Distinguishing Racism from Ideology

• Published date: 01 September 2015
• Author: L. J. Zigerell
• DOI: 10.1177/1065912915586631
• Date: 01 September 2015
• Subject Matter: Articles

Political Research Quarterly

2015, Vol. 68(3) 521 –536

© 2015 University of Utah

Reprints and permissions:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/1065912915586631

prq.sagepub.com

Article

Blacks in the United States have faced racism in forms

such as slavery, Jim Crow, violence, and separate and

unequal schools. Old-fashioned racism has become less

common over the past few decades (Valentino and Sears

2005, 678), but recent claims have been made about a

“rising tide of anti-black racism” (Bouie 2013) and “over-

whelming” evidence that “racial bias remains deeply

embedded in American life” (Kristof 2014). Much recent

discussion of racism in the United States has focused on

dramatic incidents such as shootings of unarmed black

men, but scholars have also warned of a “new racism”

that is “subtle, institutionalized and seemingly nonracial”

(Bonilla-Silva, cited in Blake 2014).

Perhaps the most prominent form of this new racism is

known as racial resentment or symbolic racism, which

Sears and Henry (2007, 963) called “the most influential

form of racial prejudice in American political life since

the civil rights era of the 1960s.” Research on symbolic

racism has provided evidence that antiblack bias has

widespread influence: symbolic racism predicts white

opposition to policies designed to help blacks (Tarman

and Sears 2005), white opposition to black political can-

didates such as Barack Obama in the 2008 presidential

election (Ford, Maxwell, and Shields 2010), white sup-

port for punitive criminal punishment policies that

negatively affect blacks at a disproportionate rate (Green,

Staerklé, and Sears 2006), and even ostensibly nonracial

behavior such as gun ownership (O’Brien et al. 2013) and

Tea Party membership (Tope, Pickett, and Chiricos 2015).

Symbolic racism has been used to identify the influ-

ence of racism, but several studies have suggested that

symbolic racism to some extent measures ideology

instead of racism (Feldman and Huddy 2005; Sniderman

and Tetlock 1986a). In Study 1, I provide guidelines for

interpreting symbolic racism research so that researchers

can identify when symbolic racism research has identi-

fied an attitude or policy preference influenced by racial

bias; in Study 2, I provide evidence to help researchers

better describe the nature of the racial bias that symbolic

racism identifies, which has been inconsistently described

in the literature in terms ranging from “racial animosity”

(Valentino and Sears 2005, 674) to “some sort of race-

based sentiment” (Rabinowitz et al. 2009, 818).

586631PRQXXX10.1177/1065912915586631Political Research QuarterlyZigerell

research-article2015

1Illinois State University, Normal, USA

Corresponding Author:

L. J. Zigerell, Illinois State University, 401 Schroeder Hall, Normal, IL

61790, USA.

Email: ljzigerell@ilstu.edu

Distinguishing Racism from Ideology:

A Methodological Inquiry

L. J. Zigerell1

Abstract

Research using symbolic racism has provided evidence that racial bias has widespread social and political impact in the

United States, influencing phenomena such as opposition to policies designed to help blacks, disapproval of Barack

Obama, and membership in the Tea Party. However, symbolic racism has a racial component and a conservative

component, so many researchers have attempted to isolate the racial component of symbolic racism with statistical

control; however, the literature lacks guidelines about the effectiveness of such statistical control. To address this

shortcoming, I report results from two studies using the 2012 ANES Time Series Study. Study 1 provides guidelines

for the effect size necessary to support an inference that variation in a dependent variable is influenced by the racial

component of symbolic racism. The nature of this racial component has been inconsistently described in the literature,

so Study 2 reports evidence that symbolic racism sometimes predicts black opposition to policies designed to help

blacks, which suggests that the characterization of the residual effect of symbolic racism as racial animosity is stronger

than warranted by the data. Together, these studies can help researchers better identify when racial bias is an influence

and better understand what this influence represents.

