With respect to the subjectivity of performance evaluations, this is the nature of virtually all performance criteria in organizational settings
(e.g., a strong majority of the criteria in the Barrick & Mount [1991] personality-job performance meta-analysis were subjective). Objective
performance data are generally only available for sales positions, which were not studied here. More importantly, we are aware of no data
suggesting that core self-evaluations uniquely predict subjective performance criteria better than objective criteria. In fact, there is evidence that the associations are similar (Erez & Judge, 2001). With respect to the concurrent nature of the study, it is possible that performance caused core self-evaluations, though, again, we are unaware of evidence on this point. Furthermore, the individual core trait of self-esteem appears to be as heritable (Roy, Neale, & Kendler, 1995) and stable (Conley, 1984) as Neuroticism, and we are unaware of such claims being made with respect to Neuroticism. Third, it is unlikely that ability is an important confounding variable here, as evidence reliably indicates low correlations between personality and cognitive ability (McHenry, Hough, Toquam, & Hanson, 1990). Finally, with respect to the self-report nature of the criteria, it certainly is true the associations here may be inflated by response bias (assuming core self-evaluations is itself not a theoretically relevant source of response bias). However, we were careful to collect data that were not self-reported, including job performance in two studies, and life satisfaction in Study 4 was measured several months after the CSES was administered. Even under these conditions, the CSES still seemed to have empirical validity, suggesting it does not merely reflect the tendency to respond more positively to survey items (though we certainly agree this is part of what core self-evaluations is).