The validity and reliability study of the Indonesian version of the sources of mathematics self-efficacy scale

Abstract

Considerable research carried out in Indonesia heavily focused on how self-efficacy is related to other variables. However, there is no study devoted to developing and validating scale measuring students' self-efficacy in mathematics learning. Therefore, the aim of the present study was to examine the validity and reliability of the Indonesian version of the Sources of Mathematics Self-Efficacy Scale. Using a cross-sectional survey with a convenience sampling, a total of 978 high school students (557 were girls and 421 were boys) from five different schools in three urban cities in Indonesia participated in this study. Based on the result of Confirmatory Factor Analysis, it was found that the four-dimensional model was well fitted. The value of goodness of fit index (GFI) of the model was χ^2 (df) = 546.08 (246), the root mean square error of approximation (RMSEA)=.07, the normed fit index (NFI)=.90, and the comparative fit index (CFI)=.92. Internal Consistency was examined through calculating Cronbach's alpha and corrected-total item correlation coefficient. The values of Cronbach's alpha for the factors ranged between 0.754 and 0.909. In addition, the corrected-total item correlation coefficients indicated a favorable level of correlation. Overall findings showed that the scale had acceptable validity and reliability score and it might be applied to identify Indonesian students' sources of mathematics self-efficacy.