การพัฒนาแบบวัดทักษะการคิดอย่างมีวิจารณญาณในศตวรรษที่ 21 ของนักเรียนชั้นมัธยมศึกษาตอนต้น โดยประยุกต์ใช้ทฤษฎีการตอบสนองข้อสอบแบบพหุวิภาค

Abstract

The purpose of this research were 1) to construct of measurement scale on critical thinking skills of the 21st century for lower secondary school students and 2) to examine the quality of measurement scale on critical thinking skills of the 21st century for lower secondary school students by applying polytomous Item response theory. The sample were 600 students selected by stratified random sampling technique. The students selected were studying in lower secondary school under the jurisdiction of the secondary education office service area 15. The measurement scale designed the situational test as a four-multiple choice was performed via the Classical Test Theory (CTT) by content validity, concurrent validity, construct validity with accomplished by confirmatory factor analysis (CFA), discrimination and reliability with Cronbach's alpha coefficient. Moreover, the parameter testing was accomplished by Polytomous (IRT), with discriminant and difficulty, and test information (TIF) was conducted via Grade Response Model (GRM). The findings were as follows: 1. The measurement scale on critical thinking skills of the 21st century for lower secondary school students composed five components with has 36 items including 1) defining skills 2) data gathering skills 3) analyze skills 4) synthetic skills and 5) evaluate skills. The content validity of the scale range from 0.80 to 1.00, analysis of discriminatory with t-test, 36 items that passed the selection criteria were obtained and the reliability was 0.911 2. The quality measurement scale by the concurrent validity with analysis the correlation coefficient between the skills test results and students’ GPA show that the skills test results correlate with students’ GPA were significantly at the level of 0.01, with a high level of correlate and the construct validity of the measurement scale using confirmatory factor analysis (CFA) consisted of five factors: defining skills, data gathering skills, analyze skills, synthetic skills and evaluate skills. Also the model fitted to the empirical data indicated by Chi-square= 5.67, df = 5, P-value = 0.33944 , RMSEA = 0.037, RMR = 0.045, CFI = 1.00, GFI = 0.98 and the weight of standard factor was between 0.57 to 0.84 when examining the quality of the measurement based on the Polytomous IRT. The slope parameter (α) is between 0.06 to 1.72 and the difficulty value the threshold value of each item (β) which all of the values in ascending items and the reliability of the measurement scale using Marginal Maximum-Likelihood (MML) showed that reliability of the test was 0.885. Moreover, the value of test information of item of measurement scale indicated was able to accurately and precise analyze in students among competent whose competency θ was between -2.0 to +0.5 or the students of low to medium competency levels.