VALIDITY EVIDENCE FOR A MEASURE OF MOTIVATION FOR SCIENCE

Lately, a decline in the scientific vocations of students has been identified. The expectancy-value model is postulated as one of the most useful theoretical frameworks for understanding academic motivations; however, recent studies suggest the need to add a third construct, called cost. Since there is a lack of research supporting this reconceptualization in the Spanish population, this investigation presents the design and psychometric analysis of a new instrument focused on measuring motivation for school science subjects in elementary education and provides reference data for this educational stage. The research adopted an instrumental design, with a non-probabilistic convenience sample of 339 students from 3rd to 6th grades, drawn from nine schools in the province of Burgos. The confirmatory factor analysis yielded optimal goodness-of-fit levels for the factorial structure composed of three factors: expectancies of success, task value and cost. Matrices of standardized coefficients and correlations between constructs provided evidence of convergent and discriminative validity. A correlational analysis with six attitudinal dimensions of theoretical convergence provided robust evidence of concurrent validity. Likewise, the three factors possessed adequate internal reliability, according to Cronbach's alpha and McDonald's Omega indices. Finally, low levels of expectancies of success expectancies and value beliefs were identified, with no significant differences according to gender or school level. These findings provide empirical support for the expectations-value-cost theoretical model and present a timely, relevant and psychometrically valid and reliable instrument that allows for the development of future research. In addition, they highlight the need to develop educational interventions from elementary stages to foster motivation for the natural sciences.