholic (34%) and government schools (35%).

Fifty-six per cent of respondents gave the names of some tests they used, the most frequently cited measure of IQ being `Wechsler Intelligence Scale for Children' (41 schools). For numeracy the most frequently cited test was `Progressive Achievement Tests-Maths' (21), followed by `DMT/Schleiger' (18). The most frequently cited measures for various aspects of literacy are set out in Table 1. It is noted that some respondents did not specify which particular part of a test battery they used to assess specific aspects of literacy. For example, some respondents nominated `Marie Clay's Observation Survey' as a measure of individual aspects of literacy, such as phonological awareness or' word identification, with the result that it is not possible to ascertain which particular subtest was used by the school to assess a particular aspect of literacy. Similarly, `Waddington' was frequently cited as a test for reading comprehension, spelling, phonological awareness and `other' categories. In Table 1 these responses have been marked `Marie Clay' or `Waddington not specified'. No measure of oral language was cited by 10 or more respondents.

The most frequently listed assessments in the `other' category included `Marie Clay Observation Survey' (13) and `Basic Skills Test' (11).

As for the times at which testing was carried out, the most common time, nominated by 43% of schools, was `end of Year 1', followed by `Year 2 diagnostic testing' (32%) and `testing at school entry' (27%). `Testing at other times' was indicated by nearly half the schools (44%). Specific explanations of testing at other times given by 10 or more schools were classified as: teacher judgement (21), annual standardised testing (15), beginning/end of year testing (14), on-going assessment (13), basic skills testing (12). In a statewise comparison, the most common response in NSW (53%) and Victorian (59%) schools was `end of Year 1 testing' while the majority of Queensland schools (76%) nominated `Year 2 diagnostic testing'.

The majority of respondents indicated that they relied on various personnel, within and outside the school, to assess students with learning difficulties (see Figure 6). Most schools indicated that classroom and specialist teachers were involved in assessment. Other professionals were also highly involved in this process, with most schools indicating the involvement of speech therapists/pathologists and school psychologists/guidance officers. Further, occupational therapists, medical practitioners and outside psychologists were involved in assessment in around half the schools and school nurses in about one-third.

[FIGURE 6 OMITTED]

Some differences were noted between states in the involvement of particular personnel in the assessment of students. For example, Queensland schools (92%) and NT schools (100%) were the most likely to indicate the involvement of class teachers in assessment. The involvement of specialist teachers was indicated by 89% of ACT schools, 86% of NT schools and 87% of Queensland schools, but they were less likely to be involved in assessment in Tasmania (54%), SA (53%) and WA (50%). Another notable difference between states was that 90% of WA schools and 89% of ACT schools, but only 14% of NT schools indicated the involvement of school psychologists. Further, whilst Victorian (55%) and WA schools (52%) indicated considerable school nurse involvement, school nurses were involved in assessment in only 3% of SA schools. WA was the least likely state to involve a medical practitioner (29%) and an outside psychologist (27%), while use of an occupational therapist in assessment was least likely in Victoria (31%) and Tasmania (36%).

A few differences were found between sectors. Government schools (84%) were more likely to use a school psychologist/guidance officer to identify children than Catholic (58%) and independent schools (50%). Catholic (90%) and government schools (80%) were more likely to use a speech pathologist/therapist in identification than were independent schools (65%).

Support programs

In terms of administration of support programs it will be seen in Figure 7 that most schools, were offering in-class support, small group withdrawal and individual withdrawal. Nearly one-third of the schools said that they offered whole-school or whole-class programs for students with learning difficulties.

[FIGURE 7 OMITTED]

Some differences between states in program administration were noted. Notably, in WA only 50% of schools offered individual withdrawal, compared to over 90% of schools offering this form of support in Queensland and SA. The use of small-group withdrawal also varied across states, with a low of 65% of WA schools offering this form of support, compared to 100% of schools in ACT and over 90% in Queensland and Tasmania. In-class support was offered by between 70% and 90% of schools in all states, apart from NT where the proportion was 57%. There were also some differences between states in the proportion of schools offering whole-school programs (from 23% in WA to 56% in ACT) and whole-class programs (from 9% in Tasmania to 38% in Queensland).

