The comparison of ‌the fit person Statistics Precision based on the Rasch model to identify the response style of acquiescence

The aim of the present study is to evaluate the accuracy of person fit statistics based on the Rasch model in identifying acquiescence response styles. To this end, the acquiescence response style was simulated in four different scenarios, including sample size, test length, the percentage of aberrant items, and the percentage of respondents with aberrant response patterns, with 25 repetitions. The findings of the study indicated that the statistics Up and Wp performed better in identifying the acquiescence response style in polytomous data compared to the statistics lp and lzp. However, the accuracy of these statistics decreases as the percentage of respondents with aberrant response patterns increases. Conversely, the lzp and lp statistics were unable to identify this response style in any scenario. Additionally, the examination of the factors affecting the accuracy of person fit statistics revealed that, in general, the accuracy of Up and Wp statistics increases with an increase in sample size, the number of questions, and the percentage of aberrant items. Therefore, in conditions where the likelihood of acquiescence response style exists, researchers can utilize Wp and Up statistics by considering the sample size and questionnaire length. However, given that these statistics fail to identify the acquiescence response style as the percentage of respondents with aberrant response patterns increases, it is recommended to examine the accuracy of other person fit statistics in identifying acquiescence response styles in polytomous data.