ERIC Number: ED193318
Record Type: RIE
Publication Date: 1980
Reference Count: 0
Normalization Regression Estimation With Application to a Nonorthogonal, Nonrecursive Model of School Learning.
Bulcock, J. W.; And Others
Advantages of normalization regression estimation over ridge regression estimation are demonstrated by reference to Bloom's model of school learning. Theoretical concern centered on the structure of scholastic achievement at grade 10 in Canadian high schools. Data on 886 students were randomly sampled from the Carnegie Human Resources Data Bank. The 41 variable model formulated by Bloom demonstrated transmission of school performance and school-related attitudes from grades 9-10, as well as selected student background factors. Because school learning models call for reciprocal effects relationships at both entry behavior and learning outcomes stages, two stage least squares (2SLS) estimates were performed. Although incorporation of feedback linkages in structural equation models aggravated the multicollinearity problem to the extent that the 2SLS procedure was inadequate, normalization regression was able to cope satisfactorily with each of six harmful effets. The substantive results did not particularly support current theories of school learning, especially those concerned with the effects of socioeconomic status on schooling success, and the impact of sex differences on schooling outcomes. (Author/RL)
Publication Type: Reports - Research
Education Level: N/A
Sponsor: Social Sciences and Humanities Research Council of Canada, Ottawa (Ontario).; Natural Sciences and Engineering Research Council, Ottawa (Ontario).
Authoring Institution: N/A
Identifiers: Bloom (Benjamin S); Canada; Normalization Regression Estimation; Ridge Regression Analysis
Note: Contains occasional small print.