Women are underrepresented in the United States government despite accounting for approximately half of the U.S. electorate. Therefore, investigation into the types of contexts in which female candidates may emerge may improve gender parity. This study seeks to investigate preferences for female candidates in times of threat, as evidence from the glass cliff literature indicates that female leaders may be preferred to male leaders in times of crisis, that this preference may be driven, in part, by system justifying ideologies (Brown, Diekman, & Schneider, 2011), and that the type of threat may influence preferences for female leaders. Therefore I propose to investigate preferences for female political leaders as a function of the membership of the group posing the threat and whether the leader is described as upholding cultural traditions. University students will read vignettes about a terrorist attack perpetrated by ingroup member (i.e., White supremacists) or outgruop members (ISIS) and then view responses from male and female presidential primary contenders who either voice support for cultural traditions or do not mention them. Participants then rate their perceptions of the candidates’ warmth and competence and indicate how likely they would be to vote for the candidate in the primary and how electable they believe the candidate to be in a general election.
In the 2016 U.S. presidential election, Donald Trump beat presumed frontrunner, Hillary Clinton, after a bruising political campaign. In her concession speech, Clinton directly addressed those who had hoped to see a female president in their lifetime, stating, “I know we have still not shattered that highest and hardest glass ceiling, but some day someone will and hopefully sooner than we might think right now” (Clinton, 2016). Certainly, there are many potential reasons why Clinton’s campaign was unsuccessful. Some have blamed an overemphasis on identity politics (Lilla, 2016), whereas others see Trump’s win as a repudiation of the policies of the Bush and Obama administrations (Guo, 2016). However, some, including Clinton herself (e.g., Suliman, 2017) have attributed her loss at least partially to sexism, prompting some to wonder more generally under what conditions a female political leader might be desirable.
The
purpose of the proposed research is to investigate conditions under which a
female U.S. presidential candidate would be preferred to a male U.S
presidential candidate among American college students. One condition in which a
female head of state might be desirable is when the nation is under threat, as
evidence suggests women are more likely to be nominated to precarious
leadership positions when organization is in crisis (Ryan & Haslam, 2005)
and that this phenomenon may be motivated by system justifying ideologies
(Brown, Diekman, & Schneider, 2011). It also seems likely that this effect
likely also depends on the type of threat, as some studies (e.g., Bruckmüller & Branscombe, 2010) show a
diminished preference for a male leader, whereas others (e.g, Brown et al., 2011) show a pronounced preference for a
female leader. I therefore investigate
preferences for male and female candidates as a function of system
justification (i.e., whether the candidate is described as endorsing cultural tradiitons or not), and
the type of threat, that is whether the threat comes from members of the
ingroup (i.e., domestic terrorism) or members of the outgroup (i.e.,
international terrorism).
Social Role Theory.In
order to understand the machanisms that might motivate preferences for female
leaders, it is necessary to consider how gender roles and stereotypes affect perceptions
of women who work, especially in stereotypically masculine professions. Social
role theory (Eagly, Wood, & Diekman, 2000) suggests that women and men are
ascribed traits that correspond toon their representation in different
occupational roles. To the extent that women are seen in caretaking roles
(e.g., occupations such as teachers, nurses, homemakers), they are ascribed communal traits such as being warm,
kind, and nurturing. Similarly, to the extent that men are seen in breadwinning
roles (e.g., occupations such as business leaders, doctors, lawyers), they are
ascribed agentic traits, such as
being dynamic, assertive, and aggressive. Thus social role theory encompasses descriptive gender role content, that
is, traits that are typical of men and women
Several
studies indicate support for this theory, For example, in a series of studies,
Diekman and Eagly (2000) demonstrated that perceptions of the traits that men
and women have vary as a function of occupational role. For example, women were
perceived as having more agentic traits over time, consistent with women’s
increasing participation in the labor force since 1950. There is also evidence
suggesting that social role theory may apply more generally to perceptions of
different groups (e.g., ethnic, racial, and religious groups). Koenig and Eagly
(2014) found that perceptions of group traits for different racial, ethnic and
religious groups varied as a function of their occupational roles. In other words,
regardless of whether the target group is gender or another group (e.g.,
religious groups), representation in occupational roles appears to impact
perceptions of the types of traits that people in that group may typically
posess.
