Mathematics Teaching Anxiety Scale: Construction, Reliability and Validity

This study aimed to develop mathematics teaching anxiety scale for prospective primary school teachers. It was designed based on survey method and conducted with four sampling group consisting of 956 prospective primary school teachers at Education Faculties in Turkey. First sampling group was consisted of 404 prospective primary school teachers and 96 out of it were involved in the application of open-ended questions and 308 were involved in exploratory factor analysis. 305 prospective primary school teachers in the second sampling group participated in the confirmatory factor analysis, 108 prospective teachers in the third group were involved in criterion validity and 139 prospective teachers in the fourth one participated in the test-retest reliability analysis. As a result of the principal component analysis of the Mathematics Teaching Anxiety Scale (MTAS), it was found that the scale indicating single factor structure and consisting of 31 items (47.43% of the total variance). After suggested modifications, the scale MTAS was constructed with 19 items. 12 items were removed from the scale and the confirmatory factor analysis (CFA) was carried out with 19 items. According to CFA results (0[less than or equal to]X2 / df = 1.483[less than or equal to]2, RMSEA = 0.040, RMR = 0.050, AGFI = 0.908, TLI = 0.972, CFI = 0.976, IFI = 0.976, GFI = 0.928, NFI = 0.930 and RFI = 0.919), it was confirmed that the scale structure was consisting of 19 items and one dimension. The Cronbach's alpha coefficient of the final form of Mathematics Teaching Anxiety Scale was calculated as 0.93.