Validation of the CreeA-Mat in Ecuador and Its Associations with Sex, Ethnicity, and Academic Performance

This dataset corresponds to the adaptation and validation process of the Self-Efficacy Beliefs Questionnaire for Mathematical Problem Solving (CreeA-Mat) in the Ecuadorian context. The study was conducted with two independent samples of high school students from public institutions in the province of Imbabura, Ecuador (Sample 1: n = 1,333; Sample 2: n = 2,183), used for exploratory and confirmatory factor analyses, respectively. The database includes item-level responses to the CreeA-Mat, relevant sociodemographic variables (sex and ethnic self-identification), and measures of perceived academic performance. Psychometric analyses were carried out within the frameworks of Classical Test Theory and Item Response Theory. Results confirmed the original unifactorial structure of the instrument, with excellent fit indices (CFI = .977; TLI = .974; RMSEA = .041) and high internal consistency (α = .942; ω = .943). Strict measurement invariance was established across sex and ethnic self-identification, supporting metric equivalence across sociodemographic groups. Item Response Theory analyses indicated adequate difficulty distribution and high item discrimination parameters (a > 1.50). A moderate correlation between mathematical self-efficacy and perceived academic performance (rₛ = .467) provided evidence of convergent validity. No statistically significant differences were found according to sex, whereas differences by ethnic self-identification were statistically significant but with a very small effect size. This repository provides empirical evidence of the cross-cultural validity of the CreeA-Mat in Ecuador and serves as a resource for secondary analyses, international comparative research, and the development of educational interventions in Latin American contexts.