The most widely used commercially available support programs, nominated by almost hall the schools, were `First Steps' and `Reading Recovery' (see Figure 8). `The Early Years Literacy Program', formerly `Keys to Life', was used by nearly one-fifth of schools. One-third of schools claimed to be using some form of direct instruction program and perceptual motor programs were said to be used in just under one-third of schools to support students with learning difficulties. Many respondents indicated that their schools were operating support programs other than those specifically named in the questionnaire. The programs nominated by 10 or more schools were: `Support-a-Reader/Writer'; `Individual Education Plan/IEP', `THRASS', `Bridging the Gap' and `Macquarie University Reading Program'.

[FIGURE 8 OMITTED]

Whilst a great many schools were using very specific programs and approaches to support the literacy development of students with learning difficulties, support programs in numeracy were operating in only a small proportion of schools. No numeracy support program was nominated by 10 or more schools.

As would be expected in view of the involvement of some state departments of education in the evolution of particular programs, the states varied in the proportions of schools using specific programs. For example, 85% of schools in WA were using `First Steps' (an Education Department of WA initiative) and 76% of Victorian schools were using the `Early Years Literacy Program' (a Victorian Ministry of Education initiative). Nevertheless, 89% of schools in ACT were also using `First Steps', but only a small proportion of schools in states other than Victoria were using the `Early Years Literacy Program'. The proportion of schools in different states using `Reading Recovery', an individualised program from New Zealand, varied from 15% in WA to 78% in Victoria. Of the more generic programs, the proportion of schools using direct instruction approaches varied from 19% in WA to 43% in NT. The proportion of schools using perceptual motor programs varied across states from 8% in NSW to over 30% in WA, Queensland and Victoria.

A variety of personnel were involved in program delivery (see Figure 9). In most schools the class teacher, a teacher assistant and specialist teacher were involved. Parents, students and other professionals, namely psychologists, guidance officers, occupational therapists and speech pathologists were also involved in some schools.

[FIGURE 9 OMITTED]

The pattern of deployment of specific personnel to support programs was not consistent across states and some quite large differences were observed. Whilst class teachers were said to be involved in support programs in most schools in all states, the use of teacher assistants varied from 43% of schools in NT to 100% in Tasmania, and the use of specialist teachers from 58% in WA to over 90% in Queensland and Tasmania. States also differed in their schools' use of other professionals, with a range for speech pathologists from 3% in WA to 20% in SA, for occupational therapists from 9% in Tasmania to 25% in Victoria and 26% in WA, and for psychologists/guidance officers from 17% in SA to 44% in Queensland, ACT and Victoria. The involvement of other students in support programs varied from 13% in WA and Victoria to 90% in ACT, while the involvement of parents varied from 28% in Tasmania to 62% in NSW.

Some differences were apparent between sectors. Catholic schools were the most likely to involve teacher assistants (89%) and occupational therapists (37%), while independent schools were the least likely to involve teacher assistants (62%) and psychologists/guidance officers (19%).

All but 5% of schools indicated that they evaluated their programs in some way, the most common methods being professional judgement (66%), standardised testing (65%) and school-made assessments (63%); with some calling in the help of outside experts (25%).

Schools nominated several methods of evaluating individual children's progress, the most common being the use of published diagnostic/standardised measures (68%) and professional judgement (67%), followed by school-made assessments (59%) and assessment materials linked to a professional development package (56%). Six per cent of schools indicated that other means of evaluation were used, with no method being cited by more than three schools. Two percent of schools indicated that they did not evaluate the progress of individual students.

Some differences between states were noted. The ACT and NSW had the highest proportion of respondents indicating the use of published diagnostic measures (81%). The lowest use of these measures was indicated by the NT (43%) and Victoria (44%). Victorian and ACT schools indicated that they had the highest proportion of schools using assessment materials linked to a professional development package (78%), while NSW schools reported the lowest use of this form of assessment (31%).

