Conference PaperPDF Available

Helping to Break the Glass Ceiling? Fathers, First Daughters, and Presidential Vote Choice in 2016

Helping to Break the Glass Ceiling?
Fathers, First Daughters, and Presidential Vote Choice in 2016*
Jill Greenlee
Brandeis University
greenlee@brandeis.edu
Tatishe Nteta
University of Massachusetts Amherst
nteta@polsci.umass.edu
Jesse Rhodes
University of Massachusetts Amherst
jrhodes@polsci.umass.edu
Elizabeth Sharrow
University of Massachusetts Amherst
sharrow@polsci.umass.edu
Prepared for presentation at the American Political Science Association Annual Conference, San
Francisco, CA, August 31 to September 3, 2017.
Working paper – please do not circulate or cite without author permission
* Authors are listed in alphabetical order. Authors acknowledge research support from a
University of Massachusetts Faculty Research Grant.
Greenlee, Nteta, Rhodes, and Sharrow
2
From the early stages of her campaign, 2016 U.S. presidential candidate Hillary Clinton
crafted messaging intended to appeal to fathers of daughters (Dickerson 2016). Beginning with
her June 2015 presidential campaign announcement speech, and again in the first Democratic
candidate debate, Clinton reminded voters that her election could mean that “fathers will be able
to say to their daughters, you, too, can grow up to be president” (Frizell 2015; New York Times
2015). Throughout her run for the presidency, Clinton routinely claimed that her election would
positively affect women and girls, suggesting that men, as fathers, should consider the interests
of their female children as they headed to the polls (Dickerson 2016). In making this appeal,
Clinton suggested that – in an election that was in so many ways about gender and women’s
place in American politics – there were also gendered dynamics relating to parenting that might
inform voters’ decisions. Clinton consistently sought to mobilize fathers of daughters to support
her candidacy. But was this strategy effective? In short, were fathers of daughters more
supportive of Clinton in the 2016 presidential election compared to fathers of sons?
Although much as been written in recent years about the ways in which gendered roles
shape women’s engagement in the political world, much less has been said about men (Greenlee
2014). And while a small number of studies have investigated how fathering daughters affects
men’s gender attitudes (i.e. Glynn and Sen 2015; Healy and Malhotra 2013), to date no study has
investigated how fathering a daughter may influence men’s vote choices – especially in an
election with a woman on the ballot. Thus, as the first woman to be nominated as the U.S.
presidential candidate from a major political party, Clinton’s candidacy presents an
unprecedented opportunity to explore how being a father – specifically being the father of a
daughter – might shape men’s presidential preferences.
Greenlee, Nteta, Rhodes, and Sharrow
3
In this paper, we explore the relationships among gender, fatherhood, and vote choice in
the 2016 election. More specifically, we ask: were men who fathered daughters (or fathered
daughters as their first child) more likely to support, and vote for, Hillary Clinton in the 2016
presidential election than were those who fathered sons (or fathered sons as their first child)?
To answer these questions, we rely on two unique sources of data: 1) an original, nationally-
representative survey data from a 2016 Cooperative Congressional Election Study (CCES)
module and 2) a survey experiment embedded in our 2016 CCES module designed to explore the
impact of exposure to a “Clintonesque” campaign appeal on father’s vote choice.
The results of our observational analyses strongly suggest that the experience of having a
daughter as a first child – but not the effect of having a daughter in general – significantly and
substantively increases fathers’ probability of supporting and voting for Clinton in 2016. To
further strengthen confidence in our findings, we also conduct a placebo test that shows that
having a daughter as a first child does not increase the probability of having voted for Barack
Obama for president in 2012 among our sample of fathers.
Finally, to show that Clinton’s message played an important part in her appeal among
fathers of first daughters, we use a survey experiment to test how exposure to a “Clintonesque”
appeal emphasizing the positive effect of the election of a female candidate on daughters affects
support for a fictional female congressional candidate. Consistent with our observational
findings, we show that fathers of first daughters who were randomly exposed to the
“Clintonesque” appeal were significantly more likely to support the fictional female candidate.
As a whole, we take our findings as clear and consistent evidence that Clinton’s appeals
to fathers of daughters were indeed effective in the 2016 election, but that these appeals uniquely
swayed fathers of first daughters and not fathers of daughters more generally. These results
Greenlee, Nteta, Rhodes, and Sharrow
4
suggest that the experience of having a first daughter may be a transformative experience that
leads men to not only view issues of gender inequality in a new light, but to apply this newfound
perspective to their candidate evaluations and vote choice.
Gender and the 2016 Presidential Election
Gender is among the many major factors contributing to voters’ choices in an election
which pitted a former Secretary of State, U.S. Senator, and First Lady, who the New York Times
(2016) declared “the most broadly and deeply qualified candidate in modern history” against
Donald Trump, a former reality television star and real estate magnate, who openly espoused
racist and misogynist views.1 While early studies and commentaries point to the importance of
racial resentment and economic anxieties (Ehrenfreund and Clement 2016; Roberts 2015), anti-
black attitudes (Tesler 2016d), anti-Muslim sentiment (Klinkner 2016), authoritarian views
(MacWilliams 2016), economic class (Manza and Crowley 2017; Stonecash 2017), and anti-
immigrant attitudes (Jones and Kiley 2016) in shaping how voters perceived the candidates –
scholars are only beginning to uncover the many ways in which gender shaped voters’
evaluations and choices (Junn 2017; Schaffner, MacWilliams, and Nteta 2017; Strolovitch,
Wong, and Proctor 2017).
Studying gender—and gender at the intersections of race, sexuality, and class—in the
2016 election requires nuanced thinking. The campaign highlighted the ways in which gender
hierarchies, heteropatriarchy, and gendered oppression structure American politics (Strolovitch,
Wong, and Proctor 2017). Importantly, scholars who monitored the day-to-day details of the
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
1 In addition to the papers which demonstrate the centrality of racist and sexist attitudes in vote choice for Donald
Trump (Schaffner, MacWilliams, and Nteta 2017; Strolovitch, Wong, and Proctor 2017), a long history of political
science research demonstrates that racism and sexism have long been central to electoral politics and voting
decisions, particularly in presidential elections (i.e., Kinder and Sanders 1996; McDermott 1998; Mendelberg 2001;
Paul and Smith 2008; Sharrow et al. 2016; Sigelman and Sigelman 1982; Tesler 2016b; Valentino, Hutchings, and
White 2002).
Greenlee, Nteta, Rhodes, and Sharrow
5
2016 campaign specifically note that much remains unexplored about the cross-cutting
demographic, social, and political factors informing presidential preferences among American
men, particularly in light of Clinton’s unique candidacy (Dittmar 2017). In this paper, we begin
to think carefully about how an import gendered relationship may shape men’s electoral
preferences.
Women, Men and Support for Clinton
To date, the research investigating the gendered aspects of support for Clinton has largely
focused on women. Some scholars argue that, rather than benefiting from in-group solidarity,
Clinton was penalized by women voters (Tesler 2016e). Evidence from the Democratic primary
(where Clinton faced a challenge from Vermont Senator Bernie Sanders) suggests that Clinton
struggled to find support among younger women, except among those who had directly
experienced sex discrimination and sexism (Poloni-Staudinger, Strachan, and Schaffner 2016;
Strolovitch, Wong, and Proctor 2017). Support for Clinton in the general election, among both
women and men, was conditioned on progressive attitudes towards women’s status and role in
society (Clarke and Steward 2017).
While scholars note important in-group differences in voting patterns among women, less
is currently understood about in-group voting patterns among men.2 In this paper, we examine
the impact of one such in-group differences among a subset of American men – the gender of
their children.3 Why should fathering a daughter influence a man’s vote choice in a presidential
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
2 One important exception is the racial divide among men in vote choice for Trump, with men of color were more
likely to vote for Clinton in 2016 (Asian American Decisions 2016; CNN 2016; Latino Decisions 2016;
Ramakrishnan 2016).
