The Different Effects of Gender and Sex on Vote Choice

• Author: J.Michael Bailey,Rose McDermott,Peter K. Hatemi,Nicholas G. Martin
• Date: 01 March 2012
• DOI: 10.1177/1065912910391475
• Published date: 01 March 2012
• Subject Matter: Articles

Political Research Quarterly
65(1) 76 –92
The Different Effects of Gender
© 2012 University of Utah
Reprints and permission:
sagepub.com/journalsPermissions.nav
and Sex on Vote Choice
DOI: 10.1177/1065912910391475
http://prq.sagepub.com
Peter K. Hatemi1, Rose McDermott2, J.Michael Bailey3,
and Nicholas G. Martin4
Abstract
Explanations for differences in political preferences between men and women continue to be debated, generating
more heat than light in attempts to locate their source and potential influence. The reason for this confusion rests on
the lack of conceptual clarity concerning the difference between sex, typically referring to biological differences, and
gender, assumed to result from socialization, and the difference these constructs might elicit in political outcomes.
Utilizing two gender scales, the authors find gender identity exerts an impact on voter preferences above and beyond
sex. They also find that individual differences in gender identity are not found to result from social influences but
largely derive from unique experiences and innate disposition. The results have substantial implications for social
scientists who theorize about and investigate sex and gender in studies of political attitudes and behaviors.
Keywords
gender, genetics, voting, identity, personality
Men and women often exhibit predictable differences
and Steffen 1986; Jennings 2006). This perspective over-
in political attitudes and behavior. Such divergences
looks contemporary findings in the medical sciences
have formed the basis of demographic predictions in vot-
(e.g., Bocklandt and Vilain 2007) that reveal that gender
ing behavior since Lazarsfeld’s early work (Berelson,
identity results in large part from biological influences,
Lazarsfeld, and McPhee 1954), although the common
including in utero effects, and does not derive from the
phrase gender gap did not enter the discourse until much
effects of socialization. Surprisingly little interrogation
later (Frankovic 1981). Indeed, much of the political dis-
has surrounded the ways in which either concept might
cussion concerning past elections surrounds the demo-
differentially inform or affect political behavior. Here we
graphics of how many, and what kind of, women supported
seek to help fill that lacunae and advance the discussion
particular kinds of candidates. This was most evident dur-
by differentiating some of the ways in which sex and gen-
ing the 2008 presidential election surrounding the demo-
der produce different sources of variation in political
graphics of which women supported Hillary Clinton or
behavior. We also explore the sources of gender and
Sarah Palin.
examine what an explicit distinction between the con-
A great deal of scholarship in both political science
cepts of sex and gender might mean for political behav-
and psychology has addressed the sources and conse-
ior. In doing so, we hope to gain some empirical grounding
quences of discrepancies in attitudes and behavior for the ways in which gender differs from sex and why
between men and women. But markedly few discussions
these differences might establish more conceptually valid
explore potential distinctions between sex and gender in
models explaining attitudinal and behavioral differences
explaining either between-sex or within-sex differences
in both men and women.
in political preferences. In discussions where the con-
cepts remain distinct, sex typically refers to biological
1University of Sydney, Sydney, Australia
reproductive status, while the notion of gender embraces
2Brown University, Providence, RI, USA
a broad spectrum of dispositional differences and socially
3Northwestern University, Evanston, IL, USA
informed attitudes and behaviors. However, in the exist-
4Queensland Institute of Medical Research, Herston, Australia
ing political science literature the meaning and influence
Corresponding Author:
of sex and gender are typically treated synonymously,
Peter K. Hatemi, University of Sydney, United States Studies
and behavioral divergences associated with differences
Centre, John Woolley Building A20, Science Rd.,
between men and women are assumed to result largely, if
University of Sydney, NSW 2006, Australia
not exclusively, from processes of socialization (Eagly
Email: phatemi@gmail.com