Funding

Twenty per cent of respondents indicated that their schools received no additional funding specifically for students with learning difficulties. Where schools did receive funding, it was obtained it in a variety of ways, the most frequent response (36%) being funding based on identified needs within the school. Other relatively common ways in which funding was allocated were: based on assessment of students' school performance (29%); based on an index of disadvantage (28%); and based on a successful project proposal (17%). `Integration funding' was the only item nominated in the `other' category by 10 or more respondents.

Staff training and professional development

The specialist training of school staff can be seen in Figure 10. Slightly over half the respondents indicated that a specialist teacher in their schools had received training in the area of learning difficulties. Less than half indicated that a classroom teacher or a support staff member had specialist training in the area, with some indicating no teachers in the school with specialist training in the area of learning difficulties. For those respondents who indicated the `other' category, the only response by 10 or more schools was `principal' and/or `deputy principal' (18).

[FIGURE 10 OMITTED]

A cross-state analysis showed that NT (57%) and Victorian (54%) classroom teachers were somewhat more likely than their counterparts in other states to have had specialist training, as were specialist teachers in ACT (100%), NT (71%), NSW (69%) and Queensland (66%). Responses indicated that WA had not only the lowest proportion of schools with a specialist teacher who had training in learning difficulties (26%), but also the highest proportion of schools to have no teachers with specialist training (26%) on staff. Only 14% of NT and 27% of WA schools noted that a support staff member had such training.

Respondents were asked about their schools' involvement in professional development in the area of learning difficulties over the previous two years. Sixty two per cent indicated that professional development had been provided by the school sector or state government. Learning difficulties organisations had provided professional development to 23% of schools, while 17% of schools had experienced professional development from private consultants and 10% from universities. Eighteen per cent of schools indicated that their staff had experienced no professional development in the area of learning difficulties during the previous two years. Six per cent of schools indicated other provision, the only response indicated by more than one school being training provided by specialist staff or other staff within the school (17 respondents).

Differences between states for professional development in learning difficulties delivered by school sector or state government ranged from 43% of schools in NT to 73% in SA. No schools in the NT had experienced professional development from learning difficulties organisations while over 30% of schools in NSW had received professional development from this source. Professional development by private consultants varied from 0% of schools in ACT, NT and Tasmania to 29% in WA. University provided professional development varied from 0% in the NT to 33% in the ACT.

Policies

Sixty-four per cent of schools said that they had a written policy for students with learning difficulties. However, the proportion of schools within each state to have a written policy varied from 50% in Tasmania to 89% in the ACT. Likewise the proportion of schools within a sector to have a written policy for these students varied. A smaller proportion of Catholic (50%) than government (68%) and independent schools (70%) indicated that they had a written policy specifically for students with learning difficulties.

Discussion

Before discussing the survey results it is important to note some features of the study. Firstly, the data that have been presented here represent a relatively broad-based picture of what was happening in schools for primary students with learning difficulties in literacy and numeracy. The Mapping the Territory study (Louden et al. 2000), of which this survey is a part, examined issues at varying levels of specificity, the most specific being that of case studies of schools identified as successfully catering for these students. Secondly, the questionnaire responses represent the perceptions of school principals or their nominees about what was happening in their schools at a particular moment in time. Thirdly, because our sample was chosen to be representative of Australian schools, Tasmania, the ACT and the NT are represented by relatively few schools. It should also be noted that the survey was confined to primary school-aged students.

Various terms were used by schools to identify students with learning difficulties and many used more than one term within their school. `Learning difficulties' was used by almost half of respondents and `students at risk' or `students at educational risk' were used by just over one-third. In contrast the term `learning disabilities', used by DETYA in the project brief was used by only one in ten schools. This use of inclusive terms by schools is in accordance with the claim by Elkins (2000) that most government and non-government schools in Australia show concern for all students who, for whatever reason, have difficulty in learning literacy and numeracy. There was evidence for this in the case studies of individual schools that were made for the Mapping the Territory study. Case study schools identified a wide range of students who were experiencing learning difficulties (Rohl et al. 2000). Some were of low general cognitive ability; some had behavioural problems often diagnosed as ADD/ADHD; some were from particular linguistic, cultural or family backgrounds; and a proportion of the identified students would have demonstrated the criteria for inclusion in exclusionary definitions of learning disability (DETYA 1998, NJCLD 2000).