Role Congruity Theory. Of course, judgments
of group members not only depend on individuals’ perceptions of traits that
groups are likely to have (i.e.,
descriptive stereotypes) but also judgments of the types of traits that group
members should have (i.e.,
prescriptive stereotypes). Therefore, social role theory was augmented by role
congruity theory (Eagly & Karau, 2002), which describes the prescriptive content of gender roles.
According to role congruity theory, individuals whose actions do not align with
their prescribed gender roles are be judged harshly or punished for role
incongruity, whereas those who behave in ways that are consistent with gender
roles are rewared.
Indeed,
a great deal of evidence supports the propositions made in role congruity
theory. For example, judgments of male and female job applicants who were
described as either self promoting (i.e., displaying agentic behaviors) or
self-effacing (i.e., displaying communal behaviors) indicated that self-promoting
candidates were generally preferred to those who were self-effacing. However,
female candidates who exhibited self-promoting behaviors were judged as having
fewer social skills and as being less hireable than their male colleagues
(Rudman, 1998), suggesting that women may be punished for role-incongruent
behavior, even if that behavior is generally viewed positively.
Evidence
also suggests that conformity to gender roles is rewarded. For exampole,
evidence suggests that young children are rewarded for conforming to gender
roles (Bussey & Bandura, 1999) and that among individuals who placed
importance on gender roles, gender role-congruent behavior was associated with
greater self-esteem and smaller discrepancies between ought and ideal selves
(Wood, Chistensen, Hebl, & Rothgerber, 1997). Therefore, it appears that self-regulating
mechanisms and societal mechanisms reward individuals for engaging in
role-congruent behaviors.
Female Leadership Disadvantage. Role congruity
theory makes two specific predictions about female leaders. In particular, it describes
a “double-bind” for female leaders (Eagly & Karau, 2002). The first part of
this bind describes how women may be passed over for promotion to leadership
positions because descriptive theories of gender roles do not align with
perceptions of leaders. That is, individuals may assume that because women are
less commonly found in leadership positions, they are less qualified for them
than are men. The second part of the bind describes the experiences of women
already in leadership positions. As leadership may require displays of agency,
women in leadership may be judged harshly for violation of prescriptive gender norms.
Therefore female leaders may be regarded as cold, cutthroat or manipulative.
Indeed,
research on gender and leadership generally supports the double bind described
by role congruity theory. For example, meta-analytic evidence suggests that
stereotypes of men align more closely with stereotypes of leaders than do
stereotyeps of women across three different paradigms measuring aspects of
agency and communion (Koenig, Eagly, Mitchell, & Ristikari, 2011). Similar
research on implicit leadership theories (ILTs) indicates that protytypes of
successful leaders are closer to prototypes of men than prototypes of women (Hall, Workman, &
Marchioro, 1998) and that leadership prototypes are more descriptive of
masculine than feminine traits (Schein, 1973).
Furthermore, there is
evidence that man and women may be held to different standards in leadership
positions. For example, meta-analytic evidence from 61 studies suggests that
perceptions of leaders are more positive when the leader is male rather than
female, particularly when judged by men (vs. women) and when the leader was
described as displaying agentic qualities, such as being directive (Eagly,
Makhijani, & Klonsky, 1992). There is also evidence suggesting that leader
behaviors may be judged differently depending on the gender of the leader. For
example, leaders exhibiting both masculine and feminine traits were judged more
positively than those exhibiting only masculine or feminine qualities. However,
this was only the case if the leader was described as a man instead of a woman
(Wolfram & Gratton, 2013). Similarly, Kark, Waismel-Manor, and Shamir
(2012) found that transformational leadership styles were associated with
masculine and feminine qualities. However, women were judged more harshly for
failing to embody these qualities than were men. Thus, there is evidence that
female leaders may be held to a higher standard than are male leaders.
Political Leaders.
Research
on female politicians, in particular, similarly suggests that women may be disadvantaged in leadership positions.
For example, in two studes, one experimental and one using data from incumbant
elections from the 2000 and 2004 congressional elections, Bauer (2015)
demonstrated that the presensence of feminine stereotypes in political ads was
associated with lower ratings of female (vs. male) leaders’ potential effectivenss
as a senator and fitness as a presidential candidate. She also found that the
presence of feminie stereotypes in ads was associated with a decreased likelihood
of voting for a female versus male candidate. Other work similarly suggests
that female leaders may be perceived as unqualified for political offices
because political offices (e.g., president, senator) may not be associated with
characteristics typically ascribed to women (Schneider & Bos, 2014).
