Elements of Research Design

For the purpose of explanation and prediction, scientific research finds it helpful to think in terms of cause and effect relationship. There is a procedure for identification of such relationship. In normal terms, a cause is something that makes some other thing to happen or change. For example, a stone thrown at the window pane will cause the pane of glass to shatter. In this situation, throwing of stone is the cause while the shattered glass is the effect. Such an implicit understanding of causality is found in everyday life. But the meaning of causality in scientific research rests on the assumption that one can observe a constant or stable association between the occurrence of events and from such an association a causative connection can be inferred.
Causal relationships are generally regarded as the heart of scientific understanding and thus it is important to understand the bases of causal interferences. Social scientists generally require at least three kinds of evidence to establish causality. These requisites are association, direction of influence and nonspuriousness.
Association: For one variable to be a cause of another, the variables must be statistically associated. If the pattern of changes in one variable is not related to changes in another, then the former is not the cause of the latter. Associations, of course, are almost never perfect. Perfect associations may be expected only under pure theoretical inference but not in the real world of observations. This is so because in real social situations it is not possible to eliminate completely the influence of extraneous factors. In actual practice, thus, causal relationships are determined in terms of strong or weak associations. For instance statistical measures only indicate the association of two variables. Cause and effect influence/deducation is based on the research design's formulation and theoretical premises. In social sciences, causal relationships are sometimes implied from