UCSD Role Played by Gender Stereotypes in Candidate Evaluations Research Paper In approx. 5-6 pages (excluding reference page) write an answer to the resea

UCSD Role Played by Gender Stereotypes in Candidate Evaluations Research Paper In approx. 5-6 pages (excluding reference page) write an answer to the research question which is How do gender stereotypes play a role in candidate evaluations? using the sources provided. Double-spaced font 12-point Times New Roman.I will attach 6 sources that you can use writing this research paper, also I will need a paper outline and I will give an outline example to show you what is required and what you should write I will be needing this first. Paper Outline Requirements
Due Friday, July 17th
by 11:59pm on iLearn
Note: This outline does not need to be more than 2 pages. You can use bullet points.
(Note: For your FINAL PAPER each question will need to be explained in more detail in
paragraph/essay format.)
● Introduction Outline
○ What is the research question?
○ What do you think the answer to the research question is and why? (thesis
statement)
○ Why is this question important when we consider American politics?
● Body Paragraph Outline (repeat for all sources/articles)
○ Topic of Source 1
○ Argument of Source 1
○ Method of Source 1 (what kind of data did they use? Did they do
interviews or experiments or surveys?)
○ What were the limitations of this source?
○ How does this source support or not support your argument?
● Conclusion Outline (answer one of these questions in your outline, answer all
questions in your final paper)
○ How do these sources/articles you have reviewed contribute to the
question being asked?
○ Was there anything you found particularly interesting or surprising about
the results of these studies?
○ What other kinds of research or questions could be useful for future work
in this area?
bs_bs_banner
Political Psychology, Vol. xx, No. xx, 2014
Political Psychology, Vol. 36, No. 6, 2015
doi: 10.1111/pops.12186
doi: 10.1111/pops.12186
Emotional, Sensitive, and Unfit for Office? Gender Stereotype
Activation and Support Female Candidates
Nichole M. Bauer
Indiana University
Women are underrepresented at all levels of elected office. It is suspected that gender stereotypes hinder the
electoral success of female candidates, but empirical evidence is inconclusive on whether stereotypes have a
direct effect on voting decisions. This empirical conflict stems, in part, from the assumption that voters
automatically rely on gender stereotypes when evaluating female candidates. This study explicitly tests the
assumption of automatic stereotype activation. I suggest that stereotype reliance depends on whether stereotypes have been activated during a campaign, and it is only when stereotypes are activated that they influence
evaluations of female candidates. These hypotheses are tested with a survey experiment and observational
analysis. The results show that campaign communication activates stereotypes when they otherwise might not
be activated, thereby diminishing support for female candidates.
KEY WORDS: female candidates, gender, stereotypes, and representation
Female candidates performed exceptionally well in the 2012 U.S. elections. Record numbers of
women ran for Congress, and 98 women won a seat to the House or Senate (CAWP, 2012). Despite
these successes, women remain underrepresented at all levels of elected office in the United States.
While female candidates do win elections and raise money at roughly the same rate as male
candidates (Burrell, 1994; Darcy & Schramm, 1977; Fox, 2009; Seltzer, Newman, & Leighton,
1997), there is ambiguity about the challenges female candidates face among voters. As Anzia and
Berry (2011) point out, comparisons of vote shares do not provide a complete picture of the
perceptual obstacles female candidates face when running for office.
Stereotypes about women are thought to pose a difficult perceptual constraint (Alexander &
Anderson, 1993; Huddy & Terkildsen, 1993a). Feminine stereotypes characterize women as warm,
nurturing, and sensitive, and these qualities contrast with the expectation that political candidates be
outspoken, decisive, and aggressive (Eagly & Karau, 2002; Huddy & Terkildsen, 1993b). Empirical
evidence documenting the effects of feminine stereotypes on candidate evaluations offers conflicting
conclusions. Some studies find that stereotypes harm female candidates (Fox & Smith, 1998; Leeper,
1991; Rosenwasser & Seale, 1988); others show they help female candidates (Dolan, 2010; Fridkin
& Kenney, 2009); and still yet, another body of research illustrates that stereotypes have little effect
on evaluations of female candidates (Brooks, 2013; Dolan, 2014; Huddy & Capelos, 2002). Underlying much of this research is an assumption that voters evaluate female candidates through the lens
of feminine stereotypes; however, these conflicting results suggest that stereotypes might not always
motivate voter perceptions.
