PSC 270 UT White & Black Voters Estimate Black Candidates Ideology Discussion Each article summary is worth 7 total points. The assignments have four compo

PSC 270 UT White & Black Voters Estimate Black Candidates Ideology Discussion Each article summary is worth 7 total points. The assignments have four components: 3 to 4 pages
a brief summary of the article’s main research question, theory, and main hypothesis (or hypotheses) (1 point);
a description of the authors’ dependent variable(s) and key independent variables (the variables that connect most directly to their research question) (2 points);
a brief description of the authors’ “control” variables (1 point); and
a summary of the article’s main findings, including using p-values and/or confidence intervals to interpret the regression results for the key independent variables (you don’t need to worry about the control variables) (3 points)Lerman, Amy E., and Meredith L. Sadin. 2016. “Stereotyping or Projection? How White and Black Voters Estimate Black Candidates’ Ideology.” Political Psychology 37(2):147- 163. Stereotyping or Projection? How White and Black Voters Estimate Black Candidates’
Ideology
Author(s): Amy E. Lerman and Meredith L. Sadin
Source: Political Psychology, Vol. 37, No. 2 (APRIL 2016), pp. 147-163
Published by: International Society of Political Psychology
Stable URL: https://www.jstor.org/stable/44132867
Accessed: 10-04-2020 23:00 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

Terms and Conditions of Use

Don't use plagiarized sources. Get Your Custom Essay on

PSC 270 UT White & Black Voters Estimate Black Candidates Ideology Discussion Each article summary is worth 7 total points. The assignments have four compo

Just from $13/Page

Order Essay

International Society of Political Psychology is collaborating with JSTOR to digitize,
preserve and extend access to Political Psychology
This content downloaded from 132.174.251.96 on Fri, 10 Apr 2020 23:00:22 UTC
All use subject to https://about.jstor.org/terms
Political Psychology, Vol. 37, No. 2, 2016
doi: 10.1 lll/pops.12235
Stereotyping or Projection? How White and Black Voters
Estimate Black Candidates’ Ideology
Amy E. Lerman
University of California , Berkeley
Meredith L. Sadin
University of California , Berkeley
In studying the electoral fortunes of Black candidates, scholars have almost exclusively focused on W
voters’ attitudes . In this article , we employ a set of randomized experiments and nationally represent
survey data to examine how both Black and White voters evaluate the ideology of racially diverse candidat
In line with previous research, we find evidence that White voters stereotype Black candidates as being m
liberal than White candidates in three of our four tests. In contrast, we find that Black voters – particularly th
who identify as politically conservative- project their own ideology onto Black candidates. These findings h
electoral importance because, as we show, vote choice for both Blacks and Whites is substantially mediated
perceived ideological distance from a candidate. These findings also enable us to better understand the man
in which Black voters navigate a trade-off between descriptive and substantive representation.
KEY WORDS: projection, stereotyping, vote choice, descriptive representation, Black politics
The notion that African American candidates have a better chance of succeeding in majori
minority districts, as opposed to majority White districts, remains a “widely accepted tr
(Highton, 2004, p. 2). However, scholars interested in the microfoundations of Black candidate
electoral fortunes have focused their attention almost exclusively on the attitudes of White vot
often relying on the implicit – and sometimes explicit – assumption that Black voters simply pr
Black candidates. In fact, very little work has probed deeper in order to explore the underlyin
dynamics of this preference.
The result of this gap in the literature is that we still know little about Black voters’ percept
of Black candidates. Do Black voters utilize similar stereotypes as White voters? If not, w
types of heuristics do they use when evaluating Black candidates? And how does the strength
content of these cognitive shortcuts vary across subgroups of Black voters? To the best of ou
knowledge, no rigorous empirical test has yet been conducted of these important questions.