Additionally,
theer is some evidence that both parts of the double-bind described by role
congruity theory may affect perceptions of female political candidates. For
example, Gervais and Hillard (2008) found in a study on perceptions of Sarah
Palin and Hillary Clinton that Clinton was perceived as less feminine and warm
but more competent than was Palin, suggesting that perceptions of both
candidates may have, in part, been driven by perception of how closely each
candidate’s behavior aligned with stereotypically feminie gender roles. Additionally,
Gervais and Hillard found that the probabiliby of voting for Palin over Clinton
was much stronger among participants who strongly endorsed gender role
traditionalism, suggesting that participants’ voting behavior may, in part, be
motivated by whether female candidates’ behaviors align with prescribed gender
roles. Therefore, it is possible that female political candidates who display
agency may be viewed as competent but unlikeable, whereas those who display
communion may be well-liked by risk being viewed as incompetent.
Female
Leadership Advantage. However,
despite definite disadvantages facing female leaders, there are some instances
in which female candidates may be advantaged relative to their male colleagues.
For example, some evidence suggests that
because women face more barriers to leadership roles than do men, female
leaders may be perceived as more qualified than male leaders in the same
position (Foschi, 2000). Indeed, Rosette and Trost (2010) found that
participants judged female leaders who displayed agentic behaviors as more
effective than male leaders in top management positions. However this effect
only emerged when female leaders’ success could be internally attributed.Thus, under some circumstances
individuals may view female leaders as being more successful than male leaders.
Other evidence suggests
that preferences for female leaders may emerge in leadership positions that
complement feminine stereotypes. For example, meta-analytic evidence suggests
that female leaders may be preferred in industries such as education and in
positions such as middle management, which may require more communication
skills than upper-management positions (Paustian-Underdahl, Walker, &
Woeher, 2014). Similarly, there is also evidence that a female leadership
advantage might exist because women are more likely than men to engage
leadership behaviors such as transformational (Vinkenburg, van Engen, Eagly,
& Johannesen-Schmidt, 2011) and democratic leadership (Eagly & Carli,
2003). Therefore, it is possible that despite negative perceptions of women in
leadership positions, there may be some instances in which women are more
likely to excel in leadership positions than are men.
Political Leaders. Research on female
political leaders similarly suggests that female
politicans may benefit from being women in some specific circumstances. For
example, female politicians were preferred when issues such as education and
health care were important to participants (Dolan, 2010). Similarly, evidence
suggests that women may be preferred at lower levels of government, where issues
that are more consistent with feminine steroetypes (e.g., education) are more
salient (Kahn, 1994). Other evidence indicates that female leaders may be less
suceptible to negative campaign ads than are male candiates (Fridkin, Kenny,
& Woodall, 2008),suggesting that despite the fact that political offices
are ssociated with agency, there are some situations in which female
politicians may be preferred to male politicians.
There is also another
context in which women might be preferred to men. Some evidence that there is a
selection bias that favors women over men when leadership positions are
precarious (e.g., when the organization is in crisis), known as the “glass
cliff phenomenon.” Ryan and Haslam (2005) coined the “glass cliff” after
demonstrating that historically, women were disproportionately nominated to
organizational boards after dips in the organization’s performance and stock
prices. Evidence for the glass cliff has been demonstrated in both archival
(e.g., Ryan & Haslam) and experimental data (e.g., Brown et al., 2011). It is important to note that the glass cliff
effect is generally considered a phenomenon rather than a theory (Ryan, Haslam,
Morgentroth, Rink, Stoker, & Peters, 2016). That is, the class cliff is a
phenomenon is an effect that can be observed, which can be explained through
other theories such as role congruity theory.
This may be one reason
that mixed evidence has been found for its existence across studies. For
example, although numerous studies have demonstrated the glass cliff phenomenon
(e.g., Ryan & Haslam, 2005; Haslam & Ryan, 2008; Ryan, Haslam, Hersby,
& Bongiorno, 2011), others studies have not. For example, in an examination
of U.S. Fortune 500 there was not evidence of a gender difference in CEO
appointments when examining objective accountancy-based measures of company
performance (Adams, Gupta, & Leeth, 2009). Other studies have similarly
failed to find evidence of the glass cliff phenomenon among and
Canadian samples (Carroll, Hennessey, & MacDonald 2013). One potential
explanation for this discrepancy is the way in which performance is measured.