3 While Tesler (2016a, 2016c) found evidence that Clinton was drawing meaningful support from men who fathered
daughters in the early months of the 2016 campaign, to date there has been no study that has examined if fathers of
daughters were more likely to support Clinton in the general election. In this paper, we further explore the impact of
daughters on men’s vote preferences in the general election using high-quality data collected at the conclusion of the
2016 campaign.
Greenlee, Nteta, Rhodes, and Sharrow
6
election featuring a female candidate? To answer this question, we develop a novel explanation
about how fatherhood of daughters affects men’s presidential preferences.
Fatherhood of Daughters and Support for Female Candidates
Decades of research have concluded that as a general matter, men are expected to have a
lower level of support for female candidates than do female voters. This is largely due to their
greater inclination to hold policy preferences consistent with Republican partisanship (Box-
Steffensmeier, Boef, and Lin 2004; Huddy, Cassese, and Lizotte 2008; Kaufmann 2002; Shapiro
and Mahajan 1986; Wolbrecht 2000) and to their baseline preference for male candidates
(Sanbonmatsu 2002).4
Yet, while much scholarly attention has been directed at accounting for the factors which
explain the gender gap in vote choice for female candidates, there exist several individual-level
factors that have been identified as important in understanding male support for female
candidates that include: Democratic partisanship (Sanbonmatsu 2002; Sanbonmatsu and Dolan
2009), lesser reliance on negative stereotypes toward female candidates (Brooks 2013; Dolan
2014), and greater favorability toward the abstract idea of electing a female president (Burden,
Ono, and Yamada 2017). To this literature, we suggest an additional individual level trait,
fatherhood of a daughter(s), as a key factor that may increase male support for female candidates
among men.
Recent research on the behavior of elites (Cronqvist and Yu 2016; Dahl, Dezső, and Ross
2012; Glynn and Sen 2015; Washington 2008) and studies of the attitudes of ordinary citizens
(Oswald and Powdthavee 2010; Sharrow et al. 2017; Warner 1991; Warner and Steel 1999)
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
4 Sanbonmatsu (2002) connects this baseline preference to gender stereotypes. And, in fact, Dolan (2010, 2014)
finds that male voters are more likely to see male candidates (as opposed to female candidates) as strong on an array
of policy issues, results which also suggest a preference based on gender stereotypes. However, Dolan (2014) finds
no evidence that gender stereotypes play in to the ultimate electoral decision of these voters.
Greenlee, Nteta, Rhodes, and Sharrow
7
suggest that fatherhood of daughters is associated with more liberal attitudes in general and with
more positive views toward gender equity and women’s reproductive rights. Scholars suggest
multiple mechanisms which might drive fathers of daughters to take stances that advance the
wellbeing and status of women, including self-interest (Warner 1991; Washington 2008), greater
awareness of gender inequality (Warner and Steel 1999), and personal relationships that teach
men about women’s issues (Glynn and Sen 2015). Given this repeated finding – that fathers of
daughters exhibit more liberal preference – we consider the implications of this relationship in
the context of an election.
“Paternal Voting” Hypothesis
Why should fathers of daughters, relative to fathers of sons, more strongly support female
candidates? In a presidential campaign where messages about women and girls were ubiquitous
– as a result of the historic nature of Clinton’s candidacy, campaign messages that drew parallels
between Clinton’s election and the opportunities of girls and women, and the wide spread
critiques of Trump for his derogatory statements about women – matters of gender may have
been a stronger consideration for fathers of daughters than in previous presidential elections. In a
campaign context that makes fathering a daughter a salient consideration, we suggest that men
may more strongly consider their paternal roles or interests as they cast their vote. Thus, we
hypothesize that:
H1: Paternal Voting Hypothesis: Men who father daughters, regardless of the birth order
of their daughters, will express more positive evaluations of Hillary Clinton, and will be
more likely to report voting for Clinton in the 2016 presidential election, compared to
men who father only sons.
Greenlee, Nteta, Rhodes, and Sharrow
8
“First Daughterhood” Hypothesis
Relatedly, we hypothesize that entering into fatherhood with a female child, a status
which we call “first daughterhood,” should contribute to greater support for Clinton. The
political socialization literature strongly suggests that political experiences, contexts, or
messages that occur early in life can have important and persistent effects on political
preferences (Schuman and Scott 1989; Sears 1981; Sears and Levy 2003). Dinas (2013) makes
one such an argument in examining why the “impressionable years” – typically the ages of 18-25
– are particularly consequential to the political evaluations of individuals. He argues that it is
“the weight” individuals attach to the political events and information that they receive during
this period because they are newly engaged in the political sphere (Dinas 2013, 868).
This thinking offers a window into why first daughters, rather than daughters born after a
male child, might be especially consequential to men’s evaluations of female candidates. The
initial entry into fatherhood with a daughter may mark a pivotal moment when men may awaken
to the ways in which gender inequality persists in society, making them particularly receptive to
campaign messages that connect gender inequality – or the wellbeing of girls and women – with
the lives of their daughters. Like young adults who enter into the political sphere with weak
political preferences and who are then exposed to strong information flows, new fathers of
daughters may also encounter new information or bring a new perspective to bear on their
evaluations of previously held positions (Sears and Funk 1999; Sears and Levy 2003).5 This
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
5 Scholars have asserted that the transition into fatherhood is a “critical juncture in men’s development” (Palkovitz
and Palm 2009, 7) because of its complex and multifaceted effects (LaRossa and LaRossa 1981; Nystrom and
Ohrling 2004; Palkovitz and Sussman 1988). Not only do men adopt a new social role, but they also experience new
responsibilities, strain, and emotional attachments (Nystrom and Ohrling 2004). Scholars have also noted that as a
child grows, fathers experience additional transitions in relation to their fathering role, as the concerns and demands
on a father shift with the development of a child (Palkovitz and Palm 2009). While this literature does not speak to a
father’s political concerns, we posit here that the “critical juncture” of becoming a father, and the subsequent new
experiences that a father has with a first daughter, may be politically consequential.
Greenlee, Nteta, Rhodes, and Sharrow
9
increased sensitivity may make these men more receptive to discourse within the political
environment of the 2016 campaign which connect Clinton’s election to the White House with
their daughters’ own possibilities. However, for men whose entry into fatherhood (with a son)
did not orient them toward gender inequality in the same way, these messages may not be as
resonant. This process among men with first daughters is shown to robustly predict men’s
support for sex equity policies (Sharrow et al. 2017), findings which we aim to test in the
electoral context. Thus, this reading of the political socialization literature suggests that first
daughters should lead to greater support for Hillary Clinton.6 We therefore hypothesize:
H2: First Daughterhood Hypothesis: Men whose first child is a daughter will express
more positive evaluations of Hillary Clinton compared to men whose first child is a son
and will be more likely to vote for Clinton in the 2016 presidential election.
Data and Methods
In evaluating these hypotheses, we first make use of original, cross-sectional survey data
that allow us to more thoroughly investigate the effects of having a daughter on presidential
candidate preferences, while also considering the implications of family structure that may shape
the impact of fatherhood in important ways. The data in this study come from questions we
designed for a module of the 2016 Cooperative Congressional Election Study (CCES). The
CCES is an online survey of over 64,000 American adults conducted via YouGov on behalf of
over 50 colleges and universities. This collaborative study has been shown to produce estimates
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
6 Shafer and Malhotra (2011) find that first daughters reduce fathers’ support for traditional gender roles, though
they employ this approach because the sex of a first-born child best approximates the conditions of a natural
experiment. This mode of analysis may unwittingly capture a meaningful difference in the parenting experience with
the first child. Birth order research finds that first born children get more time with their parents than subsequent
children (Price 2008). Thus, while Shafer and Malhotra’s intention is to cleanly delineate the random assignment of
a child’s sex from the reproductive choices of parents who may wish to have both girls and boys, their
operationalization may also be capturing unique experiences inherent to the relationship between a father and his
first-born daughter.
Greenlee, Nteta, Rhodes, and Sharrow
10
similar to telephone and mail surveys (Ansolabehere and Schaffner 2014).7 It gathers a
representative sample of respondents of “opt-in” volunteers from the YouGov database.8 The
2016 CCES was in the field during September and October 2016 (pre-election), and in
November 2016 (post-election).
