Effectiveness of multilevel modeling at accounting for school value-added effects on student English reading attainment

Lijuan Joanna LI

Research output: Contribution to conferencePapers

Abstract

This study is an extension of a previous methodologically exploratory study which established several multilevel value-added models to identify how different student and school factors influenced student English reading attainment in the nested school structure. The present study moved a step further to compare effectiveness of the models and determined the one most efficient at estimating school effects on adding value to their students’ English reading ability from Secondary 1 to Secondary 6. Effectiveness of the most parsimonious model at presenting and interpreting the substantive findings was further tested using the same set of data. The data was collected from 3,993 students of 66 Hong Kong secondary schools in 2006. Students’ English ability was tested at Form 1 intake and tested again at Form 6. The background information, such as student and school demographic and contextual factual data, was also collected. Partialing out influence from student and school level background factors, the relative progress made by students from different schools was compared. Specific findings on value-added effects of the schools are plotted thus providing distinctive graphic information. The findings will provide some insightful indications to the authorities and educational practitioners for student improvement and school effectiveness.
Original languageEnglish
Publication statusPublished - Feb 2014

Citation

Li, L. (2014, February). Effectiveness of multilevel modeling at accounting for school value-added effects on student English reading attainment. Paper presented at the Comparative Education Society of Hong Kong (CESHK) Annual Conference 2014: Policy and educational development in a global context, HKU School of Professional and Continuing Education, China.

Fingerprint Dive into the research topics of 'Effectiveness of multilevel modeling at accounting for school value-added effects on student English reading attainment'. Together they form a unique fingerprint.