In this article, we present the results of a series of randomized survey experiments, as well
analyses of the American National Election Study (ANES) and the Cooperative Congressio
1 For examples, see Berinsky et al. (2011); Bullock and Dunn (1999); Citrin et al. (1990); Colleau et al. (1990); Jones
Clemons (1993); McDermott (1998); Moskowitz and Stroh (1994); Piliawsky (1989); Sigelman et al. (1995); and Terkild
(1993).
147
0162-895X © 2014 International Society of Political Psychology
Published by Wiley Periodicals, Inc., 350 Main Street, Maiden, MA 02148, USA, 9600 Garsington Road, Oxford, OX4 2DQ
and PO Box 378 Carlton South, 3053 Victoria, Australia
This content downloaded from 132.174.251.96 on Fri, 10 Apr 2020 23:00:22 UTC
All use subject to https://about.jstor.org/terms
148
Lerman
Election
and
Sadin
Study
(CC
for
publ
candidates
nuanced
than
prev
differen
research,
we
find
e
a
White
candidate,
projection;
Black
l
rable
White
candid
than
his
White
cou
perceived
ideologic
distinctly
The Candidate Race Heuristic
A candidate’s race can serve as a “heuristic” or cognitive shortcut for political decision m
(Popkin, 1991). Previous research has shown that Whites hold clearly defined “belief stere
about Black candidates, perceiving them as more liberal than White candidates regardless
policy platform (Citrin, Green, & Sears, 1990; Huddy & Terkildsen, 1993; McDermot
Moskowitz & Stroh, 1994; Sigelman, Sigelman, Walkosz, & Nitz, 1995; Williams, 1989
belief stereotypes have been shown to be influential in predicting the probability that Whites w
for a Black candidate (McDermott, 1998).
We should be similarly concerned with the cognitive processes that underlie Blacks’ pre
for Black candidates. Yet, a review of the relevant scholarship highlights the glaring omi
Black voters from extant scholarship on voting behavior (Callaghan & Terkildsen, 2002). W
Black voters have been shown to feel favorably towards Black candidates (Tate, 2003; Will
1989), evidence on the inferences they make about the beliefs of Black candidates remains
developed.3
This is a sizable oversight, as understanding the psychological mechanisms of Black vo
behavior enables us to better understand, as well as to differentiate between, citizens’ prefer
descriptive and substantive representation. Descriptive representation – or, whether a legisl
as an agent of a particular group (Pitkin, 1967) – has been shown to be important to
Americans (Mansbridge, 1999; Sullivan & Johnson, 2008; Tate, 2001, 2003). However, w
little about how Black voters infer the ideological positions of Black candidates, an issue th
the heart of notions of substantive representation, which is concerned with whether the
positions of a candidate are in line with the ideological preferences of the voter.4
Importantly, although African Americans are overwhelmingly Democratic, they are n
logically monolithic (Bobo & Gilliam, 1990; Canon, 1999; Gay, 2004; Gilliam, 1996; G
Keane, 2006; Hajnal & Lee, 201 1). In the 2008 ANES, for example, only 35% of Black respo
called themselves liberals, while about 36% identified as moderate and 29% identified as c
tive. As such, while descriptive and substantive representation are by no means mutually ex
they are likely to conflict at times for a sizable subset of Black voters (Cameron, Ep
O’Halloran, 1996; Lublin, 1999; Swain, 1993).
When a Black voter shares her ideological position with a Black candidate, her preferenc
both descriptive and substantive representation will likely lead her to support that c
2 Throughout this article, the category “White” excludes those who identify as Hispanic.
3 In fact, there is limited experimental evidence that candidate race influences support among African Americans.
and Walton [2007] and Sullivan and Arbuthnot [2007] for important exceptions.)
4 We sidestep here the important theoretical concern of whether political representatives should be “delegates” or
in representing their constituents’ ideological preferences (Dovi, 2011). Our focus is instead on the narrowe
question of whether voters prefer a candidate whose ideology is congruent with their own and how that con
evaluated.
