ERIC Number: ED167598
Record Type: RIE
Publication Date: 1978-Mar
Reference Count: 0
The Use of Longitudinal Analysis to Identify More and Less Effective Schools.
Goldberg, Zandra S.; Spartz, James L.
Three longitudinal studies of student achievement test data were conducted to examine the extent to which the variation in student achievement can be explained by differences in student background (including socioeconomic and prior achievement data) and school resources. This model assumes that achievement test performance may be predicted by these factors plus the effectiveness of the school program. Data were analyzed by school, and included means of aptitude and achievement test scores, student background data, and school resource information. Students were tested in first, fourth, and eighth grades. One study compared student achievement from fourth to eighth grade, while two studies compared achievement from first to fourth grade. It was concluded that longitudinal analysis of these data can identify schools whose achievement appears to be above or below expectancy, but the results of one study are not reliable so it is essential that a replication study be conducted. However, where the data for the study included data from both elementary schools and middle schools, this method was not successful in identifying effective schools. (Author/CTM)
Descriptors: Academic Achievement, Educational Resources, Elementary Education, Expenditures, Family Income, Family Influence, Longitudinal Studies, Multiple Regression Analysis, Predictor Variables, Program Effectiveness, School Effectiveness, School Policy, School Support, Socioeconomic Background, Success
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Authoring Institution: N/A
Identifiers: Delaware; School Evaluation
Note: Paper presented at the Annual Meeting of the American Educational Research Association (62nd, Toronto, Ontario, Canada, March 27-31, 1978)