Instrumen sistem kepercayaan tidak rasional dalam kalangan mahasiswa di universiti awam

• Main Author: Romele, Mohd. Nur Al Sufi
• Format: Thesis
• Language: English
• Published: 2016
• Subjects: LB Theory and practice of education
• Online Access: http://eprints.utm.my/id/eprint/78224/1/MohdNurAlPFP2016.pdf; http://eprints.utm.my/id/eprint/78224/; http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:97124

The purpose of this study is to develop a Belief System Detector Instrument among university students. The constructs of this study were values, experiences, education and sosioeconomic status. The research design used in this study was exploratory sequential design by using qualitative and quantitative methods. In qualititative study, there were 24 selected respondents in focus groups based on semi structural interviews while the quantitative study involved 240 respondents in the actual survey. All respondents were selected among university students who are in pure, social and engineering sciences in Malaysia. The collected qualitative data went through content analysis using NVivo software, whilequantitative data were analysed through Rasch Model using Winstep Software Version 3.69.1.11. Preliminary results showed that 81 items were identified and after the evaluation, 36 items were used. The respondents’ reliability index for the pilot study was 0.83, while the item reliability index was 0.98. The separation of respondents was 5.20 and the separation of the items was 6.36. Then the actual analysis was conducted with the findings of respondents’ realibilty index analysis was 0.84, while the item reliability index was 0.99. The separation of respondents was 6.30 and the separation of items was 8.85. A value of Cronbach Alpha showed the reliability of item was 0.87. The Infit MNSQ value ranged between 0.89 to 1.20 and the outfit MNSQ was in the range of 0.88 to 1.28, showing the suitability of the measuring instrument. The actual analysis showed the variance explained by measures (%) indicated that the instrumen has a value of 57.0 and the unexplained variance in 1st contrast for all items was 2.3% (very good). Differential Item Functioning values for genders was not significant, indicating that this item can be used by all genders.This study discovered a few new constructs in measuring belief system, which are value, experience, education and socioeconomic status. It can be applied into the Belief System Detector Instrument that can be used to help counsellors and students in identifying irrational belief system.