อิทธิพลของวิธีวัดแบบมาตรส่วนประมาณค่าและบังคับเลือก ที่มีต่อความเที่ยงและความตรงเชิงโครงสร้างในการวัดบุคลิกภาพ 5 องค์ประกอบ
Keywords:
METHOD EFFECTS, BIG FIVE PERSONALITY, RATING SCALE, FORCED-CHOICE, CEUL TECHNIQUES, CTCM TECHNIQUESAbstract
Abstract
The purposes of this research were 1) To compare the reliability of big five personality traits measurement on rating scale and forced-choice, 2) To check construct validity of big five personality traits measurement on rating scale and forced-choice, 3) To analyze effects of rating scale and forced-choice measurement method in big five personality traits measurement by CEUL techniques and CTCM techniques and 4) To analyze effects of positive items and negative items measurement method in big five personality traits measurement by CEUL techniques and CTCM techniques. The sample consisted of 1,200 student in grade 12 under the basic education commission, Bangkok. The research instrument consisted of a big five personality traits rating scale and forced-choice. Data analysis by z-test and confirmatory factor analysis.
The research results were as follow: 1) The internal consistency reliability using Cronbach’s alpha coefficient of big five personality traits rating scale is higher than forced-choice, and the stability reliability by using the Pearson product moment correlation coefficient is higher than rating scale of the level of statistical significance .01, 2) The confirmatory factor analysis of big five personality traits rating scale and forced-choice no fit to the empirical data. ( ; p<.05, /df >2, RMSEA>.05, and CFI<.90), 3) Analysis results was effects of rating scale and forced-choice measurement methods in big five personality traits measurement by CEUL techniques more suitable than by CTCM techniques (Test 1; =11207.09, df=1640, /df=6.83, RMSEA=0.07, CFI=0.84; Test 2; =8057.27, df=1640, /df=4.9, RMSEA=0.057, CFI=0.85), 4) Analysis results was effects of positive items and negative items measurement methods in big five personality traits measurement by CTCM techniques more suitable than by CEUL techniques with rating scale only. (Test 1; =10426.07, df=1639, /df=6.36, RMSEA=.067, CFI=.84; Test 2; =7511.08, df=1639, /df=4.58, RMSEA=.055, CFI=.86) and 5) The best techniques in examination effects of rating scale and forced-choice measurement methods in big five personality traits measurement is CEUL techniques.
