Attitude to Mathematics for Primary - Aged Students: A Comparison of True Score and Rasch Measurement
Nova Science Publishers
Community Services, Education and Social Sciences
This report describes the analysis of data (N=774) from 10 items, answered in five response categories (SDA,DA,N,A,SA), for assessing primary-aged students' attitudes towards mathematics, using two methods: True Score (factor analysis and inter-item correlations) and Rasch measurement. Initial factor analysis indicated that the scale comprised three components, which reflected students' overall enjoyment of mathematics, their perceptions of its value as a subject area, and perceptions of their ability to cope with their assigned mathematics work. Data from the three components were generally found to have acceptable reliability and validity across grades four, five) and six. However, analysis with Rasch showed that the five response categories (SDA,DA,N,A,SA) were not answered consistently and logically, that the data were not valid and reliable, and that no linear scale could be constructed with attitudes measured from low to high and item difficulties from easy to hard. The Rasch analysis was repeated with the data reduced to four response categories (SDA,DA,A,SA) and then two categories (D,A), but no linear scale could be constructed. The use of factor analysis and True Score Theory is called into question as a method for producing a linear scale of attitude to mathematics.