Methodological approaches to the study of L2 classrooms are extremely varied, reflecting both a great diversity of research questions and purposes, and a range of theoretical perspectives on the conduct of research. In general, these approaches have followed methods adopted by researchers in native language schooling or other sociological and sociolinguistic studies of communicative interaction. But in L2 classrooms communication between teachers and L2 learners becomes a particularly important issue, so L2 methodology has had to evolve new concepts, instruments, and procedures to adequately describe and analyze interaction. The purpose of this chapter is to describe these methods with regard to their capacity to extract and validate generalizations about the social and linguistic processes occurring in L2 classrooms. The specific content of various studies, such as the study of teachers' speech modifications, language choice, feedback on error, student interaction, and other phenomena, will be dealt with more fully in subsequent chapters.

Four traditions in research

Because the study of L2 classrooms has arisen through the influence of researchers from different disciplines (education, sociology, psychology, linguistics, applied linguistics), research developments in each of these areas have separately contributed to procedures for investigation. Although it should be recognized that few researchers adopt only the procedures and analyses of one tradition, at least four traditions are distinguishable: psychometric, interaction analysis, discourse analysis, and ethnographic.

The first, the psychometric tradition, was applied in early evaluations of L2 instruction (e.g., Scherer and Wertheimer 1964), which followed as much as possible standard educational psychometric procedures, with comparison treatment groups and measurement of outcomes on proficiency tests.