In this study, three multilevel model approaches were derived to expand the applicability of multilevel models to longitudinal research on student achievement. These three procedures were the growth curve approach, the stability approach, and the school-effect approach. Briefly, the growth-curve approach handles true longitudinal data; its main concern is the growth of student achievement. Both the stability approach and the school-effect approach handle pseudo-longitudinal data. The stability approach addresses the stability of school effects over time. The school-effect approach is concerned with the variability of the cross-sectional school effect over time. / An illustrative example was given with computational detail for each approach using data from Leon County Public Schools, Florida. VARCL3 software was used in the estimation procedures. / The application issues included assessment of the following: goodness of fit of the model, measurement of change, correlates of change, estimation of mean growth curves of student achievement, estimation of school effects, comparison of school effects across schools, stability of school effects, and examination of variability of school effects across schools and time. Most attention was given to studying school effects longitudinally. / It was shown that the three approaches can be applied for two major purposes of school effectiveness studies: better understanding of the determinants of student outcome, and comparison of school effects across schools. The first purpose could be addressed by the fixed effects and the second by the random effects. / The three approaches provide their own unique representations of school-effect estimation. If the school effect on the individual student achievement growth is the main concern, the growth-curve approach can be used. If the change over time of the school effect is the main concern, use of the stability approach over the school-effect approach is recommended. / Source: Dissertation Abstracts International, Volume: 51-02, Section: A, page: 0656. / Major Professor: Richard L. Tate. / Thesis (Ph.D.)--The Florida State University, 1990.
| Identifer | oai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_78172 |
| Contributors | Im, Sea-Hyuck., Florida State University |
| Source Sets | Florida State University |
| Language | English |
| Detected Language | English |
| Type | Text |
| Format | 291 p. |
| Rights | On campus use only. |
| Relation | Dissertation Abstracts International |
Page generated in 0.0086 seconds