The total number of respondents available to researchers in our module was 1,500;
however, as described below, the total number of fathers of children (the subject of our study,
and the operational definition of which is described below) was smaller.9 The questions in our
module were designed to provide maximum insight into fathers’ family structures, in particular
the gender and birth order of children.
Our main dependent variables gauge respondents’ support for Hillary Clinton in the 2016
campaign/election. To measure preferences for Clinton, we used two variables from the CCES
Common Content (questions asked of all respondents and available to all research teams). We
measured respondents’ pre-election preference for Clinton using a variable from the pre-election
questionnaire querying presidential preferences (preference for Clinton was coded 1; preference
for any other candidate was coded 0). We also determined whether respondents reported voting
for Clinton, using the CCES post-election vote choice item (a self-reported vote for Clinton was
coded 1; while a self-reported vote for any other candidate was coded 0).
A third dependent variable, vote for Obama in 2012 (with a self-reported vote for Obama
coded as 1, and a vote for any other candidate coded as 0) is also included in our analysis as a
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
7 See http://cces.gov.harvard.edu/ for full survey description and data archive. At this time, only a preliminary
release of the 2016 CCES Common Content Dataset and questionnaires are available online. Data from specialized
modules, including the data on which our findings described herein are based, are not yet publically available, nor is
a full description of weighting methods.
8 See CCES Guide 2016 for information about survey weights:
https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi%3A10.7910/DVN/GDF6Z0.
9 Since becoming a father entails a variety of selection processes, we did not compare fathers and non-fathers in
order to avoid inappropriate comparisons.
Greenlee, Nteta, Rhodes, and Sharrow
11
“placebo” test of our hypotheses. We suggest that fatherhood of daughters/fatherhood of a
daughter as a first child should increase support for Clinton due to their effect on fathers’
perceptions of gender relations and their potential susceptibility to the symbolic importance of
the election of the nation’s first female president and campaign appeals that speak to issues of
gender equality. However, if the paternal voting and first daughterhood hypotheses are correct,
fatherhood of a daughter/first daughterhood should not predict self-reported Obama voting in
2012. Therefore, we include the Obama vote measure as an ancillary test of our main hypotheses.
In our module, we asked respondents several questions about family structure to obtain a
clear sense of respondents’ parental roles and the gender of their children. For the purposes of
this study, “parents” are individuals who self-identified as male or female on the CCES Common
Content (the portion of the survey asked of all 64,600 respondents) and indicated that they had
children (this question was posed only to respondents in our module and we directed respondents
to count all children, whether alive or deceased, from all marriages/relationships, and regardless
of age).10 Of our 1,500 respondents, 931 identified themselves as parents of children, and 521 did
not.11 We then identified as “fathers” all individuals who self-identified as male and indicated
that they were parents of children. Finally, we limit our analysis to respondents who indicated
that they had five or fewer children. This results in a final sample of 382 fathers.
All of the fathers in our sample of men who indicated that they were parents of at least
one child were asked a series of questions about each of their children, most importantly each
father was asked the gender and age of each of their children. With this information in hand, we
were able to identify: 1) fathers of only daughters, 2) fathers of only sons, 3) fathers who had a
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
10 One hundred percent of respondents identified as either male or female. Two percent of respondents also
identified as “transgender” on a question specifically querying identification with transgender identity, and an
additional 1.4 percent also responded as “prefer not to say” on this question.
11 Forty-eight respondents did not provide a response to the question querying the number of their children.
Greenlee, Nteta, Rhodes, and Sharrow
12
daughter first, and 4) fathers who had a son first. In our sample, 263 of our 382 fathers (69
percent) were fathers of daughters and 170 of our 382 fathers (45 percent) of daughters had a
daughter as a first child.12
In each of our models, we control for a wide array of factors that may influence fathers’
presidential preferences. We control for respondents’ race, age, educational attainment, marital
status, income, party identification, ideology, employment status, and whether the respondent
had a sister.13 We also control for a number of factors that were widely perceived to have an
important effect on presidential preferences in 2016. Given previous work indicating the
importance of gender attitudes in determining candidate preferences (Shafer and Malhotra 2011),
we control for a respondent’s general support for gender equity by including a four-item,
“Gender Equality Attitudes Scale” index (alpha=.31).14 To account for the impact of racial
attitudes on presidential preferences, we control for respondents’ racial resentment using a four-
item “Racial Resentment Scale” (alpha=.87).15 We also control for respondents’ perceptions of
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
12 Although we obtained information about the number of daughters fathered by the fathers in our sample, we did
not use this information in our analysis. Only a very small proportion of the fathers of daughters in our sample
reported having more than 2 daughters. Given the dearth of observations at high values of this variable, it is not
feasible to make sound inferences about the effects of having a large number of daughters on fathers’ presidential
candidate preferences.
13 We include this control in light of Healy and Malhotra’s (2013) finding that men with sisters may express more
conservative gender attitudes.
14 This composite measure is based on respondents’ levels of agreement with four statements: (A) “A preschool
child is likely to suffer if his or her mother works,” (B) “If his wife works full-time, a husband should share equally
in household chores, such as cooking, cleaning, and washing” (reverse coded), (C) “A husband should earn a larger
salary than his wife,” and (D) “The way society is set up, men have more opportunities than women” (reverse
coded). The order in which respondents were presented with these statements was randomized. Respondents were
asked to indicate their level of agreement to the statements on a 5-point scale, ranging from “Strongly Agree” to
“Strongly Disagree.” These measures were also used by Cassese and Holman (2017).
15 This composite measure is based on respondents’ levels of agreement with four statements: (A) “The Irish,
Italians, Jews and many other minorities overcame prejudice and worked their way up. Blacks should do the same
without any special favors,” (B) “Generations of slavery and discrimination have created conditions that make it
difficult for blacks to work their way out of the lower class,” (C) “Over the past few years, blacks have gotten less
than they deserve,” (D) “It’s really a matter of some people not trying hard enough; if blacks would only try harder
they could be as well off as whites.” The order in which respondents were presented with these statements was
randomized. Respondents were asked to indicate their level of agreement to the statements on a 5-point scale,
ranging from “Strongly Agree” to “Strongly Disagree” (Kinder and Sanders 1996).
Greenlee, Nteta, Rhodes, and Sharrow
13
whether economic conditions were poor in order to assess the impact of negative economic
evaluations on support for Clinton. Finally, we account for two contextual factors that might
influence respondents’ presidential candidate preferences in 2016: the proportion of the
respondent’s state that is female; and whether the respondent’s state was a “battleground” state in
2016.16
We took additional steps to ensure consistency in our analyses of the different dependent
variables. We limit our analysis to fathers who were self-reported “definite voters” on the pre-
test for our model of pre-election presidential candidate preferences; and to those who were self-
reported voters on the post-test for our model of post-election vote choice.17 Additionally, in all
of the statistical models presented below we limit our analysis to fathers who completed both the
pre-election and post-election surveys. Finally, in our models we cluster the standard errors by
state in order to control for any local correlation in the error term that may not apply to other
states (Arceneaux and Nickerson 2009; Green and Vavreck 2008).
Results
Study 1: 2016 Presidential Preferences & Vote Choice
We first model fathers’ pre-election preference for Clinton (compared to all other
candidates) as a function of all of our independent variables. The results of the logistic regression
model are presented in Table 1.
Insert Table 1 Here
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
16 Our measure of “battleground” relied on Politico’s list of such states and include: Colorado, Florida, Iowa,
Michigan, Nevada, New Hampshire, North Carolina, Ohio, Pennsylvania, Virginia and Wisconsin.
17 We defined “definite voters” as all individuals who indicated that they “definitely” planned to vote in the 2016
elections (or indicated they had already done so); and we defined as voters all individuals who self-reported that
they “definitely voted in the General Election.” We acknowledge that our measure of voting is based on a self-report
and is affected by social desirability bias (Cuevas-Molina 2017).
Greenlee, Nteta, Rhodes, and Sharrow
14
Generally consistent with the findings of recent research on preferences in the 2016
election, our model suggests that among fathers: whites, older fathers, Republicans,
conservatives, employed fathers, fathers with less egalitarian gender attitudes, and those who
believe that the economy was in poor shape were significantly less likely to prefer Clinton
relative to other candidates.
