We show that perceptions of insecurity are strongly correlated with victimization at the individual level and suggest that the reason this relationship is usually not uncovered in simple cross country studies is that they fail to consider observable differences across individuals who may face different perceptions of insecurity as well as different probabilities of victimization. Our findings are consistent with a model where individuals rank crime relative to other problems depending on whether they have developed coping mechanisms to attenuate the impact of crime on their wellbeing, the implication is that the weight individuals' place on crime depends more on changes in actual crime than on its level. We do not claim to establish a causal relationship between victimization and perception, but we raise the bar for claims of the absence of a correlation.