This study asks to what extent (a) individuals show consistent performance differences across typical behavioral decision-making tasks, and (b) how those differences correlate with plausible real-world correlates of good decision making. Seven tasks, chosen to span the domain of decision-making skills, were administered to participants in an ongoing longitudinal study providing extensive social, psychological, and behavioral measures. Performance scores on individual tasks generally showed small, positive inter-task correlations. An aggregate measure of decision-making competence (DMC) was appropriately correlated with plausible sources, concomitants, and outcomes of good decision making, suggesting the underlying construct's external validity. Higher DMC scores were associated with more intact social environments, more constructive cognitive styles, and fewer ‘maladaptive’ risk behaviors. In each case, DMC adds to the predictive validity of general measures of cognitive ability. These results suggest that poor decision making on common laboratory tasks is related to real-world antecedents and consequences of poor decision making. Copyright © 2005 John Wiley & Sons, Ltd.