---

Hatemi et al.
77
In contrast to the guiding principles of the discipline,
John–Joan was presented as proof that a child’s gender
we do not assume, a priori, that gender wholly results
could be changed without consequence. Thousands of
from processes of social construction. The common con-
“sex reassignments” were performed, in part based on
vention in political behavior research that labels “gender”
Money’s published findings. However, Diamond and
in dichotomous terms reduces a quite complex intersec-
Sigmundson (1997) followed up on John–Joan and
tion of both sex and gender into one discrete term. This
showed the experiment to have actually been a disastrous
practice confounds distinct traits by treating them as
failure (also see Colapinto 2000). Money failed to report
interchangeable and synonymous when there is no empir-
that David Reimer, initially identified in the medical lit-
ical basis for this fusing. Rather, we consider the possibil-
erature as John–Joan, struggled with his imposed gender
ity that individuals can manifest a propensity to either
identity all throughout childhood, and despite the incred-
correspond to their gender stereotype or not, and to vary-
ible lengths to which Money and his parents worked to
ing degrees, and that such differences hold predictive
reengineer David, including immense social pressure and
power for political preferences. Debate continues whether
conditioning to act like and be a “normal” girl, David
gender is a unidimensional construct, where masculinity
ultimately rejected the feminine roles imposed on him.1
and femininity reside on opposite poles of a single con-
At age thirty-eight, David committed suicide. This case
tinuum (Lippa 2005; Rieger et al. 2008; Rieger et al.
clearly demonstrated that no matter the degree of social
2009), or consists of two unique constructs, such that
pressures society exerts, gender or sexuality cannot be
individuals might be categorized separately and simulta-
summarily reassigned. This example also poignantly
neously on each dimension (Bem 1981). However, either
demonstrates the practical and political consequences of
way, masculinity and femininity do not emerge necessar-
imposing deterministic beliefs on people’s lives. If such
ily dichotomous.
work showing the independent biological influences on
It is important to decouple sex from gender in predictions
both sex and gender had been widely known in the 1970s,
of voting behavior because it is not clear that they predict
demonstrating that gender does not emerge from social
the same outcome. Is a more masculine woman more
conditioning, people like John Money could have been
similar in voting habits to a man or a woman? Is a more
stopped from the reckless experimentation that resulted
feminine man more likely to be a political nonconformist?
in the destruction of many lives.
The vast majority of political behavior research implicitly
The example above is important because although we
assumes that sex and gender represent commensurate and
have no desire to reenergize an atavistic debate concern-
interchangeable measures for predicting political prefer-
ing the artificial divide between nature and nurture, we do
ences or behaviors. Indeed, scholars typically explore sex
seek to explore the comingling of these forces that mani-
as a predictor of political attitudes and preferences and
fest in predictable psychological and political outcomes.
then cease the exploration there. Research has largely
If gender exerts an important political effect independent
neglected the way in which sex and gender may influence
of sex, then most of our extant measures and models of
outcomes in divergent ways. However, this conundrum
political behavior based on biological sex remain impre-
has received renewed attention; a recent report to the
cise, and possibly inaccurate, in their interpretation. This
American National Election Studies proposed the intro-
implication holds real-world applications because if dif-
duction of measures to examine the overlap between sex
ferent outcomes result depending on whether gender or sex
and gender (Sanbonmatsu and Dolan 2007). Sex and gender
is used as a variable, then predictions and interventions
may influence political attitudes and behaviors through
for a wide variety of phenomena that presumably differ
different mechanisms, producing divergent outcomes; we
according to biological sex will require reexamination.
seek here to empirically delineate the extent to which this
Here we explore both sides of the interaction between
might prove true.
the biological and environmental influences of sex and
Importantly, it is also critical to identify and distin-
gender on political preferences first by examining the
guish the sources of gender if we are to avoid disastrous
independent impact of gender on vote choice and second
public policy choices and protect the freedoms and rights
by exploring the nature of gender using a genetically
of individuals. One of the best known deterministic
informative sample of twins. We seek to investigate
...