At the same time, the model provides important insights about the impact of fatherhood
of daughters on fathers’ presidential preferences. We do not find support for H1, the paternal
voting hypothesis. Contrary to expectations, the logistic regression coefficient for the variable
Father of Daughter is negatively signed, and not statistically significant at conventional levels.
These results suggest that, among fathers, simply having a daughter as a child does not affect
fathers’ preferences for Clinton.
However, the model does provide evidence in strong support of H2, the first
daughterhood hypothesis. The logistic regression coefficient for Father of First Daughter is
positively signed and statistically significant at the p<.01 level, indicating that fathers who have a
daughter as their first child have a significantly higher probability of preferring Clinton
compared to other candidates net of other factors that also influence presidential preferences.
Moreover, the effect of having a daughter as a first child is substantively important: among
fathers, having a daughter as a first child increases the predicted probability of preferring Clinton
by 10 percentage points. Holding other variables at their mean values, fathers who had a son as
their first child have a predicted probability of supporting Clinton of .31; while fathers who had a
daughter as their first child have a predicted probability of supporting Clinton of .41. Thus, the
model provides clear indications that, net of other variables that influence preferences for Clinton
in 2016, having a daughter as a first child is an important factor in determining fathers’ support.
Greenlee, Nteta, Rhodes, and Sharrow
15
What, if any, impact did daughters have on their father’s reported vote choice in the 2016
presidential election? Table 2 presents the results from a logistic regression model that examines
the reported vote choice of fathers in the 2016 presidential election. In our model, we find that a
number of factors help to explain vote choice for Clinton in 2016. Unsurprisingly, we find that
partisanship and ideology are the strongest factors that predict vote choice among fathers. When
holding all other variables at their mean values, the movement from a strong Democrat to a
strong Republican yields a 98-point decrease in the likelihood that a father will vote for Clinton.
We discover a similar effect when examining the impact of ideology. In line with previous work
on the 2016 election (Schaffner, MacWilliams, and Nteta 2017), we also find that economic
anxiety, negative gender attitudes, and negative racial views all play significant role in predicting
opposition to Clinton. Interestingly, we find that married fathers are twelve points more likely to
support Clinton when holding all other variables in the model at the mean value.
Insert Table 2 Here
Of note, we once again find little support for H1, the paternal voting hypothesis, as our
measure of fatherhood of daughters, while positively signed, does not attain standard levels of
statistical significance in our model. However, as seen in Table 2, our item that taps whether a
male respondent had daughter as a first child is positive and statistically significant (p=.05) in
predicting support for Clinton in 2016. Examining the predicted probability of support for
Clinton based on whether a father has a first-born son or a first-born daughter, we find that
moving from having a first-born son to a first born daughter increases support for Clinton by
eight points when holding all other variables in the model at their mean values.
Thus far, our findings provide considerable support for our first daughterhood
hypothesis, suggesting that having a first daughter has a meaningful impact on men’s electoral
Greenlee, Nteta, Rhodes, and Sharrow
16
preferences in an election in which gender was highly salient. However, if our hypothesis is
indeed correct, it stands to reason that the first daughterhood effect should not extend to elections
that do not feature a prominent female candidate. In order to test this supposition, we explore
fathers’ reported vote choices in the 2012 presidential election between President Barack Obama
and former Massachusetts Governor Mitt Romney. Once again, we scale our dependent variable
0 to 1 with 1 representing men who reported voting for Obama and 0 men who reported voting
for another candidate. To maintain consistency with our model of vote choice in 2016, we
include all of the covariates from our 2016 model in our model of vote choice in 2012. As seen
in model 2 in Table 2, we find that white racial identity, negative racial attitudes, partisanship,
the belief that the economy is faltering predict opposition to Obama in 2012, while living in a
battleground state predicts support for Obama. In line with our expectations, we find that neither
having a daughter nor having a first-born daughter emerge as a significant predictor of vote
choice for Obama in 2012. These null findings suggest that the estimated effect of having a
daughter as a first child on support/vote for Clinton is neither a spurious result attributable to
sampling error nor a stand-in for a more general attitude such as ideological liberalism or
Democratic Party identification.
Study 2: 2016 Minnesota Congressional Vote Choice Experiment
While our results shed light on the relationship between fathers of daughters and vote
choice in 2016, it remains unclear if messages around Hillary Clinton’s candidacy – including
numerous campaign appeals specifically directed at fathers of daughters – led these men to more
strongly support Clinton in the general election. In order to address this shortcoming, and to
make a stronger causal case in support of the relationship between exposure to campaign appeals
and vote choice among fathers of daughters, we make use of a survey experiment embedded in
Greenlee, Nteta, Rhodes, and Sharrow
17
our pre-election module of the 2016 CCES. Given our interest in the manner in which fathers
reacted to these appeals, we once again only include respondents who self-identified as a father.
Additionally, in order to maintain linearity with our cross-sectional analysis, we restricted our
sample to fathers who also completed both the pre- and post-election surveys which results in a
final sample of 282 fathers.
In our survey experiment, respondents were randomly assigned to one of three
treatments. In all three treatments, respondents were asked their opinion on their likelihood of
voting for Molly Smith, a fictional candidate running in an upcoming congressional election that
will take place in Minnesota’s 10th District.18 Respondents in the control condition were shown
the following statement, “Molly Smith is running to become the first woman to represent
Minnesota’s 10th District” and then asked to “Imagine you live in Smith’s district. Please indicate
your likelihood that you would vote for her?” Respondents in the remaining two conditions were
also informed that Smith, if elected, would become the first woman from the 10th District to
serve in the House of Representatives, and asked the same question concerning their intention to
support Smith. However, in each of the two remaining experimental conditions respondents were
randomly assigned to receive additional information concerning Smith’s policy positions and her
views on the symbolic importance of the election.
Respondents in the first experimental treatment (STEM Treatment) were exposed to the
following statement, “Molly Smith is running to become the first woman to represent
Minnesota’s 10th District. She supports policies that would help increase the participation of
women in careers in science, technology, engineering, and mathematics (STEM)” and then asked
to indicate their likelihood of support for Smith.
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
18 In order to guard against the potential that a respondent may live in one of Minnesota’s actual nine congressional
districts, we created the fictional 10th district of Minnesota.
Greenlee, Nteta, Rhodes, and Sharrow
18
In our third experimental treatment (Clinton Treatment), respondents were asked to
indicate their likelihood of support for Molly Smith after being exposed to the following
statement,
Molly Smith is running to become the first woman to represent Minnesota’s 10th District.
She supports policies that would help increase the participation of women in careers in
science, technology, engineering, and mathematics (STEM). She has said of her
candidacy, “This campaign is about making sure there are no ceilings, no limits on any of
us, and to ensure that our daughters will forever know that there is no barrier to who they
are and what they can be in the United States of America.”
In order to maximize the external validity of our treatment, we designed this treatment to
hue closely to Clinton’s campaign appeals made to fathers of daughters during the 2016
presidential election. In our treatment, we emphasize not only the historic nature of Smith’s
candidacy, but also her pledge to pass policies that increase the status of women; two key aspects
of Clinton’s campaign appeals. Most importantly, our treatment speaks to the symbolic
importance of Smith’s elections for daughters in Smith’s district, a theme that Clinton
consistently emphasized on the campaign trail.19
Does exposure to a “Clintonesque” campaign appeal influence fathers of daughters to
more strongly support our fictional female congressional candidate? Table 3 shows the results of
an OLS regression analysis in which each of our experimental treatment conditions is
represented by a dummy variable, and fathers in the control condition serve as the baseline group
and thus excluded from the model (Ladd 2010).20 In order to better assess how exposure to each
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
19 We see evidence of this appeal in Clinton’s speech upon accepting the Democratic nomination for president when
she said: “Tonight, we've reached a milestone in our nation's march toward a more perfect union: the first time that a
major party has nominated a woman for President. Standing here as my mother's daughter, and my daughter's
mother, I'm so happy this day has come. Happy for grandmothers and little girls and everyone in between. Happy for
boys and men, too because when any barrier falls in America, for anyone, it clears the way for everyone. When
there are no ceilings, the sky's the limit. So let's keep going, until every one of the 161 million women and girls
across America has the opportunity she deserves” (Clinton 2016).