This content downloaded from 132.174.251.96 on Fri, 10 Apr 2020 23:00:22 UTC
All use subject to https://about.jstor.org/terms
Stereotyping
However,
candidate,
unless
or
Projection?
when
the
we
149
considering
two
know
a
Wh
preferences
her
decision
m
ru
behavior is determining whether
in spite of political ideology.
A
Similarity
Contingenc
When Black voters experience a
and the ideological congruence be
generally
collective
1995;
posit that linked fate status of the group – w
Guinier,
Sullivan
&
1994;
Gurin,
Arbuthnot,
representation
p. 173).
in
2007).
order
to
However, we argue that the extent to which Black voters face a conscious trade-off between
descriptive and substantive representation will be determined by the extent to which they actually
perceive a discrepancy between their own ideological position and that of a Black candidate.
Specifically, we outline here a model in which Black voters rely on a projection heuristic when
evaluating a Black candidate. Our theory thus challenges the conventional wisdom that a sense of
linked fate among Black voters leads them to sacrifice substantive for descriptive representation.
Instead, we suggest that psychological projection causes Black voters to perceive both ends as
achievable by supporting Black candidates.
Extant research has shown that individuals are prone to project – or overestimate the probability
of sharing a characteristic with another individual in the population (Katz, Allport, & Jenness, 1931,
Ross, Greene, & House, 1977; for reviews, see Krueger, 2000; Marks & Miller, 1987) – and that
projection is moderated by social categorization, whereby individuals show greater projection of
their own beliefs onto members of in-groups than members of out-groups (Berelson, Lazarsfeld, &
McPhee, 1954; Clement & Krueger, 2002; Holtz & Miller, 1985; Krueger & Zeiger, 1993; Wilder,
1984). When applied to an electoral context, projection is moderated by the degree to which a voter
is linked to a candidate by “shared political party affiliation, shared racial or ethnic identity, or some
other shared characteristic” (Krosnick, 2002, p. 119).
We extend the theory of electoral projection with a more nuanced similarity contingency model ,
which suggests that when individuals have (or perceive themselves to have) higher similarity with a
target, they are more likely to project. Conversely, when they have lower levels of similarity, they are
more likely to rely on stereotypes (Ames, 2004). We hypothesize that this basic idea helps to explain
the divergent ways in which Black and White voters evaluate Black candidates.
Hypotheses
Scholars have acknowledged a decline in old-fashioned racism among White voters and assert
that a new form of “symbolic racism” has developed, such that Whites’ negative affect towards
Blacks stems from the perception that they do not possess the same traditional, conservative values
(McConahay & Hough, 1976; Sears, 1993, 2003; Sears & Henry, 2005; Sears & Kinder, 1971; Sears
& McConahay, 1973).5 Trait stereotypes about a social group (e.g., the idea that Blacks are lazy) can
activate belief stereotypes (e.g., that Blacks are more liberal) (Berinsky & Mendelberg, 2005). Given
5 For critiques, see Sniderman and Tetlock (1986a, 1986b) and Sniderman, Brody, and Tetlock (1991).
This content downloaded from 132.174.251.96 on Fri, 10 Apr 2020 23:00:22 UTC
All use subject to https://about.jstor.org/terms
Ha
Thu
elect
150
Lerman
and
Sadin
this
and
related
re
al.,
1995;
William
et
more liberal than an identical White candidate .
In contrast, high levels of in-group identification among Black Americans lead us to expect that
Black voters will rely on projection (rather than stereotyping) when evaluating Black candidates.
Black Americans are more likely than Whites to identify with their racial group (Citrin et al., 2001;
Harris, 2008; Wong & Cho, 2005), share a stronger sense of group solidarity (Dawson, 2001, 2006;
Dawson & Popoff, 2004; Hajnal & Lee, 2011), and report that their racial identity influences their
political identity more so than Whites (Harris-Lacewell & Junn, 2007). We therefore expect that
Blacks will project their own ideological positions onto Black candidates.
