การเปรียบเทียบคุณภาพของแบบสอบชนิดเลือกตอบที่มีรูปแบบ ของตัวเลือกแตกต่างกัน ในวิชาคณิตศาสตร์ ระดับชั้นมัธยมศึกษาปีที่ 1 / นิรมล บุญตะรัตน์ = A comparison of the quality of mathayom suksa one mathematics multiple choices tests with different types of choices / Niramon Boontarat
The purposes of this study were to compare the difficulty, the discrimination power, the reliability, the validity of three multiple choices tests with different types of choices: substantive alternatives, none of these and combination alternatives between substantive alternatives and none of these; and to compare the average scores of those three tests. The sample consisted of 300 mathayom suksa one students with 166 boys and 134 girls from Sadao “Khanchaikhamplanon” Secondary School in Songkhla Province. The instruments used were mathayom suksa one mathematics multiple choices tests which were constructed by the researcher. The criterion for the concurrent validity was the students’ achievement on mathematics in the first semester. The difficulty and the discrimination power of the tests were analyzed by 33% technique and the difficulty was transformed to the standard difficulty. The reliability was estimated by Kuder-Richardson formula 20 and the concurrent validity was estimated by Pearson’s Product Moment Correlation Coefficient. The tests’ standard difficulties and the average scores of those three tests were compared by one way analysis of variance technique. The disciminantion power, reliability coefficients and validity coefficients were tested by Chi-Square Test. The results from this research were as follows: 1. The standard difficulty of three mathematics multiple choices tests with different types of choices were 14.2608, 14.7216 and 14.1296 respectively, the discrimination power were .3607, .3507 and .3488 respectively, the reliability were .859, .853 and .850 respectively and the validity were .796, .768 and .807 respectively. 2. The differences in the difficulty, the discrimination power, the reliability and the validity of three mathematics multiple choices tests with different types of choices were not statistically significant. 3. The differences of the students’ average scores on these three tests were also no statistically significant.