20 We restricted the sample to respondents who correctly answered our attention check question which asked, “In
what state will Molly Smith’s congressional election take place?”
Greenlee, Nteta, Rhodes, and Sharrow
19
campaign appeal may influence the vote choice of fathers with daughters, we interact each
treatment with our measure of whether the respondent is a father of a daughter (Model 1) or is a
father of first daughter (Model 2). Additionally, given that the partisanship of our fictional
candidate is kept hidden from our respondents, we control for the partisanship of the respondent
as well as their assessment of the ideology of our fictional candidate.21 We score our dependent
variable such that higher scores represent greater support for our fictional candidate.
As seen in Model 1 of Table 3, we once again find little evidence that fathers of
daughters react positively to campaign appeals from female candidates. The interaction between
fathers of daughters and: 1) exposure to a campaign appeal focused on increasing the numbers of
women in STEM fields, and 2) exposure to a campaign appeal that speaks to the symbolic
importance of Smith’s election both fail to reach conventional standards of statistical
significance. On the other hand, in Model 2 of Table 3 while the interaction of fathers of first
daughters and exposure to the STEM treatment does not attain statistical significance, we find
support for the first daughterhood hypothesis as fathers of first daughters who are exposed to the
“Clintonesque” treatment express stronger support for Smith candidacy when compared to
fathers of first daughters in the control condition (b=.1619, p=.01).
Insert Table 3 here
Given the difficulty in interpreting the coefficients emanating from interactions, we chart
the marginal effects for this interaction term in Figure 1.22 In Figure 1, we find that the
significant differences between fathers of first daughters and fathers of first sons are found
""""""""""""""""""""""""""""""""""""""""""""""""""""""""
21 The ideology of Molly Smith is assessed with the question, “Where would you place Smith on the ideological
scale?”
22 When using predicted values in order to determine statistically significant marginal effects, it is too conservative
to use two separate 95% confidence intervals (Knezevic 2008). When the standard errors are roughly equivalent, a
single 95% test translates into using two sets of 84% confidence intervals (Payton and Greenstone 2003). Figure 2
displays two separate 84% confidence intervals.
Greenlee, Nteta, Rhodes, and Sharrow
20
among fathers who were exposed to the Clintonesque treatment. Among fathers of first
daughters, movement from the control condition to the Clintonesque treatment yields an increase
of 21 percentage points when holding all other items in the model at their mean level. On the
other hand, among fathers with first sons moving from the control condition to the Clinton
treatment leads only to a five-percentage point increase in support for Smith when holding all
other variables in our model at their average level.
Insert Figure 2 Here
Conclusion
Recent research on fathers of daughters has found that the experience of having a
daughter has a transformative effect on the manner in which men view the political world with
fathers of daughters more likely to support sex equity policies (Sharrow et al. 2017) and exhibit a
greater commitment to norms of gender equality (Warner and Steel 1999). While these studies
suggest that fathers of daughters will support female political candidates, particularly if these
candidates appeal to their status as fathers of daughters, to date there has been little empirical
evidence in support of this expectation. Here, using original survey data, and leveraging the
presence of the first major party female presidential candidate to explore the relationship
between fatherhood of daughters and vote choice, we find that fathers with first daughters, when
compared to fathers of first sons, are more likely to strongly support Clinton in the general
election. We bolster these findings with the results from our survey experiment that suggests
exposure to Clinton’s messages that directly appealed to fathers of daughters helps to account for
the preferences of fathers of first daughters.
There are several implications flowing from these findings. As has been well
documented, there has been a long-standing debate in the literature on political communication
Greenlee, Nteta, Rhodes, and Sharrow
21
concerning the actual influence of campaign appeals on vote choice (for review see Goldstein
and Ridout 2004; Lau and Rovner 2009). On the one hand, studies of political advertisements
have found that exposure to political advertisements positively influences vote preferences
(Gerber et al. 2011) while others argue that exposure to political advertisements has very little
impact on vote choice when tested alongside traditional determinants of voter preference such as
partisan identification, sociotropic perceptions, and direct mobilization efforts on the part of
candidates (Krasno and Green 2008; Lau and Rovner 2009). Our results suggest that the
symbolic importance of the first woman presidential party candidate on a major ticket, and the
specific campaign discourse from Clinton’s campaign that tapped into men’s experience as
fathers of daughters had a real and positive impact on a subset of male voters. And while the
Clinton campaign likely crafted these messages to resonate with all fathers of daughters, our
findings suggest their appeal was felt by fathers of first daughters.
Second, the question of why first daughters and not any and all daughters should
influence men’s political evaluations and choices is one that emerges from our analysis. As we
articulate above, we hypothesize that the experience of having a first daughter has a
transformative influence on men’s views on gender inequality, and we suspect that these
attitudes can be activated and brought to bear in the political realm via carefully crafted
campaign appeals that directly speak to these views and experiences. That being said, we believe
that there is still much left to do to flesh out this theoretical argument and further theory building
and theory testing efforts are required to understand precisely how first daughters shape the
political development of their fathers.
Finally, in addition to contributing to the literature on political socialization, this work
also has important implications for research on gender and vote choice. While we note earlier
Greenlee, Nteta, Rhodes, and Sharrow
22
that scholars have identified several individual-level factors that contribute to male support for
female candidates, much of the literature focuses on reasons why voters do not vote for women
seeking electoral office, or examine the conditions under which in-group solidarity might
motivate female voters to support female candidates. Here we look carefully at the ways in
which gender and gendered roles might operate to boost support for female candidates among
male voters. In doing so, we offer a new point of consideration to a rich and important on-going
discussion about the role of gender in electoral decisions and outcomes.
Greenlee, Nteta, Rhodes, and Sharrow
23
Table 1 Logistic Regression for Presidential Candidate Preferences Among Fathers, 2016
CCES
Pre-Election
Preference for Clinton
(baseline=all other
candidates)
White
-2.880**
(0.855)
Age
-0.0985**
(0.0340)
Income
-0.00839
(0.102)
Education
-0.168
(0.226)
Married
1.263
(0.690)
Party ID
-1.138**
(0.218)
Ideology
-1.892*
(0.910)
Employed
-2.855**
(0.757)
Sister
0.366
(1.020)
Father of Daughter
-1.840
(1.098)
Father of First Daughter
2.760**
(1.015)
Gender Equality Scale
0.516*
(0.200)
Believe Economy Worse
-1.369**
(0.512)
Racial Resentment Scale
-0.106
(0.143)
Percent Female in State
-0.281
(1.100)
Battleground State
-0.265
(0.766)
Constant
30.94
(59.83)
Observations
212
Clusters
Note: These are weighted unstandardized coefficients.
Robust standard errors are in parentheses
* p < .05, ** p < .01, ***p<.001
Greenlee, Nteta, Rhodes, and Sharrow
24
Table 2 – Logistic Regression for Presidential Vote Choice Among Fathers, 2016 CCES
White
Age
Income
Education
Married
Party ID
Ideology
Employed
Sister
Father of Daughter
Father of First Daughter
Gender Equality Scale
Greenlee, Nteta, Rhodes, and Sharrow
25
Believe Economy Worse
Racial Resentment Scale
Percent Female in State
Battleground State
Constant
Pseudo R²
Clusters
Observations
Note: These are weighted unstandardized coefficients.
