The Performance of l_z^pPerson-Fit Statistic in Detecting Insufficient Effort Responses According to the Likert Item Scale ‎Points

Abstract

The current study aimed at investigating the performance of  person-fit statistic in detecting insufficient effort responses in noncognitive scales according to Likert item scale points (3, 4, 5 and 7 points). To achieve the aim of this study, cognitive biases scale (DACOBS) was used to collect the data after creating four forms of the scale to represent the number of item-scale points. The four forms of the scale were administered to a sample that consisted of (1495) undergraduate students at Yarmouk University. IRT graded response model was used to estimate item and ability parameters. Then, was calculated for every responder. The results of the current study showed that  distributions do not change over all Likert item scale points (3, 4, 5 and 7), while being not normally distributed and negatively skewed. Moreover, the results showed that the lowest detected rate of insufficient effort responses (IERs) occurred when using three-point Likert type items (10.7%), whereas the highest detection rate occurred when using four -point Likert type items no neutral option (19.7%).