Key points are not available for this paper at this time.
The Matthew effect hypothesis in reading predicts that the gap between good and poor readers increases with time. Although, intuitively appealing, the Matthew effect has hardly been empirically studied in longitudinal studies of reading. Two competing longitudinal models were used to represent the Matthew effect hypothesis: the Latent Growth Curve model and the Simplex model with structured means. It is argued that on the basis of theoretical and empirical arguments the Simplex model should be preferred to represent and analyze the Matthew effect hypothesis. However, the results of the Simplex models imply that conceptual refinement and clarification of Matthew effects in reading are needed.
Bast et al. (1997) studied this question.