2 ASPECTS OF THE SIMPLEX MODEL - GOODNESS-OF-FIT TO LINEAR GROWTH CURVE STRUCTURES AND THE ANALYSIS OF MEAN TRENDS

This article has two objectives. The first is to investigate in greate r detail the finding of Rogosa and Willett chat the quasi-Markov simpl ex model fits a linear growth curve covariance structure. It is found that under various circumstances the quasi-Markov simplex model is rej ected. Furthermore, the procedure is reversed by fitting the linear gr owth curve to quasi-Markov simplex covariance structure. It is found t hat the linear growth curve, like the quasi-Markov simplex, is not alw ays rejected even though the model is formally incorrect. The second o bjective of this article is to present a quasi-Markov simplex model wi th structured means. This model, like the linear growth curve model wi th structured means, is based on the assumption that the variation in means and individual differences are attributable to the same causal a gents. We argue that this assumption should be tested explicitly. An e xample is given.