The analysis of classroom talk

This article describes methods for analysing classroom talk, comparing their strengths and weaknesses. Both quantitative and qualitative methods are described and assessed for their strengths and weaknesses, with a discussion of the mixed use of such methods. It is acknowledged that particular methods are often embedded in particular methodologies, which are based on specific theories of social action, research paradigms, and disciplines; and so a comparison is made of two contemporary methodologies, linguistic ethnographyand sociocultural research. The article concludes with some comments on the current state of development of this field of research and on ways that it might usefully progress.

In this paper, I will describe some methods for analysing the talk and interaction of teachers and students, discussing their various affordances and limitations. One obvious way that methods vary is whether they provide qualitative or quantitative results, and so I will use that distinction as a major organizing principle. In educational research generally, the selection of a method seems often to reflect researchers’ attachment to different epistemological theories, disciplinary traditions, and research paradigms. So in the study of classroom talk, different paradigms of enquiry, or methodologies, can be distinguished; and these embody certain tenets or principles about the nature of educational talk and how it can best be studied. For example, two influential approaches to the study of classroom talk in the UK are linguistic ethnography and sociocultural research. They have arisen from different disciplinary traditions, and those traditions not only influence researchers’ methodological choices, but also the framing of their research questions and their conceptions of how educational research should relate to practice. Before describing particular methods, then, I will briefly describe these two rather different approaches.

218Linguistic ethnography has a heritage of social anthropology and descriptive linguistics. Studies are typically observational, non-interventional, and qualitative. The essence of this approach is well explained in a paper by some leading exponents (Rampton et al., 2004); Creese (2008) and Tusting and Maybin (2007) are also very informative. Illustrative examples of this kind of research are Lefstein (2008), Maybin (2006), and Rampton (2007). Researchers normally employ the ethnographic and sociolinguistic methods I describe later in this article, which involve the close and detailed examination of classroom talk in its social and cultural context. They are unlikely to use any form of experimental method, or to use statistical analysis. Indeed, they are likely to feel quantitative, pre/post analyses of changes in talk or of learning gains are antithetical to a proper exploration and understanding of classroom communication. They have addressed research questions such as the following:

How does classroom discourse enable, or inhibit, the expression of identities?

How are the languages/language varieties of different cultures recognized and used in schools?

Is current educational policy sensitive to the linguistic and cultural reality of school life?

Linguistic ethnographers commonly emphasize that language and social life are mutually shaping; that talk is always referential, interpersonal, emotive, and evaluative; that socialization is a never-ending process, mediated through talk and interaction; that language genres are important features of educational culture; and that children use talk, in classrooms as much as anywhere else, to negotiate and explore their identities. They also often argue that social situations are unique, and so generalizations of the kind commonly made by quantitative researchers are of dubious validity.

Sociocultural researchers, on the other hand, are more likely to affiliate to research traditions in social and developmental psychology and pedagogical studies, with strong attachments to the work of Vygotsky (1978; see also Daniels, 2001; Wertsch, 1985). A rationale for using this theoretical frame for studying classroom talk is provided by Alexander (2000) and Mercer and Littleton (2007). Illustrative examples of research are Black (2007), Howe, McWilliam, and Cross (2005), Skidmore (2006), Smith, Hardman, Wall, and Mroz (2004), and the collection edited by Mercer and Hodgkinson (2008). Nuthall’s distinctive work also has much in common, theoretically and methodologically, with this line of enquiry (Nuthall, 1999, 2007; see also Collins & O’Toole, 2006). Sociocultural studies may be observational, interventional, and/or quasi-experimental. Researchers quite often combine qualitative and quantitative methods (as described under ‘Mixed methods’ later in the article). They have addressed questions such as:

219How does dialogue promote learning and the development of under standing?

What types of talk are associated with the best learning outcomes?

Does collaborative activity help children to learn, or assist their conceptual development?

Sociocultural researchers commonly emphasize that language is a cultural and psychological tool which (in Vygotskian terms) links the intermental and intramental – so, for example, classroom dialogue could have an important influence on the development of children’s reasoning. They also typically emphasize that knowledge and understanding are jointly created, that talk allows reciprocity and mutuality to be developed through the continuing negotiation of meaning, and that education depends upon the creation and maintenance of intersubjectivity or ‘common knowledge’. An implication often drawn is that teachers need to guide and scaffold learning, balancing the control of dialogue between teachers and students (Myhill, Jones, & Hopper, 2005). I would suggest that it is because of the directly ‘applied’ orientation of many sociocultural researchers that they are positively inclined towards the use of pre/post interventional designs, seeking to measure differential effects of talk on problem solving, learning, and conceptual change.

Having focused on the differences between these two methodologies, it is worth noting that they also have some shared principles. Researchers in both groups are thus likely to agree that classroom education cannot be understood without due attention to the nature and functions of talk (and that means there must be a qualitative element to the analysis); that cultural and local norms shape the processes of teaching and learning; and that in the classroom, meanings are continually renegotiated through talk and interaction over variable periods of time. The implication is that one-off, ‘snapshot’ studies of classroom talk are unlikely to yield as valid results as those which involve continuous and repeated observations, such as over a series of lessons. Both groups of researchers are likely to be critical of forms of classroom research which do not appear to recognize the importance of these principles – for example, through the use of simplistic coding schemes which treat all similar-looking utterances as repeated instances of the same event (at least if used without the correctional influence of a proper qualitative analysis, as discussed in more detail in later sections). They would probably also agree that the careful observation of classroom life commonly reveals much of interest that will not normally have been apparent to the teachers involved.

To some extent, then, it is difficult to completely separate methods from methodologies. Some researchers, indeed, would probably argue that we should not. Nevertheless, in the rest of this article, I will attempt to describe methods which are in common use simply in terms of their procedures and functionalities.