Trait anxiety as a determinant of psychological test results

Abstract: This thesis deals with the problem of interpreting test scores. The intention has been to explore and elucidate the effects of trait anxiety on test scores that have been obtained in applied research settings. Paper I showed that different relationships between personality traits and cognitive functioning emerged in a group of men with solvent induced chronic toxic encephalopathy (TE; N=57) compared with a group of demographically similar healthy men (N=57). Within the referent group high and consistent correlations were observed between the anxiety scales and reaction times measures. This pattern did not appear in the TE group; instead, divergent and weak relationships emerged. By dividing the referent group into low anxiety and high anxiety subgroups, it was shown that the neuropsychological test scores in the high anxiety subgroup mostly were indistinguishable from the scores in the TE group. Paper II investigated the relationship between anxiety and defensive strategies as measured by the Meta-Contrast Technique (MCT) and self-reported trait anxiety, trait aggression and defensiveness, as measured by the Karolinska Scales of Personality (KSP) in healthy men (N=83). The results indicated no agreement between the KSP and MCT. Paper III investigated whether trait anxiety and social desirability influenced glycated hemoglobin levels (HbA1c) or - inventories of self reported symptoms, psychosocial work exposure, and social interaction in healthy men (N=98). The results indicated that the high-anxiety half of the subjects typically reported more symptoms, lower decision latitude, and fewer social interactions than the low-anxiety half. Occasionally, main effects of social desirability were observed. Paper IV addressed the relationship between personality traits on the one hand and, on the other hand, neuropsychological test performance, self-rated distress symptoms, psychosocial work exposure, and social interaction patterns, in a group of healthy women (N=101). Few relationships between personality traits and neuropsychological test results were observed. As for the rating inventories, the high-anxiety half of the subjects typically reported more symptoms and lower social interaction scores than the low-anxiety half. In conclusion, the findings indicate the importance of checking for, and possibly controlling for, the anxiety dimension when using self-report inventories in applied research settings.