The paper reports the results of a reanalysis of data from a major longitudinal data base (the Inner London Education Authority's (ILEA) Junior School Project). Multilevel models are used to examine the extent of differential school effectiveness for reading and mathematics at entry (year 3) and at year 5, and the impact of pupil background characteristics. Differential school effectiveness is investigated for prior attainment in the two areas and for selected background characteristics. The results reveal the existence of differential effectiveness (differential school slopes) for both measures of prior attainment. No evidence of differential effectiveness for pupils of different groups (sex, social class, ethnicity) is found. The presentation of schools' test results is also examined. Raw test averages are compared with school‐level residuals from multilevel analyses. Marked differences in schools' rank positions are found and the implications for the publication of ‘league tables’ of schools' examination and National Curriculum assessment results are discussed.

No abstract

Multilevel analyses of students' GCSE examination results (taken at age 16) are used to investigate both primary and secondary school effects on students' total performance scores and the issue of continuity of schools' effects over time. Follow-up data for the Inner London Education Authority's Junior School Project sample, including attainment at secondary transfer (age 11) and at GCSE and details concerning students' background characteristics are analysed.Significant school effects at both primary and secondary level and small but significant continuing effects of primary schools on later GCSE attainment are identified. The theoretical and practical significance of the results are discussed.

No abstract

No abstract