Three methods for studying developmental change: a case of reading skills and self-concept

Abstract

Aims: First, to introduce and compare three statistical methods for investigating development as a cumulative process: a simplex model, latent growth curve analysis, and clustering by cases. Second, to investigate the developmental dynamics of reading skills, and self-concept of reading ability, across the first year of primary school.

Sample: One hundred and five (61 boys, 44 girls) 6- to 7-year-old children from four first-grade classes in two primary schools participated in the study.

Method: Children were studied three times during their first school year using an identical set of measurements: a Reading Skills Test and the Self-Concept of Ability scale.

Results: A uni-construct 'Matthew' effect was found for the development of self-concept, but not for the reading skills. However, the results showed that there was a multi-construct cumulative cycle between children's reading skills and their self-concept.

Conclusions: Simultaneous use of variable- and person-oriented methods in developmental research seems to be a valuable approach, which not only provides a proper way to investigate the cumulative developmental cycles but also an option to examine how large a proportion of the sample follows the positive and negative pattern found in variable-oriented analyses.