A reliability and validity study of the mathematical literacy perception scale

This study was conducted to develop a measurement tool to measure perceptions of mathematics literacy for 7th grade students in secondary school. The development of the scale was carried out in six stages. In the first step, a survey of the literature was conducted to form a pool of items. In this context, the basic skills and competences for mathematical literacy were tried to be determined. However, by examining the literature on social learning theory, the perception of self-efficacy and affecting factors were investigated. In addition, interviews with the teachers affiliated to the Ministry of National Education were tried to reveal what the affective characteristics of a mathematical literate individual should have. The second step in the survey was to determine the type of the scale. In this context, the 5-point Likert-type scale, which is the most commonly used scale type, was chosen for measuring the level of participants, ease of use and affective characteristics. In the third stage, expert opinions were applied. Firstly, the opinions of the experts of the three subject areas were taken and the simplification of the items which did not serve the purpose or served the same purpose was provided. Then, by taking the opinions of the expert of 3 language field experts, simplification in terms of grammar, level and expression was realized and the item pool was reduced to 40 items. In the fourth step, intelligibility implementation was carried out in order to predict the problems to be encountered in the actual application. The main purpose of the implementation is to test the intelligibility of the scale. In this context implementation has not been found any items to be difficult to understand or misunderstood. The draft scale prepared in the fifth step was applied to the selected sample with the typical case sampling method, and the exploratory and confirmatory factor analysis was performed. The analysis process, which started with an exploratory factor analysis, was followed by factoring procedures after testing the suitability of the data. When there is a moderate relation between the factors, promax rotation technique is applied and a scale consisting of 4 factors and 30 items is obtained. The first level confirmatory factor analysis was performed to increase the construct validity of the obtained scale. It has been found that the t value obtained in the basic parameter measurements is at the level of .01, and it is seen that the cut off score of the compliance goodness indices examined are above the cut off score's values. As a final step, it was observed that the items - total correlations of items of scale were between .351 and .686. The Cronbach Alpha coefficient used for internal consistency measures of the scale was calculated as .925. The difference between the mean scores of the upper and lower 27% groups performed to test the distinctiveness was found to be significant at .01 level.