Abstract
Prior research has established a strong and enduring “racial divide” in support for capital punishment, but little research has explored the processes that explain the racial divide. Following the lead of Unnever and Cullen (Social Forces 85:1281–1301, 2007a), this research explores whether racist sentiment and core values (individualism, egalitarianism, symbolic patriotism, and authoritarianism) can partially explain the racial divide in public support for capital punishment. The findings suggest that racist sentiment by Whites and belief in core values by Whites partially explains the racial divide in support for capital punishment.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
On their measure of White racism, they coded all African-Americans as a value of zero. They also coded White respondents in the sample whose score was below the African-American mean for the original racism measure as a value of zero. The remaining White racists were coded from one to ten depending on the intensity of their racism (i.e., the degree to which they exceeded the African-American mean value). They found that racism by Whites was a statistically significant predictor for support for capital punishment.
In the pre-election survey there were 1,833 eligible respondents. Of these, 1,212 completed the interview (66.1% response rate). Of these 1,212 cases, 156 cases were deleted due to the respondent not being White (non-Hispanic) or African-American, leaving 1,056 cases. After cases with missing values were deleted using listwise deletion, 573 cases remained for analysis, meaning that approximately 483 cases were not usable due to missing data.
A weight variable was included in the dataset for use in re-analysis of the data. But the weight variable included in the dataset was not used because the weighting was likely impacted by missing cases and the fact that other racial and ethnic groups other than Whites and African-Americans were deleted from the analysis. To correct for this problem a new weight variable was created based only on cases that were included in the sample (cases with no missing data). A proportional weighting strategy was implemented by dividing the sample into different strata based on race, gender, and age (i.e., White males age 15–19, White male age 20–24, White male age 25–29, .......). This approach was used for White males, White females, African-American males, and African-American females in the sample. Once the sample was divided by race, gender, and age categories, the proportion that each race, gender, and age strata comprised for the sample of usable cases was calculated. Then these proportions were compared to proportions in the population (based on Census figures) for similarly constructed race, gender, and age strata. In particular, population data was obtained for White males, White females, African-American males, and African-American females by the same age groups used for the sample (Table 4: Annual Estimates of the White Alone Population by Age and Sex for the United States: April 1, 2000 to July 1, 2006, NC-EST2006–04-WA, Population Division, U.S. Census Bureau, and Table 4: Annual Estimates of the Black or African-American Alone Population by Age and Sex for the United States: April 1, 2000 to July 1, 2006, NC-EST2006–04-BA, Population Division, U.S. Census Bureau). Then, for each strata, the proportion of the strata in the population was divided by the proportion of the strata in the sample. Using this strategy, if the population proportion is equal to the sample proportion the weight is equal to one; where the sample proportion is higher than the population proportion the case is weighted less and where the sample proportion is lower than the population proportion, the case is weighted higher.
An oppose/support dependent variable (1 = support; 0 = oppose) was created. This dependent variable was regressed on the news media consumption and structural location control variables. In the model, the race variable was statistically significant after the controls were implemented. A second model regressed the oppose/support variable on the race measure, the media consumption and structural location controls, the racist sentiment variable, and the core values measures. The race variable was not statistically significant in the model. This means that, in these data, the racist sentiment and core value measures independently accounted for racial divide.
The racist sentiment measure was summed using four different items from the survey. Each respondent was asked whether they believed that: Blacks should work their way up like other groups (1 = strongly disagree; 5 = strongly agree); history makes it more difficult for Blacks to succeed (1 = strongly agree; 5 = strongly disagree); Blacks have gotten less than they deserve (1 = strongly agree; 5 = strongly disagree); and Blacks should try harder to succeed (1 = strongly disagree; 5 = strongly agree).
The individualism measure was summed using three different items. Each respondent was asked to rate services provided by government (1 = government should provide many more services; 7 = government should provide many fewer services); government role in providing services and a good standard of living (1 = government should see to jobs and good standard of living; 7 = government should let each person get ahead on their own); and government assistance to Blacks (1 = government should help Blacks; 7 = Blacks should help themselves).
The egalitarianism measure was summed using three different items. Each respondent was asked whether we have gone to far in pushing for equal rights (1 = strongly agree; 5 = strongly disagree), whether we are better of when we are worried less about equality (1 = strongly agree; 5 = strongly disagree), and whether they would agree that it is not that big of a problem if people are not given an equal chance (1 = strongly agree; 5 = strongly disagree).
The symbolic patriotism measure was summed using three different items. Each respondent was asked: 1) how does the U.S. flag make you feel? (1 = not very good; 4 = very good); 2) how strong is your love for the U.S.? (1 = not very strong; 4 = extremely strong); and 3) how important is it to you to be an American? (1 = not at all important; 5 = extremely important).