The use of inclusive terms by schools also explains the prevalence estimates. Over half indicated that between 10 and 30% of their students were experiencing difficulty with literacy and numeracy, although literacy difficulties were seen as being somewhat more prevalent than difficulties in numeracy. This broad estimate of prevalence was refined when respondents were asked about specific numbers of students in their school with learning difficulties. Prevalence was then calculated at 16%, a figure that is comparable to those obtained in some previous Australian studies (Andrews et al. 1979, Milton & Rohl 1998) and is far more than the 6% figure quoted by Elkins (2000) for US students labelled by the exclusive term `learning disabilities'.

The schools indicated that they used various criteria for identification of students with learning difficulties and the data suggest that in most schools identification procedures were in place by the end of Year 2. Nevertheless, the figure of fewer than one-third of schools reporting assessment at school entry should be seen as of concern: Snow et al. (1998) conclude that `children who are at risk for reading difficulties should be identified as soon as possible' (p. 318). Further, only around one-third of schools indicated assessment of oral language, even though most schools contained students with speech and language problems that had been identified by specialists. Since early speech and language difficulties have been shown to have a strong relationship with later learning difficulties (Snow et al. 1998), it seems that assessment of all children's oral language early in their school career should be a high priority.

Over half the schools gave the names of the particular tests that they used for identification. Of particular interest is The Observation Survey (Clay 1985), one of the most commonly cited measures of phonological awareness. This survey contains a number of subtests, none of which specifically addresses phonological awareness per se. (The Dictation subtest measures children's ability to relate sounds heard in words to the letters with which they are represented, or their ability to memorise word spellings.) This suggests that the amount of phonological awareness testing in schools may have been less than indicated by our respondents and that some school principals or their nominees may not have understood the term `phonological awareness' exactly.

All schools offered some form of program to support students with difficulties in literacy and numeracy. Most indicated that they offered in-class support and small-group and individual withdrawal and around one-third noted that they offered whole-school and whole-class programs for students with learning difficulties. A variety of personnel were involved in program delivery. Nevertheless, the manner of program delivery and the involvement of all personnel, apart from class teachers, varied as a function of state and sector. In terms of specific programs used for support the most popular were `First Steps' and `Reading Recovery', both of which were used by very nearly half of all schools. The use of these programs varied widely across states, most likely because of the requirements and funding arrangements of particular state education departments. Direct instruction and perceptual motor programs were used by nearly one-third of schools. However, support programs in numeracy were taking place in only one out of seven schools.

It seems important to make some observations on the support programs used in Australian schools. Firstly, there are important differences between programs shown as being the most widely used. `First Steps' is essentially a broad-based `first wave' teaching program that provides teachers with strategies matched to children's developmental levels in terms of indicators of progress. It was designed to be used by classroom teachers for all children across a range of early childhood and primary year levels in the areas of oral language, reading, writing and spelling. `Reading Recovery', on the other hand, was designed as a `second wave' teaching program for children in their second year of school who have been identified as being `at risk' on the basis of very specific observations of reading and writing behaviours and it is carried out by specially trained teachers in a one-to-one withdrawal situation. Many direct instruction programs have been designed for use with all learners, including `third wave' learners who still experience difficulty after exposure to both normal class instruction and early `acceleration' programs such as `Reading Recovery'.

A second observation is that previous research has shown some programs to be more effective than others in leading to academic achievement for students with learning difficulties. Specifically, research over many years has consistently shown perceptual motor programs to be ineffective for these students (Pearson 1996), while some other research has shown that direct instruction approaches can lead to increases in particular skills (Swanson et al. 1996). Research on the effectiveness of `Reading Recovery' shows short-term gains (Centre et al. 1995) but there has been some doubt about whether these gains are maintained in the long term (Chapman et al. 1999). Further, neither `First Steps' nor `Reading Recovery' has a strong emphasis on the structured explicit teaching of phonological awareness that much research has shown to be essential for decoding and encoding words (Adams 1990).

