Are we overconfident in executive overconfidence research? An examination of the convergent and content validity of extant unobtrusive measures

Building understanding of overconfident executives is central to a growing literature that spans a number of disciplines. Much of this research has utilized unobtrusive, or indirect, measures to assess executive overconfidence from secondary data sources. We analyze the convergent and content validity of seven extant unobtrusive measures of executive overconfidence. The results of our analyses indicate that these measures do not exhibit adequate convergence, suggesting that existing measures are not measuring the same construct. Further, we administer a sort task to academic colleagues to assess whether scholars believe that the seven measures are adequately assessing the intended construct. The results of our sort task indicate that scholars did not categorize any of the seven measures as sufficient for measuring overconfidence. We conclude with suggestions for future research to address the inadequate convergent and content validity found in our assessment of extant measures of executive overconfidence.