The applicability of a mathematical theorem designed to trace causality of a three-variable path that consists of an initial cause variable, an intermediate variable, and a final-effect variable with control over other system variables is evaluated. The formula was used with a horizontal rather than a normal distribution, as had been done in an earlier study which successfully demonstrated the theorem. Horizontally-distributed random numbers (200x5) were generated and substituted into five interdependent equations to produce five scores for each of 200 pseudo people. Methods and techniques, data sources, and results are presented. Results indicate that the model is reliable, potentially useful to the practitioner as a multivariate tool in educational research, and a possible instrument for the investigation of causality. (AE)