A third observation is that, whilst many schools were implementing support programs in literacy, only a small number of schools were implementing numeracy support programs. The prevalence data show that schools estimated somewhat fewer students as having difficulties in numeracy than in literacy. Nevertheless, given that over half the schools estimated more than 10% of their students to be experiencing difficulty in numeracy, it is surprising that more were not implementing programs to address the specific needs of these students.

The survey also addressed the issue of evaluation. Professional judgement, standardised testing and school-made assessments were used by approximately two-thirds of schools for evaluation of programs and individuals. For student evaluation, just over half the schools used assessment measures linked to a professional development package and external experts were used to evaluate programs in a quarter of all schools. Of concern is that one in 20 schools indicated that they conducted no evaluation of programs and one in 50 that they did not evaluate student progress.

In order to carry out programs and monitor individual children's progress over time, it is necessary for schools to have access to regular funding. One in rive schools indicated that they did not receive any additional funding for students with learning difficulties. One-third indicated that their funding was based on identified need, just over one-quarter that it was based on assessment of students or on an index of disadvantage, and nearly one-fifth that it was based on a successful project proposal. These findings suggest that many schools may not have access to regular funds for their support programs. Evidence from case studies of individual schools, indicated that lack of regular funding was an ongoing problem and that schools had to make difficult decisions as to which students they could afford to support (Rohl 2000).

Much research points to the centrality of the teacher in any program. Snow et al. (1998) state that `excellent teaching is one of the most effective means in preventing reading difficulties' (p. 122) and they see teacher education and training as a `career-long undertaking'. Only about half the schools in our survey indicated that they had a specialist teacher, classroom teacher and/or support staff member with training in the area of learning difficulties. More than one in ten schools had no staff at all with training in the area. All schools in our sample from the ACT had a specialist teacher with training, as did more than two-thirds of schools in NT, NSW and Queensland. WA schools were the least likely to contain any teachers with specialist training in learning difficulties. Given that, with only very few exceptions, children with learning difficulties in the Australian context are integrated into regular classes, it is of paramount importance that all schools contain personnel who are trained to cater for these children's needs.

In terms of professional development, nearly one in five schools had no staff member who had experienced professional development in the area of learning difficulties during the previous two years. Where professional development was provided, the most popular source was school sector/state government, with nearly two-thirds of schools nominating this source. Other sources were learning difficulties organisations, private consultants and, for a few schools, universities were the providers. Whilst the provision of professional development appears to be greater than that of formal training, much school-based professional development consists of a single half-day, with no follow-up (van Kraayenoord & Treuen 2000).

The final area of this discussion is that of school policy. Almost two-thirds of schools indicated that they had a written policy for students with learning difficulties. There was, however, no way of ascertaining from the questionnaire whether or not these policies were put into practice. Our case study data suggested that schools identified as successfully dealing with students with learning difficulties did, in fact, have a school policy for these children that was implemented at a whole school level.

Conclusions

If seems that all but a few schools were assessing literacy, with reading comprehension, spelling and word recognition assessed by most. Many schools were also assessing numeracy. In most schools classroom teachers were assisted in the identification of students with learning difficulties by various other professionals. Much of this identification took place relatively early in a child's school career and the majority of schools were conducting literacy support programs that were evaluated, as was children's progress. Around hall the schools had a specialist teacher with training in the area of learning difficulties and teachers from many schools were receiving professional development in the area. A majority of schools had written policies and received some additional funding for students with learning difficulties, who made up approximately 16% of the schools' population. The proportion of students considered to be having difficulty with numeracy was somewhat lower than that for literacy, but still lay somewhere between 10 and 30% in over half the schools.

Some areas of concern were identified by the survey, several of which relate to assessment. Firstly, many of the measures used in schools are over 20 years old and may need some revision to reflect current literacy theory and practice. Secondly, it seems that some school administrators may not have understood exactly what was being assessed by particular measures. Thirdly, relatively few schools appeared to be assessing oral language, although most schools who took part in our survey contained students who had been assessed by a specialist as having speech and language difficulties. Fourthly, although there seemed to be widespread assessment of numeracy, and a perception by principals that a sizeable proportion of students had difficulties in this area, few schools had numeracy support programs in place. (Milton 2000, discusses the issue of numeracy in more detail.)

