Researchers employ various measures of risk attitudes to investigate their relation to market behavior with mixed results. We find that a higher-order global risk attitude construct, developed using survey scales and experiments based on expected utility theory, is related to several marketing alternatives, but does not exhibit substantially greater explanatory power than underlying measures. With few exceptions, scales yield greater significance of risk attitudes for these choices, but experimental measures reveal other insights, e.g., differential attitudes in gain and loss domains. Given recent concerns with experimental measures in the literature, we suggest studies include scales as a low cost supplemental measure.