Implementing intelligence facets as predictors into the generalised I/E model

The internal/external (I/E) frame of reference model postulates that achievement influences the formation of academic self-concept (ASC) via social and dimensional comparison processes. Its recent extension as the generalized I/E model allows further domains than math and verbal achievement. However, intelligence facets as an indicator of achievement have been neglected within the GI/E model framework. Therefore, the purpose of this research is to explore the influences of three intelligence facets (verbal, numerical, figural) on domain-specific ASCs beyond grades and achievement scores. We drew on N=382 German students to analyze verbal, numerical, and figural intelligence, German and math achievement, self-reported grades and ASCs in four domains. We performed a structural equation model using Mplus with grades, achievement scores and intelligence facets specified as predictors and domain-specific ASCs as criteria. Positive paths were found from math, physics, German and English grade to their corresponding self-concepts. A positive path was found between physics grade and math ASC while negative paths were found from English grade to math and physics ASC and from math grade to German, physics and English ASC. The path coefficients from math achievement to math ASC as well as to physics ASC were positive; German achievement was negatively related to math ASC. Lastly, numerical intelligence was positively related to physics ASC, and verbal intelligence was positively related to English ASC. Thus, intelligence, specifically numerical and verbal intelligence, seem to be valid predictors of domain-specific ACSs beyond grades and achievement test scores.