Due to their distinctive historical position and continued stigmatization, Black voters are distinct
even from other minority groups in the extent of their racial group cohesion.6 For instance, unlike
Latinos, who appear to be advancing toward middle-class status through traditional markers such as
homeownership and higher education (Myers, 1999; Park & Myers, 2010), Blacks are witnessing a
generational decline in economic status (Isaacs, 2007). Given that political projection is a function
of perceived similarity based on social categorization, we expect that Blacks will project to a greater
extent than members of other racial minority groups.
Additionally, we expect the extent of projection for Black voters will vary by respondent
ideology. Black conservatives are more likely than liberal Blacks to experience a conflict between the
desire for descriptive representation and their ideological position, given that on most (though not
all) issues Black legislators are more liberal than their White counterparts (Bratton & Haynie, 1999;
Conyers & Wallace, 1976; Herring, 1990; Nelson, 1991). Projection is one way that these citizens
can avoid potential cognitive dissonance between their candidate preference and their ideological
preference (Conover & Feldman, 1982; Festinger, 1957). Therefore, we expect that projection will be
greater among Black conservatives than Black liberals.
Finally, we expect that ideological projection will matter for vote choice. The perceived con-
gruence between one’s own ideology and a candidate’s ideology has long been considered a key
predictor in models of electoral choice (see Bartels, 2010 for a brief review). Thus, we expect that
projection will increase perceptions of ideological congruence between Black voters and Black
candidates y which will in turn correlate positively and significantly with vote intention.1
Data and Methods
We conducted two identical survey experiments in January of 2009 and October of 201 1
from the two experiments were the same, so we have combined the two cohorts in the anal
follow.8 We recruited a group of participants (N = 1,580 in the first cohort and N = 1,61
second) through a web panel maintained by Survey Sampling International (SSI). The pane
randomly selected, but both Black and White samples in each experimental cohort are co
to the national Black and White citizenry on all observable demographics, including gend
education, partisan identification, and southern residence.9 We oversampled Black respon
both surveys (N = 357 in the first cohort and N = 569 in the second), and we analyze the sa
White respondents and Black respondents separately throughout the article.
6 For more on this, see Akbar (1989), Dawson (1995), Dubois (1903/1969), Sears and Savalei (2006), and Smith
7 While perceived ideological congruence may predict the likelihood of voting for a candidate, the decision t
candidate may also influence the likelihood and extent of projection. Our data do not allow us to determine causal
8 Results are not sensitive to the introduction of a dummy for year.
9 Given that ours is a nonprobability sample, however, we would encourage caution in generalizing these results. In
the Black sample is slightly younger and more educated than a random draw from the US Black populatio
demographics relative to the total national and national Black populations are provided in the online appendix.
This content downloaded from 132.174.251.96 on Fri, 10 Apr 2020 23:00:22 UTC
All use subject to https://about.jstor.org/terms
Stereotyping
or
Respondents
Projection?
were
151
randomly
ass
tions, subjects were presented wit
candidate from Pennsylvania and
nying
the candidate’s biography
they saw a Black candidat
respondents in the control condit
the one-candidate design assures t
candidates to one another, allowin
of successful covariate balance acr
respondents considered separately
Respondents were then asked w
another
feeling
variety
as
thermometer,
of
and
demographic
Independents,
and
to
p
questions
34%
ident
population, the overwhelming ma
Democrats.13 However, Black sur
dencing significant ideological di
(between 1 and 4 inclusive on a 1
as political moderates (5 or 6), an
to
10
We
in
inclusive).