For example, accountancy-based measures were unrelated to preference for a male
or female leader. However, when performance was conceptualized in terms of stock-based
performance, female leaders were preferred to male leader (Haslam, Ryan,
Kulich, Trojanowski, & Atkins, 2010). Thus, it is possible that preferences
for female leaders when an organization is in crisis may have more to do with perceived performance than actual performance.
The Glass Cliff and
Political Leaders. In
any case, one area of glass cliff research that has received relatively little
attention is the role of the glass cliff effect in the election of female
leaders. Using archival data from the 2005 UK general election, Ryan, Haslam,
and Kulich (2010) demonstrated that women ran for less winnable seats than did
men, and this was particularly pronounced for the minority party (i.e., for
Conservative women) than it was for the majority party (i.e., Labour). Thus,
there appears to be at least some archival evidence suggesting that female
political candidates more often run for precarious seats than do male political
candidates. Some evidence of the glass cliff also exists in experimental
studies. Ryan et al. (2010) found that in a laboratory study using British
political science students, participants showed a preference for a female
candidate when a seat was described as difficult to win but a preference for a
male candidate when the seat was described as winnable.
One potential criticism
of Ryan et al.’s study is that although they demonstrated that there is a
preference for female political leaders when the seat is precarious, that does
not necessarily mean that there is a preference for female political leaders
when the country is in crisis (as might be the case for a precarious leadership
position in an organization). Therefore, it is possible that individuals may
prefer women for precarious leadership seats, but not when the nation is under
threat. Indeed, some evidence suggests that male political candidates are
preferred to women on issues of war and national security (e.g., Huddy &
Terkildsen, 1993). However, other evidence suggests that there is an increased
preference for female political leaders in times of threat. In two separate
studies, Brown et al. (2011) demonstrated the glass cliff effect in two
studies. In response to being primed about threats in their communities,
participants’ preference for a male leader relative to a female leader was
eliminated, and in response to a prime about the 9/11 terrorist attacks,
participants preferred a female leader relative to a male leader. These studies
suggest that not only do individuals prefer female political candidates when
the election is precarious but that the preference for a female political leader
may occur when the country is in crisis or under threat.
Therefore, I predict the
following:
H1: I predict that there
will be a preference for a female leader relative to a male leader when
participants are primed with a threat.
Stereotypes.
There
are a few explanations that might explain the glass cliff phenomenon. The first
is that stereotypes of leaders in times of crisis may align more closely with
stereotypes of women than stereotypes of men. For example, Ryan, Haslam, and
Bongiorno (2011) codified the “think crisis think-female” effect, showing that
stereotypes of leaders when a firm was in economic crisis were more consistent
with stereotypes of women than stereotypes of men. Other work (e.g., Gartzia,
Ryan, Balluerka, & Aritzeta, 2012) similarly suggests that consistent with
role congruity theory, individuals may prefer leaders whose traits align with
organizational roles, in this case, traits of a leader in a time of crisis.
Indeed, this research is consistent with research on ILTs that suggests that,
prototypes of typical leaders differ from those of ideal leaders (Junker &
van Dick, 2014). It might therefore be possible that ILTs of a leader in charge
of an organization in crisis might differ from those in an organization that is
thriving.
Selection Bias. There is also some
evidence that the glass cliff effect might depend somewhat on participant
gender. For example, there is some evidence that when an organization is
described as in crisis, women preferred feminine traits to masculine traits
more than did men, whereas men did not differentially prefer feminine or
masculine traits (Ryan et al., 2011). Other evidence similarly indicates that
women may be more partial to female leaders, especially in times of crisis. However,
it is also important to note that these gender effects are not consistent
across studies, as others (e.g., Haslam & Ryan, 2008; Ryan et al., 2010)
did not find that the glass cliff phenomenon depended on participant gender. Additionally,
Brown et al. (2011) did not find that preference for a female versus male
leader in times of crisis depended participant gender, although they found that
women more strongly associated women with change than did men. Therefore, the
effects of participant gender on the glass cliff effect are somewhat unclear. I
therefore propose to include participant gender as a factor in my experimental
design so as to control for its potential effects. However, I make no formal
hypotheses about the role of gender in preference for a female candidate when
the nation is under threat.
