Confidence-more a personality or ability trait? It depends on how it is measured: A comparison of young and older adults

Abstract

The current study (N = 244) compared two independently developed and substantively different measures of self-confidence; a self-report measure, and a measure described as "online." Online measures are confidence-accuracy judgments made following each item on a cognitive task; in the current study, online measures were yoked to tasks of fluid and crystallized intelligence. The self-report and online measures had not previously been compared, and it was unknown if they captured the same self-confidence construct. These measures were also compared to self-efficacy and personality for the purpose of defining self-confidence as an independent construct, as well as to clarify the primary comparison. This study also aimed to replicate previous findings of a stable factor of confidence derived from online measures. An age comparison was made between a young adult sample (30 years and under) and an older adult sample (65 years and over) to determine how confidence functions across the lifespan. The primary finding was that self-report and online measures of confidence define two different but modestly correlated factors. Moreover, the self-report measures sit closer to personality, and the online measures sit closer to ability. While online measures of confidence were distinct from self-efficacy and personality, self-report measures were very closely related to the personality trait Emotional Stability. A general confidence factor-derived from online measures-was identified, and importantly was found in not just young adults but also in older adults. In terms of the age comparison, older adults had higher self-report self-confidence, and tended to be more overconfident in their judgments for online measures; however this overconfidence was more striking in the online measures attached to fluid ability than to crystallized ability.