![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
ABSTRACT
We outline four connections between xenophobia and punitiveness toward criminals in a national sample of Americans. First, among self-identified whites xenophobia is more predictive of punitiveness than specific forms of racial animus. Second, xenophobia and punitiveness are strongly connected among whites, but are only moderately and weakly related among black and Hispanic Americans, respectively. Third, among whites substantial proportions of the variance between sociodemographic, political, and religious predictors of punitiveness are mediated by levels of xenophobia. Finally, xenophobia is the strongest overall predictor of punitiveness among whites. Overall, xenophobia is an essential aspect of understanding public punitiveness, particularly among whites.
Acknowledgment
The 2014 Chapman Survey of American Fears was funded by the Earl Babbie Research Center at Chapman University and the John Templeton Foundation.
Notes
1. Quote is from authors’ transcript of the speech. Full text of Trump’s presidential announcement speech is available at: http://time.com/3923128/donald-trump-announcement-speech/.
2. For instance, in the 2014 Chapman Survey of American Fears, among white respondents who rated “blacks” below 50 on the feeling thermometer, 78% also rated “immigrants” below 50. Conversely, of white respondents who rated “immigrants” below 50 on the feeling thermometer, only 35% rated “blacks” below 50.
3. The sociodemographic frequencies in the Chapman Survey compare favorably to those from the 2014 General Social Survey. One area of difference was that the Chapman Survey has significantly more currently married respondents (58.9%) than the GSS (45.7%). Tabled results of the comparisons are available upon request.
4. Supplemental analyses with a xenophobia index weighted by factor loadings produced results identical to those presented.
5. We did not use a question from this battery about whether police should be allowed to conduct raids to find undocumented immigrants because of its similarity to the punitiveness outcomes. In support of the centrality of economic concerns in racial threat theory, the measure for whether “immigrants are a drain on the economy” had a stronger relationship to the punitive ideology outcome (r = .498) than the question about whether “immigrants are more likely to commit crime” (r = .399).
6. The xenophobia index had high reliability for scaling across the racial and ethnic categories. For black respondents, the items loaded on a single factor with an Eigenvalue of 3.5. For Hispanic and mixed race respondents there were single factors with Eigenvalues of 3.7 and 3.8, respectively.
7. We also tested models that used a series of dummy variables for education and income levels. Results did not differ substantially from those presented, with college graduates being significantly less punitive, and non-significant results for income levels. We also tested for potential non-linear effects of income using logged and quadratic measures. These results were statistically non-significant.
8. We created this index of “belief orthodoxy” that included views of the Bible for consonance with previous research on punitiveness, but supplemental analyses examining the individual indicators (belief in an active Satan, belief in future Armageddon, and Bible views) showed that beliefs about Satan were more predictive of punitive ideology than beliefs about the Bible, Armageddon, or the overall orthodoxy index. This provides a replication of the recent finding that an important aspect of religiosity for predicting punitiveness is views of supernatural evil (Baker and Booth Citation2016).
9. Due to the marginal reliability of the TV viewing index, we conducted supplementary analyses excluding this control. Focal results for the connections between xenophobia and punitiveness were unchanged. Among the specific indicators, punitiveness had the highest correlation with consuming “true crime” television shows.
10. P < .001 for paired samples T-tests of the differences between mean scores on the “immigrants” thermometer compared to the thermometers for “blacks” and “Hispanics.”
11. Z tests for significant differences of coefficients across Models 1–3 (see Paternoster et al. Citation1998) showed a statistically significant difference between the coefficients for “blacks” and “immigrants” (Z = 2.40; p = .01) and a marginally significant difference between the coefficients for “Hispanics” and “immigrants” (Z = 1.60; p = .05, one-tailed tests).
12. The xenophobia index (VIF = 1.8) had a relatively low level of multicollinearity in this model. The feeling thermometer measures had elevated, but tolerable, levels of multicollinearity (VIF = 2.6 for “blacks” and “Hispanics” thermometers; VIF = 3.0 for “immigrants” thermometer).
13. Quotations are from authors’ transcripts of both speeches. Textual transcripts are publically available for the nomination speech (http://www.politico.com/story/2016/07/full-transcript-donald-trump-nomination-acceptance-speech-at-rnc-225974) and the immigration policy speech (http://www.latimes.com/politics/la-na-pol-donald-trump-immigration-speech-transcript-20160831-snap-htmlstory.html).