The primary goals of this research were (1) to test the tenability of null associations between perception and reality of punishment and (2) to determine whether the effects of actual risks of punishment on perception of punishment are conditioned by socio-economic characteristics of an area. The importance of the first goal is fundamentally theoretical in that general deterrence mechanism is based on the assumption that perception at least roughly correspond with the reality of punishment. The significance of the second goal is quite practical since maximizing the deterrent effect of law enforcement activities is a major concern from a policy standpoint.
The current study basically replicates what Kleck et al (2004) found using multilevel data. That is, I found no evidence indicative of suppression effects that could have nullified a presumably positive APL and PPL association. Even after excluding potential suppression effects originating from three county-level variables, I failed to identify any significant and positive relationship between APL and PPL. Whether for homicide, robbery, aggravated assault, or burglary, there was no evidence indicating a close correspondence between reality and perception across different punishment types. Moreover, as shown in results of individual-level interaction effects by respondents’ past experience of arrest, HLM estimates among criminals and non-criminals were not significantly different. Criminals, who would be the most attentive to any changes in law enforcement activities in their areas, seemed not to be aware of the overall risks of punishment in their immediate environment.
Tests of county-level interaction between APL and socio-economic conditions of counties such as the percentage of minorities, the unemployment or poverty rates depict a similar picture. Most of the county-level interactions were neither statistically significant nor theoretically consequential, regardless of the types of crime and punishment. Sizes of the interactions were close to zero as well. In short, residents’ perceptions about actual levels of punishment were invariant regarding the conditions of the area in which they lived whether they were high minority, high unemployment or high poverty communities.
