No abstract

Controversy over the classification of depressive phenomena has been a prominent feature of the psychiatric scene for about forty years and the issues involved have been reviewed on a number of occasions (Partridge 1949; Grinker et al. 1961; Kiloh 1965; Rosenthal and Klerman 1966; Mendels and Cochrane 1968). Briefly it may be stated that a number of attempts have been made to support or refute the validity of classifying depressions into those which are “endogenous” and those which are “reactive” or “neurotic”; using a variety of statistical methods which have included factor analysis, principal component analysis and multiple regression analysis.

In this chapter we shall be looking at the issues which logically (and generally in practice) precede data collection itself -what cases to select, and how the study should be designed. The major concern is with validity, by which we mean the design of research to provide credible conclusions: whether the evidence which the research offers can bear the weight of the interpretation that is put on it.Every report of research embodies an argument: 'on the basis of this evidence I argue that these conclusions are true'. Within this, the evidence presented is, again, the product of a series of arguments. The authors collect certain information in certain ways from or about certain people or settings, and the research report argues that this information may be interpreted in certain ways to lead to true conclusions about a certain population. What has to be established in order that the report's conclusions can be believed is that the arguments embodied in the report are valid ones: that the data do measure or characterize what the authors claim, and that the interpretations do follow from them. The structure of a piece of research determines the conclusions that can be drawn from it (and, more importantly, the conclusions that should not be drawn from it).We shall argue, throughout this book, that the same questions have to be answered by research studies in widely different styles and arising from widely differing epistemological bases. Some research takes an essentially positivistic approach to the nature of knowledge about the social world: it takes the nature of the world as relatively unproblematic -the main problems being how to measure it adequately -and it emphasizes the neutrality and separateness of the researcher from that which is

No abstract

No abstract