A Meta-Analytic Path Analysis of the Internal/External Frame of Reference Model of Academic Achievement and Academic Self-Concept

• Published in: Review of educational research Vol. 79; no. 3; pp. 1129 - 1167
• Main Authors: Möller, Jens, Pohlmann, Britta, Köller, Olaf, Marsh, Herb W.
• Format: Journal Article
• Language: English
• Published: Los Angeles, CA SAGE Publications 01.09.2009; American Educational Research Association
• Subjects: Academic Achievement; Academic self concept; Age Differences; Child Development; Correlation; Educational Policy; Educational psychology; Educational Research; Foreign Countries; Gender Differences; Job satisfaction; Mathematics; Mathematics Achievement; Mathematics Skills; Measures (Individuals); Meta Analysis; Modeling; Observational frames of reference; Outcomes of Education; Path Analysis; Psychology; Self Concept; Self Efficacy; Self evaluation; Self Evaluation (Individuals); Standardized Tests; Systematic review; Test Results; Verbal Ability; Young Children; Germany; self-concept; comparison processes; ability beliefs; I/E model
• ISSN: 0034-6543; 1935-1046
• DOI: 10.3102/0034654309337522

A meta-analysis of 69 data sets (N = 125,308) was carried out on studies that simultaneously evaluate the effects of math and verbal achievements on math and verbal self concepts. As predicted by the internal/external frame of reference (I/E) model, math and verbal achievements were highly correlated overall (.67), but the correlation between math and verbal self concepts (.10) was close to zero. Correlations between math and verbal achievement and correlations between achievements and self-concepts within the domains were more positive when grades instead of standardized test results were used as achievement indicators. A path analysis revealed support for the I/E model, with positive paths from achievement to the corresponding self concepts (.61 for math, .49 for verbal) and negative paths from achievement in one subject to self-concept in the other subject (-.21 from math achievement on verbal self-concept, -.27 from verbal achievement to math self-concept). Furthermore, results showed that the I/E model is valid for different age groups, gender groups, and countries. The I/E model did not fit the data when selfefficacy measures were used instead of self-concept measures. These results demonstrate the broad scope of the I/E model as an adequate description of students' self-evaluation processes as they are influenced by internal and external frames of reference.