Path analysis: a strategy for investigating multivariate causal relationships in communication disorders.

This tutorial paper explains, illustrates, and discusses path analysis, a powerful strategy for examining the plausibility and degree of causal relationships which are postulated to exist among a set of variables. In addition to its strength as a method of theory testing, path analysis is highly compatible with the kind of retrospective research which is so often used to study speech and language pathologies. One of its major advantages is that it allows explicit recognition of multiple, interacting causes of communication disorders and allows the researcher to evaluate the logical consequences of assumptions made about the specific nature of those causal relationships. A summary of the meaning of causal relationships is followed by a presentation of the principles underlying path analysis, as well as an outline of its basic procedures. Also presented is an illustrative example of the application of path analysis to the evaluation of some competing causal theories of a communication disorder. Finally, the general advantages of path analysis and its applicability to the investigation of causal relationships in speech and language pathologies are summarized and discussed.