Keywords

race and politics, public opinion, survey methodology, U.S. politics, conservative ideology

522 Political Research Quarterly 68(3)

The Typical Symbolic Racism

Research Design

Symbolic racism reflects complaints about blacks at a

societal and not a personal level (Kinder 1986, 153), in

particular, “contentions that blacks do not try hard enough

to overcome the difficulties they face and that they take

what they have not earned” (Kinder and Sanders 1996,

106). However, symbolic racism is not a pure measure of

racial bias: symbolic racism is instead “the conjunction of

prejudice and values” (Kinder 1986, 156) that “[melds]

ordinary conservatism with some racial animosity”

(Valentino and Sears 2005, 674); this conflation of ideol-

ogy and racial bias can be detected in the following item

from the symbolic racism battery:

It’s really a matter of some people not trying hard enough; if

blacks would only try harder they could be just as well off as

whites.

This statement cannot be used to identify racial bias

because a person who agreed with the statement might

also agree that poor whites who try harder could be just as

well off as middle-class whites. Sniderman and Tetlock

(1986a, 181) noted that high symbolic racism scores can

indicate pure racism or pure conservatism, and factor

analyses have provided evidence that “symbolic racism is

made up about equally of racial prejudice and general

conservatism” (Sears and Henry 2003, 271).

Because symbolic racism is an impure measure, the

effect of symbolic racism cannot be assigned to the racial

component of symbolic racism without first eliminating

the conservative component of symbolic racism. The

typical symbolic racism research design has thus been to

use statistical control to attempt to eliminate symbolic

racism’s conservative component: “ . . . any analyses

done with measures of symbolic racism should be done

controlling for political conservatism in order to mini-

mize the likelihood of spurious associations being misdi-

agnosed as effects of racism” (Pasek et al. 2009, 949).

The typical symbolic racism research design has thus

been to place symbolic racism into a regression with mea-

sures of conservatism and then to interpret the residual

effect of symbolic racism as antiblack bias (e.g., Pasek

et al. 2009; Rabinowitz et al. 2009; Sears et al. 1997;

Valentino and Sears 2005).

However, there is no direct way to assess whether a

particular set of control variables has eliminated the con-

servative component of symbolic racism: for instance,

symbolic racism reflects the belief that hard work causes

success, so if this particular belief does not sufficiently

correlate with statistical control in a model, then that

model will overestimate the racial component of sym-

bolic racism and thus foster incorrect inferences about the

presence and/or strength of racial bias. But while there

might be no direct way to assess the success of statistical

control in a particular model, it is possible to indirectly

assess at an aggregate level the success of statistical con-

trol by observing symbolic racism’s effect on nonracial

dependent variables.

The Effect of Symbolic Racism on

Nonracial Dependent Variables

One way to assess the extent to which particular statisti-

cal control sufficiently dilutes the conservative compo-

nent of symbolic racism is to assess the extent to which

the residual effect of symbolic racism correlates with

dependent variables that tap conservatism but have no

obvious racial component: for instance, if symbolic rac-

ism in the presence of statistical control were to strongly

correlate with the belief that global warming is mostly

due to natural causes, then that would be strong evi-

dence that the statistical control did not sufficiently

dilute the conservative component of symbolic racism

in that model, given that there is no obvious reason

for antiblack bias to influence beliefs about global

warming.

Rabinowitz et al. (2009) presented evidence from two

studies regarding the discriminant validity of symbolic

racism. Based on a sample of seventy-seven white col-

lege students, the first study found that the post-statistical

control effect of symbolic racism positively correlated

with opposition to government agencies setting aside

contracts for black-owned businesses but did not corre-

late with opposition to government agencies setting aside

contracts for female-owned businesses. The second study

presented evidence from the 1990 and 2000 American

National Election Studies (ANES) Time Series Studies

largely consistent with an interpretation of the residual

effect of symbolic racism as antiblack bias: in the 1990

data, symbolic racism positively correlated with white

opposition to preferential hiring and promotion for blacks

and with white support for federal spending on Social

Security but did not correlate with white opposition to

federal spending on food stamps or with white opposition

to federal spending on aid to the homeless. In the 2000

data, symbolic racism positively correlated with white

opposition to preferential hiring and promotion for

blacks, white opposition to compulsory affirmative action

hiring programs for private companies that discriminated

against blacks, white opposition to federal spending on

food stamps, and white support for federal spending on

Social Security, but did not correlate with white opposi-

tion to federal spending on aid to the poor.

Rabinowitz et al. (2009) noted that these results were

consistent with a racial interpretation: symbolic racism