Abstract
An often‐repeated claim by conservative commentators attributes continuing liberal beliefs to the fact that progressives “have not been mugged.” This claim thus portrays leftist views on public policy, including crime, as utopian, if not disingenuous—as held by people who have not had to face harsh realities. Using national‐level data from the General Social Survey that span two decades, we test this “mugging thesis.” Controlling for an array of predictors of public opinion, we find no discernible relationship between being a crime victim and having a conservative worldview, support for conservative social policies, or punitiveness toward crime as measured by support for the death penalty and for harsher courts. These results question the validity of the “mugging thesis” and, more generally, of attempts to use slogans to undermine progressive political agendas.
Notes
1 Results from these analyses can be obtained by contacting the first author. Further, we replicated the analyses presented in Tables and without the measure of political orientation included in the regression equations. Being a victim of crime did not significantly predict any of the outcome measures.
2 We tested whether there was a significant association between victimization and our dependent variables while controlling for the specific years in which the GSS was administered. We created a dummy variable for each year omitting the first year in which the GSS was administered. The results from these analyses were substantively identical to those presented in Tables and .
3 Fear measured the degree to which people reported that they felt unsafe because of crime at their home at night or when walking in their neighborhood after dark. The sample size was 862. Other results are available upon request from the authors.
4 We entered the fear of crime measure into three regression equations predicting political ideology, support for the death penalty, and harsher local courts, without the control variables. Fear of crime did not significantly predict political ideology (p = .71). However, fear of crime significantly predicted support for the death penalty (p = .000), but it explained only a negligible amount of the variance, .001 percent. Fear of crime also significantly predicted harsher local courts (p = .000), but again, it only explained .004 percent of the variance. It is also noteworthy that in our analysis of the Pew data, fear was negatively related to being politically conservative.
5 Our data show that victimization is not associated with the dependent variables measuring various conservative beliefs. Our data also do not appear to lend support to the thesis, proposed by Langworthy and Whitehead (Citation1986, p. 584), that: “To the extent that victimization experience is related to punitiveness, it is indirect through fear.” It is statistically improbable to show that victimization has no direct effect on punitiveness and then to argue that people who were victimized were more likely to be punitive because they fear being victimization (see Langworthy & Whitehead, Citation1986). In the absence of a significant direct effect, it is highly unlikely that there are significant indirect effects (Alwin & Hauser, Citation1975). Thus, fear cannot cause people who have been victimized to be more punitive when the data show that people who were victimized were not more punitive. This same argument applies to our analysis of the cumulative GSS. We found that victimization was unrelated to punitiveness. However, we did find that fear predicted both a conservative worldview and punitiveness. Yet, we cannot conclude that victimization causes people to be more punitive because it may be related to fear.
6 A reviewer suggested that the respondent's exposure to the media could have a significant impact on the relationship between victimization and having a more conservative worldview. We constructed two variables that measure the respondents’ exposure to the media (“How often do you read the newspaper—every day, a few times a week, once a week, less than once a week, or never?” and “On the average day, about how many hours do you personally watch television?”). We included these measures in the logistic regression equations presented in Tables and , and found that their inclusion did not alter our substantive finding; people who were victimized did not have a more conservative worldview and were not more punitive.