Journal of East China Normal University(Educationa ›› 2018, Vol. 36 ›› Issue (1): 137-148+166.doi: 10.16382/j.cnki.1000-5560.2018.01.017

Previous Articles     Next Articles

Non-Continuity and Heterogeneity: Application of Piecewise Growth Mixture Model in Language Development Study

LIU Yuan1, LIU Hongyun2   

  1. 1. Faculty of Psychology & Key Laboratory of Cognition and Personality, Southwest University, Chongqing 400715, China;
    2. Faculty of Psychology, Beijing Normal University, Beijing 100875, China
  • Online:2018-02-20 Published:2018-01-12

Abstract: In recent researches, the piecewise growth mixture model (PGMM) has been used in longitudinal studies to detect the non-continued growing trend and heterogeneous population simultaneously. The present study used the data from Early Childhood Longitudinal Study-Kindergarten cohort (ECLS-K) as an example to illustrate the use of PGMM. An ideal model of PGMN is a two-piece growing model, with the turning point at Grade One, linear trajectory in the first period and quadratic trajectory in the second. The result showed that there should be a crucial turning point in the development of reading ability, with a rapid growing rate from kindergarten to Grade One and then a sharp-decline rate after entering the primary school. Furthermore, a three-class model was selected where the heterogeneous sample-based population was essential in describing the growing pattern. Finally, the result indicated the teachers' assessment of children's behavior was more likely to predict the latent class than that of the parents' with the control of the background effects.

Key words: Piecewise Growth Mixture Modeling (PGMM), non-continuity, latent heterogeneity, model fit