1
691
0162-895X © 2014 International Society of Political Psychology
C 2014 International Society of Political Psychology
V
0162-895X
Published by Wiley Periodicals, Inc., 350 Main Street, Malden,
MA 02148,
USA, 9600 Garsington Road, Oxford, OX4 2DQ,
Published by Wiley Periodicals, Inc., 350 Main Street, Malden, MA 02148,
9600Carlton
Garsington
Road,
2DQ,
and POUSA,
Box 378
South,
3053Oxford,
Victoria,OX4
Australia
and PO Box 378 Carlton South, 3053 Victoria, Australia
692
2
Bauer
This study tests how and when voters rely on feminine stereotypes. I start with the premise that
female candidates are not doomed to suffer from the constraints of stereotypes. Stereotypes will only
influence candidate evaluations under a narrow set of conditions, and this leads to the research
question: Under what conditions are feminine stereotypes activated when voters evaluate female
candidates? In addition, when are feminine stereotypes likely to influence support for female
candidates? I integrate research from social psychology and political science to develop a theory of
stereotype activation and application that specifies the conditions under which individuals rely on
feminine stereotypes. This article presents two empirical tests: first, an experiment tests the stereotype activation and application process; and second, an observational analysis shows how exposure
to stereotypic campaign ads reduces support for female candidates.
The Assumed Influence of Feminine Stereotypes
Feminine stereotypes reflect beliefs about the characteristics women are expected to embody,
and they include being “affectionate, helpful, kind, sympathetic, interpersonally sensitive, nurturant,
and gentle” (Eagly & Karau, 2002). These stereotypes align with communal roles, such as being a
caregiver or homemaker, that women are seen as better prepared to fulfill compared to men.
Contrasting communal roles are agentic roles, such as being an economic provider or a leader, and
traits such as aggressiveness and dominance define these roles (Koenig, Eagly, Mitchell, & Ristikari,
2011). Men are perceived as best fit for agentic roles. The alignment between communality with
women and agency with men creates the perception of women as unfit for roles requiring agentic
traits (Simon & Hoyt, 2008).
The incongruity between leadership traits and feminine stereotypes leads to the expectation that
feminine stereotypes disadvantage female candidates; however, empirical research offers conflicting
conclusions. Feminine stereotypes can reduce support for female candidates at higher levels of
office, such as for Congress or the Presidency, because voters desire candidates with agentic traits at
this level of office (Huddy & Capelos, 2002; Lawless, 2004). In this context, voters use stereotypes
to infer women are less qualified for office (Huddy & Terkildsen, 1993b; Leeper, 1991). Under other
conditions, feminine stereotypes bolster the electoral success of female candidates (Fridkin &
Kenney, 2009; Sanbonmatsu, 2002). In state and local elections, stereotypes can help female
candidates because there is greater alignment between stereotypes and the major issues at this level
of office (Kahn, 1994). For example, education is a critical state issue, and stereotypes can lead
voters to perceive women as strong on education issues. Another body of research finds that
stereotypes have little effect on vote choice because female candidates engage in strategies designed
to downplay stereotypes (Bos, 2011; Gordon, Shafie, & Crigler, 2003).
Conventional wisdom presumes an inextricable link between a female candidate’s gender and
feminine stereotypes (Huddy & Capelos, 2002), but this connection is not always tested. If the link
between female candidates and feminine stereotypes is intrinsic, then voters should have baseline
expectations for female candidates that align with these stereotypes. Brooks (2011) investigated the
extent to which voters have expectations about female-candidate emotionality and found that people
do “hold strongly gendered stereotypes about the emotionality of ordinary people, but they do not
seem to hold them about political candidates” (p. 603). Schneider and Bos (2013) found that voters
do not have expectations for female candidates that align with traditional norms of femininity. Taken
together, these findings suggest the use of feminine stereotypes in electoral evaluations is complex
and conditional.
I build on these recent studies to clarify the role of feminine stereotypes in how voters perceive and
evaluate female candidates. I argue that stereotype reliance is not automatic; rather, stereotypes require
activation before voters use them to evaluate female candidates. The next section outlines a theory of
stereotype activation and application from which I derive a set of hypotheses. An experiment then tests
Emotional,
Emotional, Sensitive,
Sensitive, and
and Unfit
Unfit for
for Office?
6933
the assumption that a female candidate’s gender triggers feminine stereotype reliance and identifies the
process of stereotype activation. Following the experiment is an observational analysis to show how
feminine stereotypes emerge in campaign advertising and how they reduce vote support for female
candidates.