Glass Cliff and System
Justification. Some
work on the glass cliff work indicates that women may also be nominated to
precarious leadership positions as a means to uphold the status quo (Ryan &
Haslam, 2005). Indeed, Ryan et al. (2011) found that one of the reasons that
women were nominated to leadership positions is that women are expected to take
responsibility for the crisis and therefore may serve as a scapegoat for the
failure of the company. The potential for female leaders in precarious
positions to fail also reaffirms stereotypes that women cannot be leaders,
justifying existing social systems in which men are more likely to be found in
high-status roles than are women. Therefore, it is possible that the nomination
of a female leader may be a way to maintain rather than upend existing social
structures and it is therefore possible that the preference for a female (vs.
male) leader in times of crisis might be driven by system-justifying motives.
System justifying
ideologies allow individuals to believe that existing political, economic, and
social systems are fair and legitimate, regardless of whether individuals
benefit from these systems (Jost & Hunyady, 2005). Examples of such
ideologies include social dominance orientation (Ho, Sidanius, Pratto, Levin,
Kteily, & Sheehy-Skeffinton), the belief that some groups (e.g., Whites,
men) are inherently superior to other groups (e.g., Blacks, women). Other
examples of system justifying ideologies include political conservatism and
right-wing authoritarianism, both of which involve a preference for greater
societal traditionalism (e.g., Jost & Thompson, 2000). Among the
consequents of such ideologies are an increased support for societal institutions
and a decreased support for social change (Jost & Hunyady, 2005). In other
words, individuals who strongly system-justifying ideologies have a strong need
to maintain the status quo.
Sometimes a change in
leadership might be desirable in order to maintain the existing social system. Donald
Trump’s election might be an example of such a change. For example, although some
argue that Trump voters were motivated by a desire to change policies that
favored an increasingly globalized economy (e.g., Tumulty, Rucker, &
Gearan, 2016), others argued that Donald Trump’s election and the election of
similar populist figures is a backlash against
social change (e.g., Inglehart & Norris, 2016). For example, it is possible
that some voters considered eight years of a Black president and the
possibility of a female president to be too
much change to existing social structures in which Whites and men enjoy
higher social status than do Black and women, respectively. Therefore Trump’s
election potentially signaled a return to the status quo rather than a change
away from the existing system.
In a study on the
effects of system justification on the outcomes of the 2016 presidential
election, Azvedo, Jost, and Rothmnd (2017) found support indicating that
Trump’s election may have been motivated by a need to return to the status quo.
Although they found that general system justifying ideologies were associated
with stronger support for Hillary Clinton (perhaps supporting claims by some
that she was the “status quo” candidate), strong support for economic and
gender system justification were associated with greater support for Donald
Trump. Indeed, despite Trump’s criticism of an increasingly globalized economy,
Trump supporters showed an increased preference for the types of social
structures that drove those inequalities. Therefore, the desire to “Make
American Great Again” may have served to signal a return to the status quo rather
than a departure from it.
These results make a case
that in some cases, change (e.g., the election of a new political candidate)
may serve to uphold rather than change the existing social system. In this
sense, it is possible that a female candidate may be viewed as a form of change
that may serve to justify existing social systems rather than change them.
Indeed, Brown et al. (2011) found support for this hypothesis, demonstrating
that the tendency to prefer a female candidate relative to a male candidate in
times of crisis was especially pronounced among individuals high in system
justifying ideologies and social dominance orientation. Other evidence suggests
that the presence of female candidates may cue system-justifying ideologies.
For example, in a series of three experiments Brown and Diekman (2013)
demonstrated that the presence of female politicians increased system
justification. In particular, they showed that the presence of female
candidates lead to greater perceptions of fairness, a greater acceptance of
gender inequality, and a preference for the constancy rather than change. Therefore,
counter-intuitively, although female candidates may represent a change in the
sense that historically, U.S. presidents have been male, female candidates in
some ways may represent a return to the status quo.
It is likely that given
that individuals are generally motivated to perceive the system is fair that candidates
who are perceived as upholding traditional American values (consistent system
justifying ideologies such as political conservatism) will be more electable
and more likeable than candidates who are not perceived as upholding
traditions. Furthermore, given the fact that preferences for female candidates
may be driven by system justifying ideologies, it is possible that preference
for a female candidate may be especially strong when she is perceived as
upholding the system (i.e., upholding cultural traditions) than when she is
perceived as not upholding cultural traditions. Therefore, I predict the
following:
H2: Candidates who
endorse cultural traditions will be viewed more positively than candidates who
are not viewed as upholding cultural traditions.