A particularly important issue is the relative effectiveness of the various programs being used with students who have learning difficulties. The principals indicated that two programs, `First Steps' and `Reading Recovery', were being used in almost half the schools to support such students. These programs are very different in terms of aims, content, mode of delivery and cost for each student and neither contains an emphasis on the structured teaching of the sound structure of words. Nevertheless, we could find little research into the relative effectiveness and efficiency of these and other programs that were widely used with children who had difficulties in literacy. Further, despite the fact that much research has shown perceptual motor programs to be ineffective in increasing literacy skills in children with learning difficulties, almost one-third of schools indicated that they were using such programs for this purpose. Another issue related to programs is that only around half of the schools in the survey indicated that they had on staff a specialist teacher with training in the area of learning difficulties, and in some schools there were no teachers at all with any training in the area.

Of particular concern is the issue of equity of access to appropriate educational services. Whilst there appeared to be a few differences between school sectors, there seemed to be some large differences between both schools and states in all areas addressed by the survey. It can thus be seen that the results of this survey, when analysed in terms of research literature and other data from the Mapping the Territory study, have raised some concerns about provision in Australian schools for students with learning difficulties. On the basis of these results we make the following recommendations for action:

* In order to provide accurate prevalence rates and to remove confusion in the community there needs to be standardisation and clarification of the terms used by federal and state departments of education and schools to refer to students who have difficulties in learning literacy and numeracy

* All schools need to have in place a policy that relates to students with learning difficulties that is put into practice at a whole school level

* All schools need to have in place systematic procedures for assessing oral language, literacy and numeracy in order to identify students with learning difficulties as early as possible in their school careers so that appropriate intervention can be implemented for children who need it

* All schools need access to staff who have specialised knowledge in the area of learning difficulties, with particular reference to appropriate assessment measures and teaching strategies

* All schools need the resources with which to provide for their students who have learning difficulties

* Research needs to be conducted into the relative effectiveness of the programs most widely used for supporting students with learning difficulties

* All school systems and sectors need to provide support for teachers who wish to undertake further study in the area of learning difficulties

* There is a need for an examination of pre-service teacher education programs to ensure that all students have access to courses that prepare them to diagnose and teach students with learning difficulties.

We acknowledge that some systems, schools and teacher education institutions are putting into practice many of these recommendations. Nevertheless, the results of our survey showed differential provision across the schools and states of Australia for children with learning difficulties in terms of identification, programs, evaluation, funding, training of support personnel and the availability within schools of a written policy to address the needs of these students. Figure 1: National and survey school samples by state Total database 8199 schools by state Total database 8199 Sample of 377 schools by state schools by state WA 10% 17% ACT 1% 2% NSW 32% 28% NT 2% 2% QLD 19% 19% SA 9% 8% TAS 3% 3% VIC 21% 21% Note: Table made from pie chart. Figure 2: Terminology used by schools Percentage of schools nominating each item learning difficulties 47 students at risk 32 other 25 special needs/in need 17 learning disabilities 10 confused with assessment 10 confused with school term 9 students at educational risk 5 Note: Table made from bar graph. Figure 4: Conditions identified by specialists Percentage of schools nominating each item speech/language problems 79 ADD/ADHD 78 specific learning disabilities 64 behavioural disorders 46 perceptual motor problems 36 otitis media 24 dyslexia 21 other 14 Note: Table made from bar graph. Figure 5: Areas of formal assessment Percentage of schools nominating each item reading comprehension 89 spelling 84 word identification 78 phonological awareness 72 numeracy 72 oral language 37 IQ 36 Note: Table made from bar graph. Table 1: Literacy assessment measures nominated by 10 or more schools Aspect of Literacy Measure Number of respondents Reading comprehension TORCH (Tests of Reading Comprehension) 78 Neale Analysis of Reading Ability 58 Waddington 30 Phonological Awareness Marie Clay (not specified) 18 Waddington (not specified) 10 Word identification Burt 29 Marie Clay (not specified) 19 Oral language No measures nominated by 10 or more respondents Spelling South Australian/Westwood Spelling Test 1 Waddington 42 Schonell Spelling Test 16 Figure 6: Professionals involved in assessment Percentage of schools nominating each item teacher 81 speech pathologist/therapist 80 school psychologist 74 specialist teacher 69 occupational therapist 52 medical practitioner 51 outside psychologist 46 school nurse 34 other 10 Note: Table made from bar graph. Figure 7: Types of support programs offered Percentage of schools nominating each item individual withdrawal 82 small group withdrawal 81 in class support 80 whole school program 31 whole class program 28 other 10 Note: Table made from bar graph. Figure 8: Programs used for support Percentage of schools nominating each item First Steps 49 Reading Recovery 46 Other 33 Direct Instruction 32 Perceptual Motor Program 26 Early Years Literacy Program 19 Numeracy Program 14 Note: Table made from bar graph. Figure 9: Personnel delivering support programs Percentage of schools nominating each item class teacher 87 teacher's aide 78 specialist teacher 73 parents 51 psychologist/guidance officer 33 other students 21 occupational therapist 21 speech pathologist/therapist 11 other 9 Note: Table made from bar graph. Figure 10: Staff with specialist training Percentage of schools nominating each item specialist teacher 54 classroom teacher 44 support staff 41 none 13 other 9 don't know 1 Note: Table made from bar graph.

