CHAPTER 1
New Directions in Special Educational Needs Research
Graham Vulliamy and Rosemary Webb
My attitude to children with learning difficulties is very different now; Iām not prepared to accept that they canāt do certain things and Iām much better at looking for potential strengths in them.
The above comment was made during an interview with a primary school teacher researcher whose pedagogy had changed markedly as a result of her research. It illustrates how teachers enquiring into their own classroom practice can lead to a greater understanding of childrenās needs and capabilities. The aim of this book is to explore the potential of teacher research to bring about informed changes in practice and policy in the area of special educational needs.
In this opening chapter, we will examine both why teacher research has been relatively neglected in special education and what the essence of its contribution might be. In outline, our argument, which is developed in more detail in subsequent sections of the chapter, is that special educational needs research in this country has tended to be dominated by those trained in the discipline of psychology. This discipline has characteristically used positivist research strategies, such as experiments and surveys, which attempt to produce law-like generalisations based upon the statistical analysis of large samples. Any shortened statement on the methodology of positivism is clearly problematic in that the term is used in very different ways, both in philosophy and in sociology (Halfpennyās book, 1982), for example, delineates twelve different usages). Moreover, it cannot do justice to the varied positions that positivists take on the crucial issues such as causation and prediction in the social sciences. This said, a positivist approach is generally taken to imply that āthe methodological procedures of natural science may be directly adaptedā (Giddens, 1974, p.3). More recently, however, two major problems with such approaches have been identified by prominent special educational needs researchers. Firstly, there has been concern that experimental settings and the study of large samples often do little to add to our understanding of the realities of teaching and learning in the natural settings of ordinary or special schools. Secondly, worries have been expressed concerning the lack of impact of research on practice. During the 1980s many special needs researchers have increasingly pointed to the connection between these two perceived limitations of traditional research approaches. This has led to suggested new directions: quantitative studies based upon āsingle-subjectā and āsmall-Nā designs; a move from laboratory experiments to studies of the processes of learning in classroom settings; the blending of qualitative and quantitative research; and programmes for teacher-researcher collaboration.
Also, during the 1980s, the established tradition of special needs research increasingly came under attack from sociologists. They argue that much conventional special needs research tends to be atheoretical and, by neglecting the social and political context of disability, tends to focus exclusively on the characteristics of individual pupils at the expense of the institutional context of the school and the labelling processes in the wider society.
Such recent developments in special needs research are still wedded to the traditional educational disciplines of psychology and sociology. In other areas of educational research, however, notably in curriculum evaluation and in the study of classroom practice, it has been argued that a more genuinely āeducationalā theory located within a teacher-research tradition has great potential for breaking down the theory ā practice split (Elliott, 1978; Bassey, 1983). Such teacher research generally adopts a qualitative research strategy, which tends to be eclectic in its choice of data collection techniques, usually using some combination of interviews, observation and the analysis of documents. We believe that teacher research based, as much of it is, on case studies of the processes of teaching and learning, is particularly suited to the area of special needs, where teachers are often concerned to understand pupils with unique learning difficulties. And yet, with very few exceptions (e.g. Ainscow, 1989), this has not been an approach which has been generally encouraged by the special needs community.
The broadening of traditional approaches to special needs research
A historical overview of special needs research over the last two decades indicates that many of the marked changes in emphasis during the 1980s were prompted by the implications of the Warnock Report, published in 1978, and the subsequent 1981 Education Act. Wedell (1985a) notes that āthis Act gave āofficialā, recognition to the concept of āspecial educational needā, and to the concern of special education with meeting childrenās needs rather than with categorizing themā (p.1). Thus the focus of research needed to shift from descriptive studies of childrenās conditions and disabilities to studies of the ways in which various educational needs might best be met. This, in turn, directed research attention towards ways of improving the learning experiences of children, and evaluating and disseminating improvements, whether in special or in ordinary schools.
Such a change of emphasis is readily apparent in a comparison of reviews of special needs research conducted prior to the 1981 Act (Cave and Madison, 1978; Wedell and Roberts, 1981) ā where the more traditional research, including investigations of children grouped by category of diagnostic classification, predominates ā with an agenda for future research drawn up by a 1982 special needs research symposium. The latter concluded that priority should be given to six areas:
the evaluation of intervention approaches;
the methodology for evaluating intervention;
within the area of descriptive research, emphasis should be placed on the study of functional impairment, rather than the characteristics of diagnostic categories of children with special needs;
the process of innovation;
the methods of disseminating research information;
the preparation of critical summaries of existing relevant research, particularly in areas which extend across disciplines.
