Psychometric Structure of the Scale of Attitutedes Before Statistics (EAE-2020)

Abstract

Attitudes are a complex construct whose primary components comprise the affective, cognitive, and behavioral parts, which imply, among other things, an assessment for or against a specific social object, whose proper measurement allows predictions to be made. In comparison, the study of statistics involves a predisposition on the part of the teacher and the learner. In this document, the former is prioritized due to its role in the educational task of preparing present and future generations. The research aims to define the psychometric structure of the scale of attitudes toward statistics (EAE-2020). With 1,116 statistics professors who made up the faculty at the Specialized University of the Americas (UDELAS) in the Republic of Panama, professors were invited to participate, and 120 responded, representing 10.75% of the universe, with an intentional non-probabilistic snowball selection. For this research, the authors modified elements of the scale of attitudes toward statistics in teaching by Auzmendi (1992). Statistical analysis indicates a global reliability Alpha = 0.855, meaning it has adequate internal consistency. In addition, construct validity was measured through factorial analysis with a total explained variance of 65.27%, represented by four factors: confidence, anxiety, utility, and motivation. The findings allow us to infer that the scale for measuring attitudes towards statistics (EAE) has validity and reliability psychometric characteristics. A distinctive feature of this research is that it was carried out during the global health crisis caused by the COVID-19 pandemic. The results are discussed considering the literature.