Acknowledgement

We would like to thank Danielle Brady for her help with survey design and analysis, and the school principals who completed the surveys.

(1.) The project was funded under the Literacy and Numeracy Programme administered by the Commonwealth of Education, Training and Youth Affairs (DETYA). The views expressed herein do not necessarily represent the views of DETYA.

References

Australian Council for Educational Research 1997, `Mapping Literacy Achievement: Results of the 1996 National School English Literacy Survey', Australian Council for Educational Research, Camberwell.

Adams, M.J. 1990, Beginning to Read: Thinking and Learning about Print, MIT Press, Cambridge, Massachusetts.

Andrews, R.J., Elkins, J., Berry, P.B. & Burge, J.A. 1979, `A survey of special education in Australia: Provisions, needs and priorities in the education of children with handicaps and learning difficulties', Fred & Eleanor Schonell Educational Research Centre, St Lucia, Queensland.

Centre, Y., Wheldall, K., Freeman, L., Outhred, L. & McNaught, M. 1995, `An evaluation of Reading Recovery', Reading Research Quarterly, vol. 30, no. 2, pp. 240-263.

Chan, K.S. 1991, `Promoting strategy generalisation through self-instructional training in students with reading disabilities', Journal of Learning Disabilities, vol. 24, no. 7, pp. 427-432.

Chan, L.K.S. & Dally, K. 2000, `Review of literature', in Mapping the Territory: Primary School Students with Learning Difficulties in Literacy and Numeracy, vol. 2, W. Louden, L. K. S. Chan, J. Elkins, D. Greaves, H. House, M. Milton, S. Nichols, J. Rivalland, M. Rohl & C. van Kraayenoord, Department of Education, Training and Youth Affairs, Canberra.

Chapman, J.W., Tunmer, W.E. & Prochnow, J.E. 1999, Success in Reading Recovery depends on the development of phonological processing skills, Revised Research Report, Ministry of Education, New Zealand.

Clay, M.M. 1985, The Early Detection of Reading Difficulties, Heinemann, Portsmouth.

Cole, P.G. & Chan, L.K.S. 1990, Teaching Methods and Techniques in Special Education, Prentice-Hall of Australia, Sydney.

Elkins, J. 1983, `The concept of learning difficulties: An Australian perspective', in Current Topics in Learning Disabilities, vol. 1, eds J.D. McKinney & L. Feagans, Ablex, Norwood, New Jersey.

Elkins, J. 2000, `All empires fall, you just have to know where to push: Antecedent issues of a study of learning difficulties in Australia', Australian Journal of Learning Disabilities, vol. 5, no. 2, pp. 4-7.

