The measurement tool in this study was developed based on the theory of Haber & Ruyon (1984). It aimed to validate the psychometric properties of self-adjustment using Mixture Rasch Analysis. A total of 217 participants were university students. The results showed that the unidimensionality assumption was met after removing one item from each sub-scale: ACS_6, PSI_12R, and GIR_19R. The 5-point Likert scale functioned effectively, allowing participants to differentiate among various response options. Item reliability for most sub-scales was achieved, ranging from 0.70 – 0.90, except for the PSI sub-scale (0.36). However, person separation for all sub-scales did not meet the criteria, ranging between 0.70 – 1.86 (≤ 3). All items in this study showed no DIF and no latent subclass based on the Rasch Mixture Model analysis results. Overall, the self-adjustment measurement tool demonstrated fairly good performance, except for person reliability and separation, which did not meet the criteria. As a result, the items were less able to capture participants with low and high abilities.
Copyrights © 2025