H3: I further expect
this effect to be qualified by candidate gender such that the preference for a
female versus a male candidate in times of threat will be stronger when the
candidate is portrayed as upholding cultural traditions versus not mentioning
cultural traditions.
There is also some
evidence from the glass cliff literature that the type of threat may affect
preference for a female leader. For example, effects in the glass cliff
literature are unclear as to whether threats diminish the preference for a male candidate relative to a female
candidate or increase the presence
for a female candidate relative to a male candidate. For example, a distinct
preference for a female leader was found when threat was primed with an
unsuccessful company (Haslam & Ryan, 2008) or poor financial performance
(Ryan et al., 2011). However, in response to a crisis, Bruckmüller and Branscombe (2010)
merely found a decreased preference for male leaders.
Effects across Brown and
Diekman’s (2011) experiments were inconsistent as well. For example, when
participants were asked to write about threats in their local communities
(i.e., economic instability, violent crime, or job cuts), preferences for a
male leader decreased in the threat (vs. control) condition. However preference
for a female leader remained unchanged. However, in response to threat primed
by the 9/11 terrorist attacks, participants preferred a female leader to a male
leader. One explanation for these effects is that not just the mere presence of
the threat but the type of threat might affect preference for a female leader.
One such type of threat
that might affect the preference for a female (vs. male) leader might be
whether the threat to the system comes from an ingroup member (e.g., White
supremacist group) rather than an outgroup member (e.g., ISIS), as threats from
outgroup members may be perceived as more threatening to the status quo than
threats posed by an ingroup member. For example, non-Black American
participants were quicker to associate a Black face with a gun than a White
face and more often mistook a tool for a gun when presented with a Black face
rather than a White face (Payne, 2001). This tendency is also consistent
evidence in the United States indicating that although Americans are extremely
unlikely to die in an attack perpetrated by a foreign-born terrorist
(Nowraseth, 2016) and that nearly half of the deaths from terrorism can be attributed
to domestic rather than foreign groups (Willis, 2017), the media often portrays
foreign terrorism as a much more serious threat than domestic terrorism
(Powell, 2011).
It is therefore possible
that threats from outgroup members may produce a greater desire for a
protection of the existing system than might threats from ingroup members.
Therefore, if system justifying mechanisms produce a preference for a female
leader in times of crisis, preference for a female leader may increase when a
threat is posed by an outgroup member than an ingroup member. Furthermore, this
effect might be stronger if the leader is perceived as upholding cultural
traditions (i.e., representing a return to the status quo) rather than not
mentioning cultural traditions.
I therefore predict the
following:
H4: Preference for a
female leader relative to a male leader will be stronger when the terrorist
attack is described as coming from an outgroup member (i.e., ISIS) than an
ingroup member (i.e., a White supremacist group).
H5: There will be a
three-way interaction between target gender, mention of traditionalism, and
threat type such that preference for a female leader will be stronger when the
threat is posed by an ingroup member than an outgroup member, and this effect
will be stronger when the candidate endorses traditions (vs. no mention).
The purpose of this research was to investigate the glass cliff phenomenon in a political setting in which preferences political candidates may vary as a function of threat (i.e., posed by ingroup vs. outgroup member), and whether that candidate was described as endorsing traditions (i.e., upholding the status quo) or not. Participants were asked to read an online newspaper article describing a terrorist attack perpetrated by either ISIS (i.e., outgroup member) or a White supremacist group (i.e., ingroup member). Next participants were directed to watch two videos of two ostensive presidential primary candidates (one male and one female) responding to the attacks and whose political affiliation matched that of the participant. In their response, the candidates either endorsed traditions or did not endorse traditions. Participants then indicated their perceptions of the candidate, including how warmly the felt towards the candidate, how likely they were to vote for the candidate, how fit the candidate was for office, and how much they though the candidate would win in the general election. Finally, although I made no formal predictions involving participant gender, I propose to keep it as a factor in the design, as some studies have found that the glass cliff effect depends on participant gender. Thus, I have proposed a 2 (Target Gender) X 2 (Threat Type: Ingroup vs. Outgroup) X 2 (System Justification: Mention vs. No Mention) X 2 (Participant Gender) design with repeated measures of the first factor.
