ERIC Number: ED218776
Record Type: RIE
Publication Date: 1982-Mar
Reference Count: 0
Achievement Prediction with School Level Equations: A Non-Technical Example Using the Public and Private Schools Data.
Crain, Robert L.; Ferrer, Robert L.
To assess the effects of school traits on achievement, researchers reanalyzed the data used in "Public and Private Schools" (Coleman et al.) at the school level rather than the individual level. The data for the Coleman report and the present reanalysis are drawn from the "High School and Beyond" study, a 1980 national survey of 30 sophomores in each of 1,002 high schools. The researchers find that using regression equations on aggregate school-level data instead of individual-level data reduces the error in individual-level equations (caused by error in variable measurement) and controls for the contextual effects of student-body socioeconomic status. The reanalysis indicates that the apparent superiority of private schools in academic achievement is much smaller when computed at the school level rather than at the individual level. (Author/RW)
Descriptors: Academic Achievement, Data Analysis, High School Students, High Schools, Individual Characteristics, Institutional Characteristics, National Surveys, Private Schools, Public Schools, Regression (Statistics), Research Methodology, School Effectiveness, Socioeconomic Status, Student Characteristics, Tables (Data)
Publication Type: Reports - Research
Education Level: N/A
Sponsor: National Inst. of Education (ED), Washington, DC.
Authoring Institution: Johns Hopkins Univ., Baltimore, MD. Center for Social Organization of Schools.
Identifiers: High School and Beyond (NCES); Public and Private Schools (Coleman et al)