skip to main content skip to footer

Analyzing School Effects: ANCOVA with a Fallible Covariate ANCOVA

Author(s):
Werts, Charles E.; Linn, Robert L.
Publication Year:
1969
Report Number:
RB-69-59
Source:
ETS Research Bulletin
Document Type:
Report
Page Count:
16
Subject/Key Words:
Analysis of Covariance (ANCOVA), Educational Benefits, Program Effectiveness, Research Methodology, Statistical Analysis, Student School Relationship

Abstract

The analysis of covariance method has been employed as a means of controlling for input when studying the differential impact of schools on students. The interpretation of such results is confounded by unreliability in the input measures which may result in spurious findings because in effect input has not been completely controlled. Some of the logical and statistical problems in dealing with covariate unreliability are discussed.

Read More