| Popis: |
The Social Skills Improvement System (SSIS) Rating Scales for students can be used to evaluate student’s social skills and problem behavior. The purpose of this study was to validate and examine the reliability of the Malay version of the SSIS instrument for students aged 8–12 years old using the item analysis method. The 75-item self-administered translated SSIS instrument was introduced to 188 students from two conveniently selected primary schools in Malaysia. The polytomous data were analyzed using the Winstep version 3.80.1, which applied Rasch measurement model based on Item Response Theory (IRT) Models. The instrument was subjected to threshold calibration analysis to ensure the suitability of the scales proposed. Item reliability index was used in examining the instrument reliability, while fit statistics which include the point-measure correlation (PTMEA Corr) index and mean square (MNSQ) values and unidimensionality were examined for instrument construct validity. The results showed the proposed 4-point rating scales were workable. All SSIS subdomains have person reliability values of >0.53 and separation indexes >1.07, with positive PTMEA values for all items. The infit and outfit MNSQ that ranged between 0.5 and 1.5 were used for the purpose of reviewing and retaining items. The findings also showed that 73 were fit items (MNSQ 0.5–1.50), none were overfit and two items were misfits. Further analysis on the ICC of the misfit items suggested that those items could be retained due to careless and erratic responses. In conclusion, the Rasch measurement model could be used to produce empirical evidence of validity and reliability of the translated instrument. Hence, the Malay version of the SSIS instrument for students aged 8–12 years old could then be used to further assessing student’s social skills and problem behavior. |