The authoritarianism measure was summed using three different items. Each respondent was asked whether it was more important for a child to have: 1) independence or respect (1 = independence, 2 = both, 3 = respect); 2) curiosity or good manners (1 = curiosity, 2 = both, 3 = good manners); and 3) self-reliance or obedience (1 = self-reliance, 2 = both, 3 = obedience).
Four different primary factors were extracted with factor one accounting for 27% of the variance, factor two accounting for 14% of the variance, factor three accounting for 13% of the variance, and factor four accounting for 9% of the variance. The three Individualism measures loaded on factor one (factor loadings of.499,.858, and.617, respectively). The three Symbolic Patriotism measures loaded on factor two (factor loadings of.729,.868, and.763, respectively). The three Authoritarianism measures loaded on factor three (factor loadings of.629,.614, and.600, respectively). The three Egalitarianism factors loaded on factor four (factor loadings of.694,.848, and.538, respectively). Principal Axis Factoring with oblimin rotation was selected because this approach allows the factors to be correlated and provides more conservative factor loading estimates.
The TV News measure is a summed scale from three different items in the survey. Each respondent was asked the number of days they watched national news on television in the past week, how many days they watched local television news in the afternoon/early evening in the past week, and the number of days that they watched local television news in the late evening in the past week (0 = no days, 1 = one day; 2 = 2 days; 3 = 3 days; 4 = 4 days; 5 = 5 days; 6 = 6 days; and 7 = 7 days).
The original social class measure was on an 8-point scale. There were a sizable number of respondents who were coded as missing specifically because the respondent was unsure of whether they were in the working or middle class. These respondents were not treated as missing in the analyses. Instead, they were coded as a 5 (between the working and middle class values) and the value labels for the remaining original values of the variable (5, 6, 7, and 8) were increased by a value of one.
The Religious Practices measure is comprised of two different items. The first item is a survey question that asked whether religion provides guidance in day-to-day living (1 = religion is not important; 2 = some; 3 = quite a bit; 4 = a great deal). The second item is a survey item that asked how often the respondent prays (1 = never; 2 = once a week or less; 3 = a few times a week; 4 = once a day; 5 = several times a day).
The Conservativism measure may under-represent respondents with moderate political views. The final reported value in the NES data was based on an approach that urges people who originally indicate that they are “moderate” to select a classification (liberal or conservative) that best applies to them. If the respondent refuses to do so, the final response is reported as “moderate,” but if the respondent indicates that they are closest to being liberal or conservative, that is what was reported.
The African-American mean value for the racist sentiment measure was 10.45. All African-Americans were coded with a value of zero on the White racist sentiment measure. All White respondents whose score on the original racist sentiment variable exceeded the African-American mean were coded as holding racist sentiment. The magnitude of the coding was determined by the respondent’s score on the racist sentiment measure: 11 = 1; 12 = 2; 13 = 3; 14 = 4; 15 = 5; 16 = 6; 17 = 7; 18 = 8; 19 = 9;and 20 = 10. The number to the left of the equal sign was the score on the racist sentiment measure and the value to the right represented the value on the created White racist sentiment variable. This resulted in an eleven-point White racist sentiment variable that ranged between zero and ten in magnitude.
The African-American mean value for the individualism measure was 5.98. All African-Americans were coded as a value of zero on the White individualism measure. All White respondents whose score on the individualism variable exceeded the African-American mean were coded as a White individualist. The magnitude of the coding was determined by the respondent’s score on the individualism variable: 7 = 1; 8 = 2; 9 = 3; 10 = 4; 11 = 5; 12 = 6; 13 = 7; and 14 = 8. The number to the left of the equal sign was the score on the individualism measure and the value to the right represented the value on the created White individualism variable. This resulted in a “White individualism” measure that ranged in value from zero to eight.
The African-American mean value for the symbolic patriotism measure was 10.17. All African-Americans were coded as a value of zero on the White symbolic patriotism measure. All White respondents whose score on the symbolic patriotism variable exceeded the African-American mean were coded as a White symbolic patriot. The magnitude of the coding was determined by the respondent’s score on the symbolic patriotism variable: 11 = 1; 12 = 2; and 13 = 3. The number to the left of the equal sign was the score on the symbolic patriotism measure and the value to the right represented the value on the created White symbolic patriotism variable. The result was a measure that ranged between the value of zero and three.