Robust standard errors are in parentheses
* p < .05, ** p < .01, ***p<.001
Greenlee, Nteta, Rhodes, and Sharrow
26
Table 3 – OLS Regression for Support for Molly Smith Among Fathers, 2016 CCES
Model 1
Model 2
Vote for Smith
Vote for Smith
STEM Treatment
0.0191
0.0238
(0.072)
(0.058)
Clinton Treatment
0.0563
0.0512
(0.060)
(0.044)
Smith Ideology
0.2280
0.2404
(0.129)
(0.129)
PartyID
-0.0378***
-0.0385***
(0.007)
(0.007)
Father of Daughter
-0.0350
N/A
(0.037)
Father of First Daughter
N/A
-0.0699**
(0.028)
Father of (First) Daughter *
Stem
0.0489
0.0629
(0.088)
(0.073)
Father of (First) Daughter *
Clinton
0.0951
0.1619**
(0.092)
(0.064)
Constant
0.6228***
0.6284***
(0.063)
(0.060)
Clusters
43
43
Standard Error
.238
.237
R²
.15
.15
Observations
234
234
Note: These are weighted unstandardized coefficients.
Robust standard errors are in parentheses
* p < .05, ** p < .01, ***p<.001
Greenlee, Nteta, Rhodes, and Sharrow
27
Figure 2 - Predicted Probability of Support for Smith By Exposure to Clinton Treatment
and Fatherhood of First Daughter, 2016 CCES
Note: Figure plots predicted probabilities based on the model 2 in Table 3. Plot shows predicted probability strong
opposition while holding all other variables in the model at their mean values. Shaded area represents 84%
confidence intervals.
.5 .6 .7 .8
Support for Molly Smith
0 1
Exposure to Clinton Appeal
fatherfirstdaughter=0 fatherfirstdaughter=1
Greenlee, Nteta, Rhodes, and Sharrow
28
References:
AAD, Asian American Decisions. 2016. “Asian American Decisions 2016 Election Eve Poll.”
http://asianamericandecisions.com/wp-content/uploads/2016/11/2016-Asian-American-
Election-Eve-Poll-Infographic.pdf (July 10, 2017).
Ansolabehere, Stephen, and Brian Schaffner. 2014. “Does Survey Mode Still Matter? Findings
from a 2010 Multi-Mode Comparison.” Political Analysis 22(3): 285–303.
Arceneaux, Kevin, and David W. Nickerson. 2009. “Modeling Certainty with Clustered Data: A
Comparison of Methods.” Political Analysis 17(2): 177–90.
Box-Steffensmeier, Janet M., Suzanna De Boef, and Tse-min Lin. 2004. “The Dynamics of the
Partisan Gender Gap.” American Political Science Review 98(3): 515–28.
Brooks, Deborah. 2013. He Runs, She Runs: Why Gender Stereotypes Do Not Harm Women
Candidates. Princeton, NJ: Princeton University Press.
Burden, Barry, Yoshikuni Ono, and Masahiro Yamada. 2017. “Reassessing Public Support for a
Female President.” The Journal of Politics 79(3): 1073–78.
Cassese, Erin, and Mirya Holman. 2017. “Religion, Gendered Authority, and Identity in
American Politics.” Politics and Religion 10: 31–56.
Clarke, Harold, and Marianne Steward. 2017. “How Attitudes about Gender May Have Helped
Hillary Clinton in 2016.” Monkey Cage Blog, The Washington Post.
https://www.washingtonpost.com/news/monkey-cage/wp/2017/06/12/how-attitudes-about-
gender-may-have-helped-hillary-clinton-in-2016/?utm_term=.4e94dc9cd72d (July 6, 2017).
Clinton, Hillary. 2016. “Address Accepting the Presidential Nomination at the Democratic
National Convention in Philadelphia, Pennsylvania.” Online by Gerhard Peters and John T.
Woolley, The American Presidency Project.
http://www.presidency.ucsb.edu/ws/index.php?pid=118051.
CNN. 2016. “2016 Exit Polls.” http://www.cnn.com/election/results/exit-polls (July 10, 2017).
Cronqvist, Henrik, and Frank Yu. 2016. “Shaped by Their Daughters: Executives, Female
Socialization, and Corporate Social Responsibility.” Journal of Financial Economics
Forthcomin.
Cuevas-Molina, Ivelisse. 2017. “Voter Turnout Overreports: Measurement, Modeling and
Deception.” Ph.D. diss. University of Massachusetts Amherst.
Dahl, Michael, Cristian Dezső, and David Ross. 2012. “Fatherhood and Managerial Style: How a
Male CEO’s Children Affect the Wages of His Employees.” Administrative Science
Quarterly 57(4): 669–93.
Dickerson, John. 2016. “Hillary Clinton’s Fight for the Dad Vote.” CBS News.
http://www.cbsnews.com/news/hillary-clintons-fight-for-the-dad-vote/ (July 10, 2017).
Dinas, Elias. 2013. “Opening ‘Openness to Change’: Political Events and the Increased
Sensitivity of Young Adults.” Political Research Quarterly 66(4): 868–82.
Dittmar, Kelly. 2017. Finding Gender in Election 2016: Lessons from Presidential Gender
Watch. New Brunswick, NJ: Center for American Women and Politics, Eagleton Institute of
Politics, Rutgers, the State University of New Jersey, with the Barbara Lee Family
Foundation. http://presidentialgenderwatch.org/wp-content/uploads/2017/05/Finding-
Gender-in-Election-2016.pdf.
Dolan, Kathleen. 2010. “The Impact of Gender Stereotyped Evaluations on Support for Women
Candidates.” Political Behavior 32(1): 69–88.
Greenlee, Nteta, Rhodes, and Sharrow
29
———. 2014. When Does Gender Matter?: Women Candidates and Gender Stereotypes in
American Elections. New York: Oxford University Press.
Ehrenfreund, Max, and Scott Clement. 2016. “Economic and Racial Anxiety: Two Separate
Forces Driving Support for Donald Trump.” The Washington Post: 2016–18.
Frizell, Sam. 2015. “Transcript: Read the Full Text of Hillary Clinton’s Campaign Launch
Speech.” Time. http://time.com/3920332/transcript-full-text-hillary-clinton-campaign-
launch/ (August 10, 2017).
Gerber, Alan, James Gimpel, Donald Green, and Daron Shaw. 2011. “How Large and Long-
Lasting Are the Persuasive Effects of Televised Campaign Ads? Results from a
Randomized Field Experiment.” American Political Science Review 105(1): 135–50.
Glynn, Adam, and Maya Sen. 2015. “Identifying Judicial Empathy: Does Having Daughters
Cause Judges to Rule for Women’s Issues?” American Journal of Political Science 59(1):
37–54.
Goldstein, Kenneth, and Travis Ridout. 2004. “Measuring the Effects of Televised Political
Advertising in the United States.” Annual Review of Political Science 7: 205–26.
Green, Donald P., and Lynn Vavreck. 2008. “Analysis of Cluster-Randomized Experiments: A
Comparison of Alternative Estimation Approaches.” Political Analysis 16(2): 138–52.
Greenlee, Jill. 2014. The Political Consequences of Motherhood. Ann Arbor, MI: University of
Michigan Press.
Healy, Andrew, and Neil Malhotra. 2013. “Childhood Socialization and Political Attitudes:
Evidence from a Natural Experiment.” The Journal of Politics 75(4): 1023–37.
Huddy, Leonie, Erin Cassese, and Mary-Kate Lizotte. 2008. “Gender, Public Opinion, and
Political Reasoning.” In Political Women and American Democracy, eds. Christina
Wolbrecht, Karen Beckwith, and Lisa Baldez. New York: Cambridge University Press, 31–
49.
Jones, Bradley, and Jocelyn Kiley. 2016. “More ‘warmth’ for Trump among GOP Voters
Concerned by Immigrants, Diversity.” Pew Research Center.
http://www.pewresearch.org/fact-tank/2016/06/02/more-warmth-for-trump-among-gop-
voters-concerned-by-immigrants-diversity/ (July 10, 2017).
Junn, Jane. 2017. “The Trump Majority: White Womanhood and the Making of Female Voters
in the U.S.” Politics, Groups, and Identities 5(2): 343–52.
Kaufmann, Karen M. 2002. “Culture Wars , Secular Realignment , and the Gender Gap in Party
Identification.” 24(3): 283–307.
Kinder, Donald, and Lynn Sanders. 1996. Divided by Color: Racial Politics and Democratic
Ideals. Chicago, IL: University of Chicago Press.
