ABSTRACT
This chapter examines what constitutes survey research as a particular type of research method and what are the problems peculiar to it. Survey researchers have probably all got some vague negative justification for survey research in their heads which amounts to a knowledge that the other styles of research that are open to sociologists are in practice inadequate. In other words, in survey research the process of testing causal hypotheses, central to any theory-building endeavour, is a very indirect process of drawing inferences from already existing variance in populations by a rigorous process of comparison. If correlational analysis is used to test theories which link variables in a causal model, then survey research has a contribution to make to the development of scientific theories. The chapter discusses some of the difficulties which arise from converting individual information to variables and thereafter analysing these either in the whole dataset or in specific subgroups.
