Tutor ratings do not entirely tap one underlying trait; one- and two-factor models fit unsatisfactorily¶

Claim · All claims
Evidence · 1 study · causal r2 · q2
Currency · newest study 1995 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Confirmatory factor analysis showed neither a one-factor nor a two-factor (process/content) hypothesis fitted the tutor-rating data entirely satisfactorily. → Schmidt 1995

Evidence¶

Schmidt 1995¶

Schmidt, H., van der Arend, A., Kokx, I. & Boon, L. (1995). Peer versus staff tutoring in problem-based learning. Instructional Science 22: 279-285. https://doi.org/10.1007/bf00891781

q2 · i? · causal · r2

Confirmatory factor analysis of the four tutor-rating items. The one-factor model gave Chi-square 91.018, df = 2, CFI 0.94; the two-factor process/content solution gave Chi-square 42.996, df = 1, CFI 0.97 — neither fitted entirely satisfactorily, suggesting ratings do not entirely tap the same underlying trait.

"The one-factor hypothesis produced a Chi-square of 91.018, df = 2, p < :0.001. Bentler-Bonett's Comparative Fit Index for this case equaled 0.94."

Discussion¶

Related Claims¶

• A one-factor unidimensional model is the most reasonable fit for MAP computerized adaptive Reading and Mathematics tests across grades, terms, and ten states, though Michigan Mathematics misses the RMSEA criterion — related
• The six-process structure of inductive reasoning fits fifth graders' figurative-test data better than a unidimensional model — related
• In kindergarteners, trait beliefs and goal orientations form two separate but related factors rather than a coherent growth/fixed mindset — related
• A confirmatory factor analysis supported a two-factor Gf–Gc representation of the cognitive measures, with a significant .33 correlation between the factors — related
• A two-factor (reactive–proactive) model of the Turkish RPQ fits adolescent data better than a one-factor general aggression model — related