Development of attitude scales towards mathematics and Turkish courses for eighth grade primary school students: A study on validity and reliability

Abstract

This study aimed to develop two attitude scales to measure 8th-grade students’ attitudes toward Mathematics and Turkish courses included in the newly implemented primary education curriculum. Both the Mathematics and Turkish Attitude Scales were designed as 30-item, five-point Likert-type instruments. Each scale consists of 15 positive and 15 negative attitude statements. The study sample consisted of 300 8th-grade students, randomly selected from four public schools located in the metropolitan area of İzmir, Karşıyaka district. To determine the construct validity of the Mathematics Attitude Scale, a factor analysis was conducted, revealing factor loadings ranging from 0.432 to 0.816. The Kaiser-Meyer-Olkin measure was found to be 0.95, and Bartlett’s Test of Sphericity was statistically significant. The corrected item-total correlations and t-values for upper and lower groups were also statistically significant, with item discrimination indices ranging from 0.4177 to 0.8093. The Cronbach’s Alpha reliability coefficient was α = 0.97. Similarly, the Turkish Attitude Scale showed factor loadings ranging from 0.446 to 0.780, with a KMO value of 0.958 and a statistically significant Bartlett’s Test of Sphericity. The corrected item-total correlations and t-values for upper and lower groups were also significant, with item discrimination indices ranging from 0.435 to 0776. The Cronbach’s Alpha coefficient for the Turkish Attitude Scale was found to be α = 0.96. The findings from the validity and reliability analyses indicate that both scales possess a valid and reliable structure.