Overall, I expect that there will be a general preference for a female leader to a male leader, as all participants were primed with threat. I also expect a main effect of traditionalism such that participants will generally prefer a candidate who upholds tradition to one that does not. I further expect that preference for a female leader will be stronger when the threat to the group comes from an outgroup member (i.e., ISIS) rather than an ingroup member (i.e., ISIS), as an ougroup threat constitutes a larger threat to the existing system and may potentially prime a greater desire to return to the status quo. Finally, I expect a three-way interaction between target gender, threat type, and traditionalism, such that preference for a female leader will be stronger when the threat comes from outside (vs. within) of the group, particularly when the female leader is described as upholding cultural traditions (vs. no mention of traditionalism).
I propose to pilot the videos of
the candidates as well as the candidates’ scripts to ensure that any of the
effects of the study are not due to extraneous variables such as candidate
attractiveness.
I
propose to use 50 participants from a small mid-Western university.
Four videos will be made
of men and women talking for thirty seconds about their favorite television show.
Two targets of each gender will be included to ensure that no extraneous
variables particular to target might affect results. Participants will be called
into the lab for a study ostensibly on first impressions and will be asked to
watch videos of one man and one women (order counterbalanced across
participants) and rate their perceptions how attractive each speaker is and how
likeable each speaker is on a 1 (not at
all) to 7 (very much) scale. Thus,
the design is a 2 (Target Gender) X 2 (Target A vs. Target B) mixed model
design with repeated measures on the first factor.
After judging videos participants will be routed to a presumably unrelated study on political speeches. As participants in the main study will watch two clips of political candidates, slightly different (but equivalent) versions of each response (upholding vs. not holding traditions) will be necessary in order to be believable.
Therefore, I propose to
write four responses to a terrorist attack, two where candidates are described
as upholding traditions and two in which they are not. The text of one of the
responses in which participants were described as upholding traditions is
described below.
Our thoughts and prayers
go out to the families affected by the recent attacks. I want to promise the
American people that if elected, my administration will stop at nothing to make
sure perpetrators of such attacks are hunted down and swiftly brought to justice.
We stand by those who were hurt and lost loved ones today and will be a source
of comfort to them in their darkest hour. Let us remember that terror can never win against the American
traditions and values that I pledge to uphold and which make this country
so great. Thank you and may God
bless America.
In the no mention
condition, the candidate will state that terror an never win against the will
and spirit of the American people, which are indomitable and which make this
country so great.
Participants will read two of the four vignettes: one in which candidates were described as upholding traditions and one in which they are not. Thus, the design is a 2 (Version A vs. Version B) X 2 (Traditionalism: Mention vs. No Mention) design with repeated measures on the last factor. After each vignette, participants will indicate how effective they thought each statement was how caring they believed each statement was, and how presidential they thought each statement was on a 1 (Not at all) to 7 (Very much) likert-type scale.
Participants’ ratings of attractiveness and likeability of videos of candidates bill be submitted to a 2 (Target Gender) X 2 (Version A vs. Version B) mixed model ANOVA with repeated measures on the first factor. In order to move forward, perceptions of likeability and attractiveness will not vary as a function of target gender or version, nor will perceptions vary as a function of the Target X Version interaction.
Responses to statements will be analyzed as a function of a 2 (Version) X 2 (Traditionalism) mixed model ANOVA with repeated measures on the last factor. I expect that participants will view candidates who endorse traditionalism more positively than those who do not. In order to move forward with the main study, the critical tests must demonstrate that perceptions of the candidates do not vary as a function of version, nor should the effects of traditionalism depend on version.
I
propose to recruit approximately 300 college students from a mid-sized
Midwestern University.
Experimental manipulation. Participants
will be given a news report that describes the aftermath of a terrorist attack
in which 10 people were killed and 30 injured after a makeshift explosive
device went off in the middle of a busy city. According to the report, the
investigation is still ongoing, but federal authorities are crediting the
attacks to either members of ISIS or members of a White supremacist group. All
other aspects of the report, besides the ingroup (White supremacist) or
outgroup (ISIS) status of the attackers will be consistent across reports. After
the article, to participants will indicate their perceptions of how serious the
threat is on a 1 (not at all serious)
to 7 (very serious) scale to serve as
an attention check and manipulation check.
