Limitations in the implicit theories evaluation instrument: a critical approach to constructivism

Abstract

Studies of the implicit theories assume that the evaluation model is valid, when there are no statistical analyzes to confirm it. The aim of this study is to evaluate the validity and reliability of the Implicit Theories Dilemmas Questionnaire (CDTI) initially proposed by Martín et al. (2004) with 18 items, then reduced to 12 items (Vilanova et al., 2007) and finally to 10 items (Basilisa et al., 2014). The 10-item instrument was applied to 474 students from the UNAE National University of Education in Azogues, Ecuador. To measure the validity of factors proposed by the theory, the methods of Maximum Likelihood, Unweighted Least Squares and Bayesian Estimation were used. As a result, it was noted that the questionnaire is neither robust nor reliable, which calls into question the model of implicit theories. The item analysis shows that UNAE teaches future teachers to adopt constructivist conceptions, mainly in the cognitive and curricular aspects, not so much in those of evaluation. The weakness of the instrument is discussed and an alternative much more attached to the behavior of the variables in the formulation of the questionnaire and the search for explanatory factors to the problem of education is proposed, as well as the review of paradoxical positions within Ecuadorian education, beyond the constructivist ideology.