How likely is ‘likely’? Subjective perception and communication of victimization probabilities

Abstract

What is the meaning of ‘likely’, when respondents estimate the risk to become victim of crime? On the one hand, such risks can be presented and interpreted as gains, in terms of being spared of becoming a victim of crime. On the other hand, such risks can be presented and interpreted as losses, in terms of becoming a victim of crime. We demonstrate a framing-effect with data from an experimental survey in Leipzig. Because losses are perceived as more severe, respondents state lower subjective victimization probabilities in the loss-frame, compared to the gain-frame. In addition, the meaning of vague quantifiers varies with the frequency of an event: The frequent event is ‘being spared of a crime’; the infrequent event is ‘becoming a victim of crime’. We show that the same verbal likelihood labels (e.g. ‘unlikely’ or ‘very unlikely’) are more often chosen (in terms of percentages) for frequent than for infrequent events. Finally, with increasing severity of offences, respondents assign for their likelihood estimate lower percentages to the same verbal label. Our evidence shows that vague quantifiers cannot be interpreted literally. Because respondents do not use vague quantifiers consistently for different victimizations, it is problematic to compare subjective victimization risks across different victimizations.