Stereotype Activation and Application
I rely on a theory of stereotype activation and application to clarify when and how voters turn
to feminine stereotypes to evaluate female candidates. That feminine stereotypes exist as part of
general cultural knowledge is clear, but these stereotypes do not always shape how individuals
perceive female candidates. Kunda and Spencer (2003) explain the process of stereotype reliance,
writing: “stereotype activation is the extent to which a stereotype is accessible in one’s mind, and
stereotype application is the extent to which one uses a stereotype to judge a member of a stereotyped group” (p. 522). For most individuals, activating stereotypes requires more than encountering
a member of a stereotyped group (Blair, 2002; Blair & Banaji, 1996; Devine, 1989).1 Activation
follows exposure to stereotypic information that leads individuals to rely on broader stereotypes
about the appropriate roles and behaviors of women to form impressions of female candidates. Blair
and Banaji (1996) experimentally tested whether gender, alone, triggers stereotypes and found
stereotype activation did not occur. Stereotype activation only happened when individuals received
information about a woman that reflected traditional norms of femininity.
Translated to the electoral arena, stereotype activation and application depends on the type of
information voters have about a candidate (Kunda & Sinclair, 1999; Lepore & Brown, 1997). When
voters have minimal information, such as partisanship, stereotypes will not be activated. This
prediction does counter prior work finding that individuals use stereotypes to make ideological and
issue-based inferences when they have little information (Banducci, Karp, Thrasher, & Rallings,
2008; Koch, 2000; McDermott, 1997, 1998). Voters may make these inferences, but feminine
stereotypes will not affect how voters evaluate female candidates. The first hypothesis describes how
feminine stereotypes are not activated under baseline conditions and identifies the conditions that
lead to feminine stereotype activation:
H1: When voters have only the partisan cue for a female candidate, feminine stereotypes
will not be activated compared to when individuals have information that reflects feminine
stereotypes.
During a campaign, stereotype activation is likely to occur through communication channels, such as
television advertising, news media coverage, or speeches connecting female candidates with stereotypic traits (Heldman, Carroll, & Olson, 2005; Kahn, 1994). Describing female candidates as caring
or nurturing might trigger a range of other associations about women as mothers or caregivers and
are subtle ways stereotypes become more salient (Kunda, Sinclair, & Griffin, 1997; Prentice &
Carranza, 2002). Activating feminine stereotypes for female candidates may lead voters to make
other stereotype-based inferences that female candidates may not have the necessary traits for being
an effective leader. Character traits play an important role in determining a candidate’s qualifications
for political office (Kinder & Fiske, 1986; McGraw, 2003; Miller, Wattenberg, & Malanchuk, 1986;
Pierce, 1993). Hypothesis 2 describes how activated stereotypes will affect candidate evaluations:
1
Future work should turn to identifying individual characteristics that lead some individuals to more readily experience
stereotype activation and application. For example, measures such as the Bem Sex Role inventory can demonstrate how
gender-role perceptions affect stereotype activation (Bem, 1981).
694
4
Bauer
Table 1. Experimental Design
Wave 1
Wave 2
Democrat Male and Democrat Female
Republican Male and Republican Female
Democrat
Stereotyped Male
Stereotyped Female
Neutral Male
Neutral Female
Republican
Stereotyped Male
Stereotyped Female
Neutral Male
Neutral Female
H2: Activated feminine stereotypes will be applied to female candidates and affect how
voters perceive the qualifications and abilities of female candidates.
Activated stereotypes exacerbate the incongruity between communality and agency, leading individuals to perceive women are poor leaders. This incongruity will diminish support for female
candidates, especially at higher levels of office. Without stereotype activation, individuals will
evaluate female candidates according to politically relevant cues such as candidate partisanship.
An Experimental Test of Stereotype Activation and Application
I start with an experiment to test the connection between female candidates and feminine
stereotypes. An experiment is particularly beneficial because the method permits “stronger causal
inferences by allowing researchers to rule out potential confounds through tight control over conditions and random assignment of participants to exposure” (Brader, 2005, p. 391).
Experimental Design
The experiment took place in two waves and implemented a 2 × 2 factorial design. Wave 1
occurred before the experimental manipulation. In the first wave, participants evaluated a female
candidate, Susan Foster, and a male candidate, Tom Larson.2 In waves 1 and 2, the study matched
participants to candidates with whom they shared partisanship—for Democratic participants, Susan
and Tom were Democrats, and for Republican participants, the candidates were Republicans.
Wave 2 took place one week after the first wave and manipulated stereotypes. The 2 × 2 factorial
design manipulated candidate gender and feminine stereotypes or no stereotypes.3 Participants read
a news article about one of the two candidates from the first wave, either Susan Foster or Tom Larson.
