Probabilistic causal systems and the conditional probability approach to causal analysis

An alternative approach to the analysis of probabilistic causal systems, the conditional probability approach, is examined & shown to be applicable to various kinds of general causal systems. A distinction is made between pure-"or" & pure-"and" systems. The conditional probability approach offers a useful method for causal analysis in the social sciences. This method is compared with the path analysis & structural equation approaches; it is shown that the latter two yield incorrect results in cases corresponding to three mathematically described sets of relationships. The conditional probability approach is further compared to an approach developed by Ian Birnbaum (see SA 31:3/83N0496), a method for causal analysis of contingency tables that is shown to be inadequate. 17 References. Modified HA.