Short Version of Self-Assessment Scale of Job Performance

Abstract: This paper aims to reduce the job performance self-assessment scale as well as control the response bias and acquiescence bias using vignettes anchors and inverted items. The original scale database was composed of 20 items divided into two factors: task and context. For the reduction, the ten items with higher factor loads and thresholds were chosen. The reduced scale was estimated by a general factor and two specific dimensions: task and context, representing a bifactor model, with adequate adjustment indicators (RMSEA = .05; TLI = .98). To control response bias and acquiescence, a second study was carried out, in which the responses were recoded and factor analyses were performed in order to make a comparison of the results with and without the use of the vignettes and inverted items. The results indicated that the vignettes improved the factorial loads; however, the reversed items did not perform better than the vignettes.