Rational choice theory in criminology is based on a model of decision-making, expected utility theory (EUT), that research in psychology and economics has shown to be descriptively inaccurate for many types of human decisions. Because of its ability to account for intuitive reasoning and anomalies in crime decisions, prospect theory (PT) may be a viable alternative to EUT for explaining criminal choices. The present study tests several hypotheses derived from PT about the effects of framing, reference dependence, and decision weights in crime decisions. We present findings from a series of survey-based experiments conducted with three different samples. The results support our hypotheses derived from EUT, but not those from PT. We find that although initial wealth, criminal benefits, and arrest risk impact offender decision-making, sanction-cost frames and gains/losses from monetary reference points have little if any effect. We discuss the implications of our findings and directions for future research.