(Wedell, 1985b, p.23)
Such a research agenda inevitably required a shift from the more traditional emphasis of psychology upon studies of specific children, either as individuals or groups, to a broader range of approaches. This range included both new substantive themes, such as implementation and change, and new methodological ones, such as qualitative evaluation styles, which for some time had characterised educational research on curriculum evaluation and innovation in general. This, in turn, brought the special needs research community more directly into contact with alternative research strategies arising from critiques of the methodological positivism underlying more traditional approaches. As early as 1968, the psychologists Bracht and Glass had raised some fundamental problems about such approaches when they developed the concept of āecological validityā in their discussion of the external validity of experiments. Ecological validity refers to the extent to which behaviour observed in one context can be generalised to another. Put simply, the problem with more traditional research methods in education, whether experiments or questionnaire surveys, is that they are unlikely to give an accurate portrayal of the realities of teaching and learning in a natural or conventional setting. The strictures of experimental design are such that only rarely does the experimental setting approximate to the normal conditions of schooling to which generalisations need to be made. Questionnaire surveys cannot penetrate the gap between āwords and deedsā (Deutscher, 1966) and, especially in the evaluation of innovations, are prone to the reproduction of the rhetoric contained in the aims and documentation of the innovation. It is considerations such as these which have led to compelling arguments that traditional positivist research strategies in education and the social sciences have been over-preoccupied with reliability (the consistency of a measuring instrument) at the expense of validity (Deutscher, 1973). Somewhat ironically, perhaps, Bracht and Glassās (1968) early discussion of āecological validityā was picked up much later by qualitative researchers in education who see the maximisation of ecological validity as one of the main rationales for their approach (see, for example, Atkinson, 1979; Hammersley, 1979; Evans, 1983).
The influence of such critiques of positivism, together with the new research agenda, can be clearly seen in a 1984 special needs research symposium on the methodology of evaluation studies in special education. Thus, for example, Corrie and Zaklukiewicz comment in their paper that much previous work:
has tended to be limited by the somewhat narrow conception of research as being above all, quantitative and statistical in character. Quantitative procedures, particularly survey techniques, have featured prominently throughout the entire history of research in this area ā¦ as have psychometric approaches to individual functioning and development.
(1985, p.123)
Following an overview of the major criticisms being made of previous research in special education (Schindele, 1985), the symposium was devoted to two approaches to evaluation which, in very different ways, marked new departures. The first of these, the use of single-subject and small-N research designs, whilst still operating within a positivist and quantitative research strategy, nevertheless attempts to overcome some of the previous weaknesses of more traditional evaluation strategies by focusing upon an in-depth evaluation of practice. Kiernan (1985) provides an overview of the defining features of single-subject research designs. They involve the investigation of the behaviour of a single individual (or a small group in small-N designs), where the intention is to test specific hypotheses concerning critical variables. However, where in more traditional experiments the use of control groups is intended to enable the replication of effects across individuals, any replication in single-subject designs is across time, context or other variables and the analysis concerns only the behaviour of the single individual.
An example of such a study is Barrera et al.ās (1980) investigation into the effectiveness of three different models of language training ā using signs, words and a combination of the two ā for a four-and-a-half-year-old āmuteā autistic boy. This involved giving the child 20 minutes of direct language training with each of the three models in a random order each day and using a trained observer, supplemented by video recordings to ensure reliability, to monitor the childās responses. The results showed that the total communication model was substantially superior to both the oral and sign-alone training models.
Evans (1985) suggests that single-subject experimental designs have a particular relevance for special needs research for a number of reasons. Firstly, the nature of some childrenās special needs may be particularly individualistic in character, with few other children displaying exactly the same range of needs. Secondly, the designs enable teachers to monitor and record the progress of an individual child over time against pre-set objectives. Thirdly, whilst sophisticated statistical techniques have been developed for the interpretation of causal effects in single-subject designs (McReynolds and Kearns, 1982; Halil, 1985), it is still possible for teachers to āeyeballā the numerical data and reach conclusions which can immediately be put into practice.
The second major evaluation approach considered at the 1984 research symposium was the use of qualitative research. There are several important distinguishing features of such research. It provides descriptions and accounts of the processes of social interaction in ānaturalā settings, usually based upon a combination of observation and interviewing of participants in order to understand their perspectives. Culture, meanings and processes are emphasised, rather than variables, outcomes and products. Instead of testing preconceived hypotheses, qualitative research aims to generate hypotheses and theories from the data that emerge, in an attempt to avoid the imposition of a previous, and possibly inappropriate, frame of reference on the subjects of the research. This implies a far greater degree of flexibility concerning research design, data collection and analysis, with aspects of each of these sometimes occurring simultaneously throughout the duration of a research project, than tends to be the case with quantitative research. (For more detailed summaries of the defining characteristics of qualitative research see, for example, Corrie and Zaklukiewicz, 1984, pp.125ā7; Burgess, 1985, pp.7ā10; Bryman, 1988, pp.61ā9; Vulliamy, 1990a, pp.7ā14.)
Hegarty (1985) argues that qualitative research is particularly suited to a number of topics in special education. These include investigations into pupilsā and teachersā perspectives and experiences of special education programmes; clarifying the implications of various policy options; evaluating innovations; and providing detailed accounts of various forms of special education provision. However, to date, there have been relatively few published qualitative research studies on the theme of special educational needs, despite the prominent impact of qualitative research on educational research more generally. Moreover, it is noticeable that, of the few such published studies that do exist, a high proportion have been initiated or conducted by researchers from outside the special needs research community, as, for example, with Jamieson et al.ās (1977) study of the integration of visually-impaired children into ordinary schools and Bell and Colbeckās (1989) action inquiry project into the integration of a number of special needs units into a large primary school.
Given the strong legacy of positivist research in the special needs community, but the recent recognition also of the need for ...