Klinkner, Philip. 2016. “The Easiest Way to Guess If Someone Supports Trump? Ask If Obama
Is a Muslim.” Vox. https://www.vox.com/2016/6/2/11833548/donald-trump-support-race-
religion-economy (July 6, 2017).
Knezevic, Andrea. 2008. “Overlapping Confidence Intervals and Statistical Significance.”
StatNews: Cornell University Statistical Consulting Unit 73(1).
Krasno, Jonathan S., and Donald P. Green. 2008. “Do Televised Presidential Ads Increase Voter
Turnout? Evidence from a Natural Experiment.” The Journal of Politics 70(1): 245–61.
Ladd, Jonathon McDonald. 2010. “Power of Elite Opinion Leadership The Neglected Toward
the News Media: To Produce Antipathy Evidence from a Survey Experiment.” Political
Behavior 32(1): 29–50.
LaRossa, Ralph, and Maureen Mulligan LaRossa. 1981. Transitions to Parenthood: How Infants
Greenlee, Nteta, Rhodes, and Sharrow
30
Change Families. Beverly Hills, CA: SAGE.
Lau, Richard, and Ivy Brown Rovner. 2009. “Negative Campaigning.” Annual Review of
Political Science 12: 285–306.
LD, Latio Decisions. 2016. “Latino Decisions 2016 Election Eve Poll.”
http://www.latinodecisions.com/files/8614/7866/3919/National_2016__Xtabs.pdf (July 10,
2017).
MacWilliams, Matthew. 2016. “The Best Predictor of Trump Support Isn’t Income, Education,
or Age. It’s Authoritarianism.” Vox. https://www.vox.com/2016/2/23/11099644/trump-
support-authoritarianism (July 10, 2017).
Manza, Jeff, and Ned Crowley. 2017. “Working Class Hero? Interrogating the Social Bases of
the Rise of Donald Trump.” The Forum 15(1): 3–28.
McDermott, Monika. 1998. “Race and Gender Cues in Low-Information.” Political Research
Quarterly 51(4): 895–918.
Mendelberg, Tali. 2001. The Race Card: Campaign Strategy, Implicit Messages, and the Norm
of Equality. Princeton, NJ: Princeton University Press.
NewYorkTimes. 2015. “Full Transcript: Democratic Presidential Debate.” New York Times.
https://www.nytimes.com/2015/10/14/us/politics/democratic-debate-transcript.html (July 6,
2017).
Nystrom, Kerstin, and Kerstin Ohrling. 2004. “Parenthood Experiences during the Child’s First
Year: Literature Review.” Journal of Advanced Nursing 46(3): 319–30.
NYT, Editorial Board. 2016. “Hillary Clinton for the Democratic Nomination.” The New York
Times. https://www.nytimes.com/2016/01/31/opinion/sunday/hillary-clinton-
endorsement.html?_r=0 (July 11, 2017).
Oswald, Andrew J, and Nattavudh Powdthavee. 2010. “Daughters and Left-Wing Voting.” The
Review of Economics and Statistics 92(2): 213–27.
Palkovitz, Robin, and Glen Palm. 2009. “Transitions within Fathering.” Fathering 7(1): 3–22.
Palkovitz, Robin, and Marvin B. Sussman. 1988. Transitions to Parenthood. Psychology Press.
Paul, David, and Jessi L. Smith. 2008. “Subtle Sexism? Examining Vote Preferences When
Women Run Against Men for the Presidency.” Journal of Women, Politics & Policy 29(4):
451–76.
Payton, Mark, and Matthew Greenstone. 2003. “Overlapping Confidence Intervals or Standard
Error Intervals: What Do They Mean in Terms of Statistical Significance?” Journal of
Insect Science 3(1): 34.
Poloni-Staudinger, Lori, J. Cherie Strachan, and Brian Schaffner. 2016. “In 6 Graphs, Here’s
Why Young Women Don’t Support Hillary Clinton as Much as Older Women Do.” Monkey
Cage Blog, The Washington Post. https://www.washingtonpost.com/news/monkey-
cage/wp/2016/04/11/in-6-graphs-heres-why-young-women-dont-support-hillary-clinton-as-
much-as-older-women-do/?utm_term=.c00a2c907e62 (July 10, 2017).
Price, Joseph. 2008. “Parent-Child Quality Time: Does Birth Order Matter?” Journal of Human
Resources 43(1): 240–65.
Ramakrishnan, Karthick. 2016. “Trump Got More Votes from People of Color than Romney
Did. Here’s the Data.” Monkey Cage Blog, The Washington Post.
https://www.washingtonpost.com/news/monkey-cage/wp/2016/11/11/trump-got-more-
votes-from-people-of-color-than-romney-did-heres-the-data/?utm_term=.8b625e45812a
(July 19, 2017).
Roberts, David. 2015. “Are Trump Supporters Driven by Economic Anxiety or Racial
Greenlee, Nteta, Rhodes, and Sharrow
31
Resentment? Yes.” Vox. https://www.vox.com/2015/12/30/10690360/racism-economic-
anxiety-trump (July 6, 2017).
Sanbonmatsu, Kira. 2002. “Gender Stereotypes and Vote Choice.” American Journal of Political
Science 46(1): 20–34.
Sanbonmatsu, Kira, and Kathleen Dolan. 2009. “Do Gender Stereotypes Transcend Party?”
Political Research Quarterly 62(3): 485–94.
Schaffner, Brian, Matthew MacWilliams, and Tatishe Nteta. 2017. Explaining White
Polarization in the 2016 Vote for President: The Sobering Role of Racism and Sexism.
Working Paper.
Schuman, Howard, and Jacqueline Scott. 1989. “Generations and Collective Memories.”
American Sociological Review 54(3): 359–81.
Sears, David. 1981. “Life Stage Effects upon Attitude Change, Especially among the Elderly.” In
Aging: Social Change, eds. S.B. Kiesler, J.N. Morgan, and V.K. Oppenheimer. New York:
Academic Press, 183–204.
Sears, David, and Carolyn Funk. 1999. “Evidence of the Long-Term Persistence of Adults’
Political Predispositions.” The Journal of Politics 61(1): 1–28.
Sears, David, and Sheri Levy. 2003. “Childhood and Adult Political Development.” In Oxford
Handbook of Political Psychology, eds. David Sears, Leonie Huddy, and Robert Jervis.
New York: Oxford University Press, 60–109.
Shafer, Emily Fitzgibbons, and Neil Malhotra. 2011. “The Effect of a Child’s Sex on Support for
Traditional Gender Roles.” Social Forces 90(1): 209–22.
Shapiro, Robert, and Harpreet Mahajan. 1986. “Gender Differences in Policy Preferences: A
Summary of Trends from the 1960s to the 1980s.” Public Opinion Quarterly 50(1): 42–61.
Sharrow, Elizabeth et al. 2016. “Gender Attitudes, Gendered Partisanship: Feminism and
Support for Sarah Palin and Hillary Clinton among Party Activists.” Journal of Women,
Politics & Policy 37(4): 394–416.
Sharrow, Elizabeth, Jesse Rhodes, Tatishe Nteta, and Jill Greenlee. 2017. The First Daughter
Effect: The Impact of Fathering First Daughters on Men’s Preferences on Gender Equality
Issues. Presented at the European Conference on Politics and Gender, June 8-10, 2017.
Sigelman, Lee, and Carol K Sigelman. 1982. “Sexism, Racism, and Ageism in Voting Behavior:
An Experimental Analysis.” Social Psychology Quarterly 45(4): 263–69.
Stonecash, Jeffrey M. 2017. “The Puzzle of Class in Presidential Voting.” The Forum 15(1): 29–
49.
Strolovitch, Dara Z., Janelle S. Wong, and Andrew Proctor. 2017. “A Possessive Investment in
White Heteropatriarchy? The 2016 Election and the Politics of Race, Gender, and
Sexuality.” Politics, Groups, and Identities 5(2): 353–63.
Tesler, Michael. 2016a. “A Key Reason Young People Don’t Support Hillary Clinton? They
Don’t Have Daughters.” Monkey Cage Blog, The Washington Post.
https://www.washingtonpost.com/news/monkey-cage/wp/2016/02/11/a-key-reason-young-
people-dont-support-hillary-clinton-they-dont-have-daughters/?utm_term=.69017896f20f
(July 10, 2017).
