The assessment of learning difficulties in literacy among children learning English as an additional language: a literature review
The focus in this review is on children who learn to read at school in their second or third language where that language is different from their first language or the language spoken in their home. The review is not mainly concerned with learning difficulties experienced by children who speak a variety of English at home that is different from that used in the classroom.
The main findings of the review are presented in the form of answers to questions that are commonly asked about work with children learning EAL who are thought to have learning difficulties in reading.
Do children learning to read in EAL experience difficulties that are different from those with which teachers are familiar in monolingual children?
The literature suggests that, when analysing the reading difficulties of children learning EAL, a crucial distinction has to be kept in mind. On the one hand, there is likely to be a substantial number of children learning EAL who experience reading
difficulties in the early stages because of linguistic and cultural obstacles which they are not always given sufficient help to negotiate. On the other hand, there is a much smaller number who will experience severe and continuing difficulties at the word level that may go undiagnosed.
Do children learning EAL with learning difficulties in literacy have access to the specialist support that they need?
There is no regular monitoring at national level of the allocation of SEN support to children learning EAL, and the available data is patchy. However, local surveys have indicated that children learning EAL are underrepresented in LEA lists of those
receiving specialist support for pupils with specific learning difficulties. For example, in one urban study 5% of the total school population was learning EAL but this group comprised only 1% of the roll of SpLD provision. The numbers are small in each individual survey, but the picture is consistent across areas. Studies of the impact of initiatives such as the National Literacy Project (directed at all pupils) and the Reading Recovery programme (directed at pupils at risk) have shown a positive outcome for pupils learning EAL in general. But there is evidence that the benefits from such initiatives are significantly reduced in the case of children at the early stages of learning English.
Should literacy learning difficulties be identified as early as possible with children learning EAL?
According to conventional wisdom, learning difficulties should be identified at the earliest possible stage. However, there are grounds for concern that, in the case of children learning EAL, the likelihood of inaccurate identification is higher than with
other children and the action of labelling them as slow learners will carry a particular risk of becoming a self-fulfilling prophecy. There is a balance to be struck between the advantages of early identification and the need to take every possible step to prevent the development of inappropriate low expectations of pupils learning EAL. There are some problems when Baseline Assessment schemes are employed to identify learning difficulties among children learning EAL. Although some authors are seeking to produce improved instruments, there are good reasons to believe that it is going to be extremely difficult to achieve reliable and valid prediction from baseline assessment, except in extreme cases. This is partly because many members of this population show rapid learning after school entry. Claims from authors and publishers that Baseline Assessment schemes can deliver valid and reliable screening of learning difficulties with multilingual populations should be treated with scepticism. It has been suggested that, instead of concentrating on identifying within-child learning problems, we should think of "teachers and carers as noticing children's individual needs and then adjusting their responses accordingly" (BPS, 1998). The advantage of this approach is that it seems likely to retain the advantages of early
"identification" while reducing its risks. The literacy skills of children who are perceived to be making limited progress can be closely monitored both in the Literacy Hour and across the curriculum. When combined with careful profiling and systematic
recording, this can lead to appropriate further action, if necessary, within the framework of the SEN Code of Practice.
What role should specialist EAL teachers, bilingual support staff and interpreters play in the identification and assessment of learning difficulties, and what training do they and others need?
While most commentators agree that this group of staff has an important role to play in collaboration with others, conflicting opinions are expressed on what that role should be and how it would best be exercised. It appears that the views of bilingual
staff themselves have not been surveyed. Both the staff themselves and outside observers have commented on a general lack of confidence and expertise in this field. No published UK surveys or reports were traced that related specifically to the identification of training needs in EAL and SEN.
Should a special test be developed to overcome the problems of assessment in this field such as cultural bias?
An analysis of the literature strongly suggests that a single test could not deal with the challenges posed by this task. Both because individual tests and observations tend to be less reliable with this population and because the performance of children learning EAL is exceptionally variable across settings, it is important to look for multiple sources of evidence wherever possible and sample children's performance and behaviour in different roles and different situations. Thus the answer to the challenges posed in this review will not be a single, simple set of assessment materials but a multifaceted strategy for assessment and consultation. Some of those working on methods of normative assessment have suggested that the problems encountered in using their materials with children learning EAL could be overcome by translating well-established tests or by developing pluralistic or local norms. The evidence suggests that the technical obstacles to the successful use of tests in translation and to the development of pluralistic or local norms are not insurmountable. However, the use of such materials will be of limited value in supporting the planning of classroom initiatives to help individual children with EAL to overcome learning difficulties. Translated tests will not meet the needs of all children from a particular linguistic community, and local norms are likely to change fast. It is doubtful that these strategies are cost-effective or educationally valid - except in a very limited range of situations. When different strategies of assessment are evaluated for their efficacy in use with children learning EAL, the approach that is usually least favourably reviewed is the one that appears to be most commonly used - normative assessment. There is a case for prioritising research and development work on curriculum related assessment, dynamic assessment and a hypothesis testing framework for assessment over further work on variants of a normative approach.
The use of an IQ-achievement discrepancy definition of dyslexia has been criticised on several grounds. One of the alternatives advocated in the literature is a discrepancy definition in terms of listening and reading comprehension. For children learning EAL who have reached a sufficient stage in oral English proficiency the application of this criterion appears likely to help identify those whose reading difficulties are not mainly a factor of limited oral language proficiency.
No test can be "culture-free", but test developers can do a great deal to reduce unfair bias in test content. It is recommended that they adopt a good practice that covers this issue fully. Where tests and assessment tasks are developed locally, it is possible to employ school-based review strategies for checking on possible item bias.
Should children learning EAL be assessed in their first language or in English or in some combination of them both?
The first step in any assessment is to evaluate a child's knowledge and use of their first language and of English. All other steps in the process of assessment will need to take account of the implications of the language findings - an apparently obvious principle which studies in different areas of the country have shown to be often ignored in practice.
There has been only limited development work on the evaluation of children's command of minority languages where they have learning difficulties, and there has been controversy about the most effective way of delineating progress in developing
English as an additional language. There is no simple and universal answer to the question of whether the abilities of
children learning EAL should be assessed in L1 or L2. It cannot be assumed that, because children have a first language other than English, they will be advantaged if educational assessment is carried out in their L1. They may rarely use that language for academic purposes and have no vocabulary in it for the things studied at school. A decision as to the appropriate language of educational and psychological assessment for individual children is best taken on the basis of an initial evaluation of their exposure, use, proficiency and affiliation in relation to each of their languages.
What background information is needed when the learning difficulties of a child with EAL are assessed?
When children have a minority cultural background or a complex language history, accurate identification and assessment of learning difficulties will be impossible without detailed background information. There is broad agreement about what items
of information should be taken into account and reported when the learning difficulties of a child with EAL are assessed. Yet researchers have identified key information as missing in a significant proportion of the case papers examined in
studies in different parts of the country. An important part of the background information to be taken into account will be an
account of the child's educational history and current educational provision. There is a lack of systematic, theoretically informed developmental work on strategies for evaluating the school learning environments of children with EAL for the purposes of SEN assessment.
There is evidence that some children learning EAL may experience severe cultural discontinuities between literacy practices at home and in a community or religious class on the one hand and at their school on the other. When severe reading
difficulties are observed in L2 and the child attends a community or religious class, an insight into their progress with literacy in that setting may prove illuminating.
Cline, T. and Shamsi, T. DfES, 2000, RR 184
