The Effectiveness of the Number of Extra Negative Items in Detecting Insufficient Effort Responses in Non-Cognitive Scales

Abstract

This study attempted to determine the effectiveness of the number of extra negative items in identifying insufficient effort responses in Attitudes toward Statistics (SATS-36). The (SATS-36), which consists of 36 5-point Likert Scale items, was used to actually achieve the goal of this study. Furthermore, the researchers developed three forms, each with a different set of extra negative items (2, 4, and 6). The three forms were distributed to a sample of (750) students at Yarmouk University. The results revealed that form 1, which enclosed 6 extra negative items, had the lowest detected rate of insufficient effort responses (IERs) (7.20%), while form 3, which contained 2 extra negative items, had the highest detection rate (15.6%). The detection rate was discovered respectively among the Lie Detection Scale, Mahalanobois, lpz and the method of extra negative items. The highest detection rate was found in form 3 with two extra negative items, and data reliability decreased after the insufficient effort responses were removed (IERs). Furthermore, the results showed that the maximum changes in model-data fit indices after data filtering were in form 3, which contained two extra negative items. Moreover, the results indicate that the reliability of data after filtering those with insufficient effort responses (IERs) is reduced.