Validation and adaptation of the epistemic beliefs survey for mathematics in Colombian secondary students

Abstract

This study adapted and validated the epistemic beliefs survey for mathematics for Colombian secondary students. A total of 817 public-school students aged 10-20 participated, and the sample was randomly divided for exploratory factor analysis (EFA) (n = 407) and confirmatory factor analysis (CFA) (n = 410). Content validity was examined through expert judgment, and analyses used procedures appropriate for ordinal data. Item screening and EFA supported a revised 27-item structure in which ability differentiated into personal ability and general ability, whereas structure of knowledge did not emerge as a coherent factor. CFA showed that the revised six-factor model fit the data better than the five-factor alternative (CFI = .966, TLI = .962, RMSEA = .070, SRMR = .072), with satisfactory reliability (ωt = .898). Measurement invariance held through the scalar level across educational levels. Findings suggest context-sensitive factorial configurations, supporting context-specific validation when transferring domain-specific instruments.