The African-American mean value for the Authoritarianism measure was 7.98. All African-Americans were coded as a value of zero on the White authoritarianism measure. All White respondents whose score on the authoritarianism variable exceeded the African-American mean were coded as a White authoritarian. The magnitude of the coding was determined by the respondent’s score on the authoritarianism variable: 9 = 1. The number to the left of the equal sign was the score on the authoritarianism measure and the value to the right represented the value on the created White authoritarian variable.
References
Abramowitz, A. I. (1994). Issue evolution reconsidered: Racial attitudes and partisanship in the U.S. electorate. American Journal of Political Science, 38, 1–24.
Aguirre, A., & Baker, D. V. (1993). Racial prejudice and the death penalty: A research note. Social Issues, 20, 150–156.
Barkan, S. E., & Cohn, S. F. (2005). On reducing white support for the death penalty: A pessimistic appraisal. Criminology and Public Policy, 4, 39–44.
Bohm, R. M. (1991). American death penalty opinion, 1936–1986: A critical examination of the Gallup polls. In R. M. Bohm (Ed.), The death penalty in America: Current research (pp. 113–145). Cincinnati, OH: Anderson.
Carmines, E. G., & Merriman Jr., W. R. (1993). The changing American dilemma: Liberal values and racial policies. In P. M. Sniderman, P. E. Tetlock, & E. G. Carmines (Eds.), Prejudice, politics, and the American dilemma. Stanford, CA: Stanford University Press.
Cochran, J. K., & Chamlin, M. B. (2006). The enduring racial divide in death penalty support. Journal of Criminal Justice, 34, 85–99.
Feagin, J. R. (2001). Racist American: Roots, current realities, and future reparations. Routledge.
Feldman, S. (1999). Economic values and inequality. In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of political attitudes (pp. 159–201). New York: Academic Press.
Green, E. G. T., Staerkle, C., & Sears, D. O. (2006). Symbolic racism and Whites’ attitudes towards punitive and preventive crime policies. Law and Human Behavior, 30, 435–454.
Harlow, R., & Dundes, L. (2004). “United” we stand: Responses to the September 11 attacks in Black and White. Sociological Perspectives, 47, 439–464.
Huddy, L., & Khatib, N. (2007). American patriotism, national identity, and political involvement. American Journal of Political Science, 51, 63–77.
Kosterman, R., & Feshbach, S. (1989). Toward a measure of patriotic and nationalistic attitudes. Political Psychology, 10, 257–274.
Messner, S. F., Baumer, E. P., & Rosenfeld, R. (2006). Distrust of government, the vigilante tradition, and support for capital punishment. Law & Society Review, 40, 559–590.
Pantoja, A. (2006). Against the tide? Core American values and attitudes toward U.S. immigration policy in the 1990’s. Journal of Ethnic and Migration Studies, 32, 515–531.
Roth, B. M. (1994). Prescription for failure: Race relations in the age of social science. New Brunswick, NJ: Transaction.
Sears, D. O., van Laar, C., Carrillo, M., & Kosterman, R. (1997). Is it really racism?: The origin of White Americans’ opposition to race-targeted policies. Public Opinion Quarterly, 61, 16–53.
Sidanius, J., Feshbach, S., Levin, S., & Pratto, F. (1997). The interface between ethnic and national attachment. Public Opinion Quarterly, 61, 103–133.
Skitka, L. J., Bauman, C. W., Aamovich, N. P., & Morgan, G. S. (2006). Confrontational and preventative policy responses to terrorism: Anger wants a fight and fear wants “them” to go away. Basic and Applied Social Psychology, 28, 375–384.
Sniderman, P. M., & Piazza, T. (1993). The scar of race. Cambridge, MA: Harvard University Press.
Soss, J., Langbein, L., & Metelko, A. R. (2003). Why do White Americans support the death penalty? Journal of Politics, 65, 397–421.
Stack, S. (2000). Support for the death penalty: A gender-specific model. Sex Roles, 43, 163–179.
Surlin, S. H. (1981). Race and authoritarianism: Effect on the perception of Roots. Journal of Black Studies, 12, 71–82.
Unnever, J. D., & Cullen, F. T. (2007a). The racial divide in support for the death penalty: Does white racism matter? Social Forces, 85, 1281–1301.
Unnever, J. D., & Cullen, F. T. (2007b). Reassessing the racial divide in support for capital punishment – The continuing significance of race. Journal of Research in Crime and Delinquency, 44, 124–158.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Buckler, K., Davila, M. & Salinas, P.R. Racial Differences in Public Support for the Death Penalty: Can Racist sentiment and Core Values Explain the Racial Divide?. Am J Crim Just 33, 151–165 (2008). https://doi.org/10.1007/s12103-008-9043-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12103-008-9043-1