No abstract

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. GROWTH IN PUBLIC AND CATHOLIC SCHOOLS 75in method indicated some robustness of the inferences. However, for testing the inferences with longitudinal data, these variations can become problematic. Definitions and procedures necessitate some arbitrary decisions, ways of approaching a problem differ, and styles of reporting results and pursuing a sequence of findings differ. To resolve each of these differences by compromise would be to lose many of the virtues of the two approaches. Therefore, to compare the longitudinal and crosssectional results of both analyses, it appears most sensible to carry forward both techniques.The next section is a longitudinal extension of the Greeley (1982) analysis, and the section following that is a longitudinal extension of the CHK (1982b) analysis. There are two modifications: (1) Although the Greeley crosssectional analysis used unweighted data, the longitudinal extension uses data weighted to represent the U.S. student population; and (2) although the CHK cross-sectional analysis compared both Catholic schools and other private schools tQ public schools, the longitudinal analysis is limited to Catholic schools because of the small size and heterogeneity of the sample of non-Catholic private schools.The final section of the article examines the consistencies and inconsistencies between the two analytical sections and draws together the results from both approaches. It also examines studies by Alexander and Pallas (1985) and Willms (1985) and discusses the points of convergence and divergence between their results and those presented here. LONGITUDINAL EXTENSION OF THE GREELEY ANALYSISIn this section, we extend the analysis begun in Greeley (1982). The methodological skepticism of that study is retained: We assume the presence of a school effect only after we have attributed as many school-related differences as possible to differences in individual and family input and only when we can link the residual difference to actual school programs and policies. We therefore attribute to student-input variables, as opposed to schooleffect variables, such influences on academic performance as the students' college plans, which might themselves be the result of their school experience.The objections to our earlier reports can be summarized as follows:1. Any apparent effects are merely the result of a higher learning curve for Catholicschool students, due to selection. The effect of Catholic schools is confinedto the most able students and to those from homes where there is strong motivation and support for academic achievement. 3. Catholic schools seem to have an effect because they eliminate their disciplinary problems by expelling them from the school. T...

No abstract

No abstract