Next, participants will be
asked to indicate their political party and will be shown two videotaped
statements from male and female ostensive political candidates (the same men
and women from the pilot study) in which candidates will be described as either
upholding cultural traditions or not mentioning cultural traditions. Slightly
different versions of each statement from the pilot study will be used to
ensure participants never see the same statement twice. Therefore, the proposed
design is a 2 (Target Gender) X 2 (Threat Group: Ingroup member vs. Outgroup
member) X 2 (Traditionalism: Present vs. not) X 2 (Participant Gender) design
with target gender varying within subjects.
Dependent
variables. After
each video, participants will indicate their perceptions of how likeable each
candidate is, how effective they perceive each candidate to be, how likely they
would be to vote for each candidate in the presidential primary, and how likely
they think each candidate might be to win the presidential election on a 1 (not at all) to 7 (very much) scale.
Finally, participants
will answer a series of demographics questions such as their gender and race.
I propose
to run univariate statistics to assess for skew and kurtosis for the dependent
measures. I also propose to evaluate the factor structure of the dependent measure
using exploratory and confirmatory factor analyses, as it is possible that a
multi-factor structure (e.g., likelihood of voting for the candidate in the
primary and election) may load on a different factor from perceptions of
candidates’ likeability or competence. Reliability (Cronbach’s alpha) will also
be assessed.
Next all responses will
be analyzed as a 2 (Target Gender) X 2 (Threat Group: Ingroup member vs.
Outgroup member) X 2 (Traditionalism: Present vs. Not) X 2 (Participant Gender)
ANOVA with repeated measures on the first factor. I propose to analyze the dependent
variable both as a composite scale and as single item measures, as it is
possible that analysis of single item measures may yield somewhat different
results from analysis of the composite scale. If participant gender yields no
significant main effects or interactions, I propose dropping it as a factor
from the design to improve power. I also propose to test the effects of
participant ethnicity, although I have no formal hypotheses regarding
participant ethnicity.
This investigation will
seek to determine the types of conditions in which female leaders might be
perceived as more attractive than male leaders, specifically when a threat to
the system was posed by outgroup rather than ingroup members, and when candidates
signaled a return to the status quo (i,e., upheld cultural traditions). This
study would contribute to the literature in a few ways. First, there is
relatively little literature on the glass cliff among female political leaders
and it would therefore contribute new understanding to the conditions in which
women might be nominated to precarious political leadership positions.
Secondly, although there is evidence that system-justifying ideologies may play
a role in preferences for female leaders in times of threat, this study would
directly manipulate such ideologies, providing stronger evidence for its role
in preferences for potential female political candidates as previous studies
have measured system justification as a covariate. Additionally, better understanding
the conditions under which female candidates are more desirable more generally
may contribute to greater gender parity in political representation.
Although one implication
of this study is that female leaders may be more desirable when they are
traditional, I am not advocating for female candidates to be “traditional” in
any sense, particularly not in their endorsement of traditional gender roles. Female
candidates, like male candidates should run on issues that that they consider
to be important and should advocate for the public interest as they see fit. However
one potential implication of this study might be that the first female
president of the United States may be a conservative woman, despite
progressives’ push for more gender parity. Regardless, a better understanding
of the existing system may facilitate interventions that help female candidates
run for office and may contribute to greater gender parity in politics and in
leadership positions more generally.
Hillary
Clinton’s call for a female president candidate is a lovely vision, but not one
without potential struggles for the candidate to be. There is no doubt that
despite some advantages, female leaders, and female political leaders more
specifically have an uphill battle. However, there may be some situations and
aspects of the candidate that increase the likelihood of a female candidate,
suggesting that female candidates need not necessarily be dissuaded from
running when the nation is faced with threats (e.g., military threats) that
complement agency. Indeed, these may increase preference for a female
candidate, who might finally shatter that “highest hardest glass ceiling.”
You have to be 100% sure of the quality of your product to give a money-back guarantee. This describes us perfectly. Make sure that this guarantee is totally transparent.
Read moreEach paper is composed from scratch, according to your instructions. It is then checked by our plagiarism-detection software. There is no gap where plagiarism could squeeze in.
Read moreThanks to our free revisions, there is no way for you to be unsatisfied. We will work on your paper until you are completely happy with the result.
Read moreYour email is safe, as we store it according to international data protection rules. Your bank details are secure, as we use only reliable payment systems.
Read moreBy sending us your money, you buy the service we provide. Check out our terms and conditions if you prefer business talks to be laid out in official language.
Read more