A Longitudinal Analysis Using the Lisrel-Model with Structured Means

The Lisrel-model with structured means was used to study similarities and differences between two groups of students on a longitudinal set of data. Inferences were made about the differences between the two groups in the development of mathematical ability as measured by two tests on three successive occasions, by testing the hypothesis that the slopes in the regression of true final scores on true initial scores are equal across the two groups, and by estimating the difference in the intercept alpha in the structural equation. The results indicated that the two groups started with equivalent mathematics ability. The Immersion group had a significantly higher level (2.52 points) of growth in mathematics ability than the Regular group at Grade 5; in the subsequent year, however, the Regular group exceeded the Immersion group by 2.15 points in its average growth. In comparison with the Regular group, the Immersion group grew more rapidly from grade 4 to 5 and more slowly from grade 5 to 6.