conducted
this
a
third
iteration
candidate’s
we
position
experimen
also
conservative
we
questions,
on
asked
variet
and welfare; social issues like gay
control and affirmative action.14
Finally,
on
a
continuum
asked
as
on
the
respondents
well
as
on
to
“econom
Our survey experiments enable
validity. However, as with many
the treatment to the “real world
White voters. To confirm that our
post-election
interviews
in
the
Cooperative Congressional Electio
CCES).15 Both datasets were chos
ANES and N = 3,693 for the CCES
with
data
from
the
Congressio
which includes detailed biographi
for each legislator in the 110th C
these descriptive data do not allo
10
Details
of
the
web
page
are
provided
11 Facial images were created using
evaluations of facial features and a
in
Face
face
Oosterhof, 2008), we ensure that each ph
survey, we confirm that a large majority
12 Demographic balance checks are provid
13 In the 2008 ANES, about 87% of Black
The order of issue questions was random
15 While the ANES uses a probability-base
16 Legislators in the CLASS data were mat
This content downloaded from 132.174.251.96 on Fri, 10 Apr 2020 23:00:22 UTC
All use subject to https://about.jstor.org/terms
152
Lerman
and
between
Sadin
Black
member’s
positions
and
NOMIN
of
each
c
Experiment
Our
survey
exper
stereotype
a
Black
effect
is
limited.18
scale
(coded
from
0
at
.40
on
the
same
a
one-way
ANOVA
vote
for
the
Black
in
the
other
condi
logical
subgroups
f
In
contrast
to
Wh
the
Black
candidat
difference
the
of
Black
nine
candida
percentage
points
to
Whites
candidates.
As
we
onto
in-group
cand
“cancel
out”
in
the
contrast
When
we
disaggr
hypotheses
percentage
repeats
(see
Fig
points,
when
Differences
exam
for
Bla
percent
of
Black
l
who
perceive
this
t
the
reverse
is
true
candidate
to
be
con
We
also
estimate
squares
(OLS)
regre
treatment
assignm
significant
(p< .01 treatment indicato the Black candidat didate (p < .05). Wh 17 NOMINATE scores a legislator's first-dimen economic matters. The (McCarty, Poole, & Ro 18 Charlie Johnson was place the candidate (irr In fact, we find some e this is due to the candi only to the effect of u allegiance to the This content downloaded from 132.174.251.96 on Fri, 10 Apr 2020 23:00:22 UTC All use subject to https://about.jstor.org/terms Demo Stereotyping Condi,ion n n O O or Projection? 153 Condition u* * control white u* ~ O ~ white I M ' T L i : & ! T 0 6" 1 ! ' 1 1 Tlil I ?§ °! 6" ® ! !t! I í i T a T! 'fi 1 /I ! Ç ¡ T i ! 22 ® i ii ! i ! «o + t 1 * T t1 a 11 ! /I 1 ¡ !+ i ¡ ! re ? 0 >
o
I
+
S
i !
i iiT
TT
t t
!i! !
1??-4?~
i 1
11
!
“O
i
i
¡ T
¡T
tt
C
Ó
T
?
¡
¡
!i I
5 tí? ! T 1 i 1 i T i i i tí? I 1
?
i
;
i
:
i
1
i
+
1
T ! f ? T ¿ T ? f 1
• t 1 ¡ ? i ; I i 0 20i >L 1 ! i I 1
i
I I I ^ ? I ? 0.00-! I ? I I I ?
All Whites I White Conservatives I Black Liberals I All Whites I White Conservatives I Black Liberals I
White Liberals All Blacks Black Conservatives White Liberals All Blacks Black Conservatives
Condition
1.00- #black
* control
O white
0.80-
!
! i T ? ‘ ? ; I 1 ? i
i06” 1 1 M’ 1 1 ? ti
?
Ý
.E
è
Ö)
T
®
®
î
T
–
1
T
ò
i
:
J-
1
A
lì!
0.40-
i
0.20-
All Whites I White Conservatives I Black Liberals I
°.°(H
White Liberals All Blacks Black Conservatives
Figure
3 on a
1. Effect of candidate race, by res
1-10 scale; conservative is 8, 9 or
Blacks N = 927; Black Liberals N = 217;
1 (most conservative). Source: original ex
perceive the Black candidate as 13 percentage points more con
(p < .Ol).20 Data from our third experiment further allow us to examin sionality. Do Black and White liberals and cons... Purchase answer to see full attachment