As in wave 1, participants evaluated a candidate with whom they shared partisanship. Including
partisan labels gives a more conservative estimate of how participants evaluate candidates they
already view favorably through a partisan lens and limits the inferences participants might otherwise
make about partisanship. Table 1 presents the full design.
The two-wave design offers a number of advantages. First, this design allows the measurement
of baseline stereotype activation under low-information conditions. The design of wave 1 is not
dissimilar from an actual voting situation when voters often know only the partisanship and names
of candidates. Stereotype activation is measured prior to the main study, and the one-week time lag
between the two waves limits the potential that stereotypes might be primed in the first wave and
2
3
The photos and names were pretested using an undergraduate sample from a large midwestern university (N = 120). Both
candidates were rated between the ages of 45 and 55 (p = 0.1811). The candidates were rated comparably on attractiveness
(p = 0.6338) and on competency (p = 0.34).
This study does not manipulate masculine stereotypes. Research on masculine strategies for female leaders is mixed. Some
studies find that female candidates who evoke masculine stereotypes limit reliance on feminine stereotypes (Brooks, 2013),
and other studies find that female candidates or female business leaders can face a backlash for being counterstereotypic
(Krupnikov & Bauer, 2014; Rudman & Glick, 1999).
Emotional,
Emotional, Sensitive,
Sensitive, and
and Unfit
Unfit for
for Office?
6955
Table 2. MTurk Sample Characteristics Compared to Other Samples
% Women
% Region****
Midwest
Northeast
South
West
% Democrats
% White
% Making $60,000+
% Approve Obama’s Performance
Age
18–24
25–44
45–64
65+
Experiment
Sample
Berinsky et al.
sample
2011
PEW
2010
CCES
2008 ANES
(Internet)
2010
Census*
46.15%
60.1%
55.74%
52.76%
57.6%
50.8%
22.68%
28.43%
27.80%
21.09%
60.89%
80.25%
38.26%
52.55%
26.8%
22.4%
30.4%
20.4%
***
83.5%
**
**
25.28%
18.76%
38.32%
17.63%
31.97%
80.87%
44.51%
45.41%
22.80%
16.19%
36.30%
24.80%
35.71%
75.13%
53.51%
42.40%
28.3%
16.9%
31.4%
23.4%
35.71%
84.3%
21.67%
17.9%
37.10%
23.3%
**
74.83%
42.8%
**
27%
52%
19.5%
1.5%
**
**
**
**
10.25%
23.41%
39.01%
27.33%
3.4%
23%
50.9%
22.7%
**
18–29
30–44
45–59
60+
9.4%
24.8%
36.4%
29.5%
13.08%
35.01%
34.74%
17.17%
Note. *2010 Census data percentages are based on the total U.S. population, not just the adult population. **Does not ask
this question; data presented in a different manner; survey does not ask appropriate question ***On partisanship, Berinsky
et al. (2012) provide the average on a 7-point scale, rather than a percentage. ****Based on U.S. Census Designations.
influence evaluations in the second wave (Hutchings and Piston 2011). A two-wave design traces the
stereotype-activation process by measuring before and after changes in stereotype activation.
The stimulus manipulated stereotypes through trait associations (see the appendix). For
example, Susan Foster and Tom Larson were described as “nurturing” and “compassionate” in the
feminine stereotype-activation condition. The stimulus recounted a campaign event the candidate
attended with his or her children at an elementary school where he or she met with parents. The
nonstereotype group received the same article with no stereotypic traits or mention of the candidate’s
family and gender-neutral terms used to describe the candidate.
I used traits to activate stereotypes rather than issues for several reasons. First, research in social
psychology shows that traits lead to stereotype activation for women (Eagly & Karau, 2002; Prentice
& Carranza, 2002). Additionally, personality traits play a key role in candidate evaluations (Funk,
1999). I refrain from using issues because issues overlap with partisanship. The manipulation was
embedded in a newspaper article. In a “real world” campaign setting, stereotype activation is likely
to occur through campaign communication such as newspaper articles or campaign ads. Additionally, studies show that newspaper coverage of personality traits shapes voter perceptions of candidates (Fridkin & Kenney, 2011).
I recruited participants following the method of Berinsky, Huber, and Lenz (2012) using
Amazon’s Mechanical Turk.4 Table 2 compares the demographic characteristics of the MTurk
sample to Internet-based opt-in samples, nationally representative samples, and the U.S. census.
While the sample skews White, young, and liberal, it is comparable to other MTurk studies and
Internet-based samples such as the CCES. A total of N = 302 completed both wave…
Purchase answer to see full
attachment