DISTINGUISHING FACTS AND ARTIFACTS IN GROUP-BASED MODELING*

Group-based methodology (SPGM) has been presented as suitable to test for the existence of subpopulations not directly observable. Several criminological studies have used this methodology, and it is fair to say that typological theorizing has been spurred by its development. In particular, much of the empirical support for Moffitt's taxonomy (1993, 2006) is from studies using SPGM. In a small simulation experiment, I investigate whether SPGM is suitable for such tests, and I examine the extent to which similar trajectories might equally well result from mechanisms suggested by general theories. I conclude that, as it is usually applied, SPGM cannot provide evidence either for or against a taxonomy and that the usual findings can be explained by competing theories. I argue that this result is not only because of the methodology characteristics but also because of the modeling strategy applied.