———. 2016b. “A Striking Poll Shows That Many Trump Supporters Already Doubted
Women’s Claims of Sexual Harassment – Even Before the Infamous Tape.” Monkey Cage
Blog, The Washington Post.
———. 2016c. “Parents of Daughters Support Hillary Clinton More than Parents of Sons.”
Monkey Cage Blog, The Washington Post. https://www.washingtonpost.com/news/monkey-
Greenlee, Nteta, Rhodes, and Sharrow
32
cage/wp/2016/01/05/parents-of-daughters-support-hillary-clinton-more-than-parents-of-
sons/?utm_term=.1060e2895c02 (July 6, 2017).
———. 2016d. “Views about Race Mattered More in Electing Trump than in Electing Obama.”
Monkey Cage Blog, The Washington Post. https://www.washingtonpost.com/news/monkey-
cage/wp/2016/11/22/peoples-views-about-race-mattered-more-in-electing-trump-than-in-
electing-obama/?utm_term=.5683cd0a9749 (July 10, 2017).
———. 2016e. “Why the Gender Gap Doomed Hillary Clinton.” Monkey Cage Blog, The
Washington Post.
Valentino, Nicholas, Vincent Hutchings, and Ismail White. 2002. “Cues That Matter: How
Political Ads Prime Racial Attitudes during Campaigns.” American Political Science
Review 96(1): 75–90.
Warner, Rebecca. 1991. “Does the Sex of Your Children Matter? Support for Feminism among
Women and Men in the United States and Canada.” Journal of Marriage and Family 53(4):
1051–56.
Warner, Rebecca, and Brent Steel. 1999. “Child Rearing As A Mechanism for Social Change:
The Relationship of Child Gender to Parents’ Committment to Gender Equity.” Gender &
Society 13(4): 503–17.
Washington, Ebonya. 2008. “Female Socialization: How Daughters Affect Their Legislator
Fathers’ Voting on Women’s Issues.” American Economic Review 98(1): 311–32.
Wolbrecht, Christina. 2000. The Politics of Women’s Rights: Parties, Positions, and Change.
Princeton, NJ: Princeton University Press.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
We re-deploy a list experiment conducted a decade ago to reassess the degree to which the American public opposes electing a woman as president. We find that opposition has been cut in half from approximately 26% to 13%. In addition, opposition is now concentrated in specific sociodemographic categories rather than being evenly distributed. Newly developed statistical methods that permit multivariate analysis of list experiment data reveal that resistance has all but disappeared among Democratic-leaning groups in the electorate. These patterns appear to reflect the reduction of uncertainty among groups most favorable toward the recent success of Democratic women.
Article
Full-text available
Activists in the Democratic and Republican parties have distinct concerns about women?s place in American politics and society. These views lead them to evaluate female candidates through different ideological lenses that are conditioned, in part, on their divergent attitudes about gender. We explore the implications of these diverging lenses through an examination of the 2008 candidacies of Hillary Clinton and Sarah Palin, using data from an original survey of Democratic and Republican National Convention delegates. We find that delegate sex did not affect their evaluations but that evaluations were influenced by the interaction of partisanship and attitudes about women?s roles.
Chapter
Full-text available
Men and women differ in their political attitudes and behavior, but these differencesare modest and inconsistent (Sapiro 2003). The much-discussed gender gap in voting choice and partisan preference in which women identify more strongly with the Democratic Party and give greater electoral support to Democratic candidates is real, persistent, and consequential. However, it is also modest in size, with women and men differing in their support for Democratic presidential and congressional candidates by 8 to 10 percentage points on average. This difference is small compared with other political differences across demographic groups such as race, in which the gap in vote choice between blacks and whites is closer to 40 percentage points (Tate 1994). There are also small or inconsistent differences between men and women in many areas of public opinion, leaving researchers to analyze various “gender gaps” of differing origins, in addition to many areas of public opinion in which there are simply no such differences (Schlesinger and Heldman 2001). In reality, men and women both differ and converge politically ininteresting ways that deserve the scrutiny of empirical researchers (for a similar point aboutdifferences between male and female politicians, see Reingold, this volume). Differences between men and women are slight, but that does not necessarily neutralize their political power.A difference of even 8 to 10 percentage points can determine the outcome of elections because women represent a larger segment of the voting population than do men.
Article
The conventional wisdom is that class divisions once prevailed but in recent decades have gradually declined. Indeed, many now suggest that the working class has been voting Republican since the 1980s. The historical evidence on voting in presidential elections does not indicate that there was a decline since the 1950s. If anything, the well-off and more educated have moved somewhat more Democratic, lessening class divisions. There has, however, been a change since the 2008 election, with the working class – whether only whites or all – moving away from the Democrats. This may not mean a decline of the working class voting their interests. It may be that the economic recovery of recent years has done little to help the working class and they have taken a gamble that a businessman will help their job prospects more.
Article
The estimated 52% of white female voters who supported Donald Trump for President of the United States in 2016 animates this dialogue on the politics of groups and identities. The Trump majority among white women exists in contrast to strong support of Democratic Party candidate Hillary Clinton among women of color as well as minority men. While many observers were surprised at the high proportion of white female Trump voters, this pattern of electoral behavior supporting Republican Party candidates is a consistent phenomenon since the 1950s in U.S. Presidential elections. The pattern is both clear and easily visible, and provides an important clue to better understand the dynamics of race and gender in electoral politics. I argue for political scientists to pursue intersectional analytical perspectives by situating voting behavior within the context of a polity beyond the black-white binary of race and ethnicity. I suggest analysts consider the positionality of white women as second in sex to men, but first in race to minorities, and the invocation of white womanhood in political rhetoric and practice as a potential explanation of the Trump majority.
Article
The 2016 election has triggered new interest in and speculation about longstanding questions about the roles of race, gender, and sexuality in American politics. We argue that rather than anomalous and exceptional, the 2016 election represents an extension – and perhaps the beginning of a consolidation – of enduring and intersecting configurations of racialized and gendered power, marginalization, and oppression. We examine some of the ways in which these intersecting configurations structure and are structured by American politics, exploring some of the political consequences of proximity to or distance from the benefits of white heteropatriarchy.
Article
We present a political sociological analysis of the social bases of support for Donald Trump during the critical phase of his victory in the Republican nominating contest. In particular, we test the widely voiced hypothesis that a critical source of Trump’s support in the GOP primaries came from his appeal to working class and/or downwardly mobile and insecure middle class voters responding to a “populist” message. Drawing on both the ANES January 2016 pilot survey and exit poll data, we argue that Trump’s rise to the GOP nomination was facilitated by a broad-based appeal that centered on voters who have levels of education and income that are well above national and primary state averages.
Article
Religious identity serves as a central cleavage in American politics. However, little attention has been granted to how gendered views of authority conveyed in religious doctrine shape political identities and attitudes. Using a nation-wide sample of adult Americans, we demonstrate that gendered notions of divine and human authority exert considerable influence on political thinking. In particular, belief in a masculine God and preferences for traditional gender roles strongly relate to political conservatism. Adherence to gendered notions of authority influences political identity and policy preferences, even when controlling for more conventional indicators of religiosity. Accounting for gendered beliefs about authority also partially explains well-documented gender gaps in American politics, providing insight into women's apparently contradictory tendencies toward both political liberalism and religiosity. The relationships uncovered here, coupled with the continued salience of both gender and religion in contemporary political campaigns, underscore the importance of attending to the gendered dimensions of authority.
Article
Gender differences in vote choice, opinion, and party identification have become a common feature of the American political landscape. We examine the nature and causes of gender differences in partisanship using a time series approach. We show that gender differences are pervasive-existing outside of the context of specific elections or issues-and that they are a product of the interaction of societal conditions and politics. We find that from 7979 to 2000, the partisan gender gap has grown when the political climate moved in a conservative direction, the economy deteriorated, and the percentage of economically vulnerable, single women increased. The gender gap is likely to be a continual feature of the American political landscape: one that shapes everything from elite political behavior to election outcomes.