Multiple Determinants and Effect Size: A More General Method of Discourse

Ahadi and Diener (1989) argued that the prevalent concern about low trait-behavior correlations may reflect the faulty assumption of a one-to-one relation between traits and behaviors. Using a simulation methodology, they demonstrated that low to moderate trait-behavior and behavior-behavior correlations are to be expected even when 2 behaviors are each completely determined by 3 traits with 1 trait in common. It is demonstrated that the results estimated by Ahadi and Diener can be derived exactly by using the mathematics of linear combinations. The advantage of the mathematical approach is the precise and easy calculation of results that follows from presumed causal models of any complexity. This approach provides a general method for exploring the expected empirical consequences of different causal models.