Ferguson, R. 1991, `Paying for public education: New evidence on how and why money matters', Harvard Journal on Legislation, vol. 28 (Summer), pp. 465-498.

Greaves, D.F. 1996, `Children with learning disabilities have special needs', in Learning Disabilities: Issues for the Nineties, ed. K. Healy, The Spinney Press, Sydney.

Hill, S., Comber, B., Louden, W., Rivalland, J. & Reid, J. 1998, 100 Children Go to School, DETYA, Canberra.

Hill, P.W. 1997, The literacy challenge in Australian primary schools, paper presented at the APPA/ACPPA National Conference, Sydney, October.

Louden, W., Chan, L.K.S., Elkins, J., Greaves, D., House, H., Milton, M., Nichols, S., Rohl, M., Rivalland, J. & Van Kraayenord, C. 2000, Mapping the Territory: Primary Students with Learning Difficulties: Literacy and Numeracy, vols. 1-3, Department of Education, Training and Youth Affairs, Canberra.

Milton, M.J. 2000, `How do schools provide for children with difficulties in numeracy?' Australian Journal of Learning Disabilities, vol. 5, no. 2 pp. 23-27.

Milton, M.J. & Rohl, M. 1998, `Children K-2 who are of concern to their teachers: Identification, programs and the professional needs of teachers', Australian Journal of Learning Disabilities, vol. 3, no. 1 pp. 9-20.

National Joint Committee on Learning Disabilities 2000, Learning Disabilities: Issues on definition (1990, January [Online]. Available: http://www.ldonline. org/njcld/defn_91.html

National Research Council 1999, Starting out Right: A Guide to Promoting Children's Reading Success, National Academy Press, Washington DC.

Pearson, P.D. 1996, Preface to `Perceptual-motor activities in the treatment of severe reading disability', The Reading Teacher, vol. 50, no. 2, p. 88.

Rohl, M. 2000, `Programs and strategies used by teachers to support primary students with difficulties in learning literacy', Australian Journal of Learning Disabilities, vol. 5, no. 2, pp. 17-22.

Rohl, M., House, H., Louden, W., Milton, M. & Rivalland, J., 2000, Successful Programs and Strategies for Children with Learning Difficulties in Literacy and Numeracy, Department of Education, Training and Youth Affairs, Canberra. (A brochure sent to all schools in Australia.) Available: http: //www.detya.gov.au/ schools/LiteracyNumeracy/index.htm

Snow, C.E., Burns, M.S. & Griffin, P. (eds) 1998, Preventing Reading Difficulties in Young Children, National Academy Press, Washington DC.

Swanson, H.L., Carson, C. & Saches-Lee, C.M. 1996, `A selective synthesis of intervention research for students with learning disabilities', School Psychology Review, vol. 25, no. 3, pp. 370-391.

van Kraayenoord, C. & Treuen, M. 2000, `Provision of professional development in literacy and numeracy for students with learning difficulties and disabilities', in Mapping the Territory: Primary School Students with Learning Difficulties in Literacy and Numeracy, vol. 2, W. Louden, L.K.S. Chan, J. Elkins, D. Greaves, H. House, M. Milton, S. Nichols, J. Rivalland, M. Rohl & C. van Kraayenoord, Department of Education, Training and Youth Affairs, Canberra.

Mary Rohl is Co-director of the Centre for Applied Language and Literacy Research at Edith Cowan University and a senior lecturer in the School of Education, where she conducts research and teaches early childhood and postgraduate courses in language and literacy. Her research interests and publications are in the areas of early literacy, including ESL and bilingual students, teacher education in literacy and numeracy and learning difficulties.

Address: Edith Cowan University, Pearson Street, Churchlands, WA 6018 Email: m.rohl@cowan.edu.au

Marion Milton is a senior lecturer at Edith Cowan University. She is also the Coordinator of the Adult Literacy and Numeracy Research consortium in WA. Her research and teaching interests include metalinguistic awareness and literacy acquisition, literacy learning difficulties in children and adults, workplace literacy and teaching English as a second language.

Address: Edith Cowan University, 2 Bradford Street, Mount Lawley WA 6050 Email: m.milton@ecu.edu.au