Content uploaded by Marta Fraile
Author content
All content in this area was uploaded by Marta Fraile on Oct 15, 2021
Content may be subject to copyright.
1
1
Title:
Risk and the gender gap in internal political efficacy in Europe
Authors
Marta Fraile (marta.fraile@csic.es) ORCID ID = 0000-0002-4123-2874
Consejo Superior de Investigaciones Científicas – Instituto de Bienes y Políticas Públicas,
Madrid (Spain) **Corresponding autor**
Carolina de Miguel (carol.demiguel@utoronto.ca) ORCID ID = 0000-0002-
3676-3187. University of Toronto
Note: This article would like to honour the memory of my friend and co-author Carolina de
Miguel Moyer, who sadly left us in August 2020.
This research was supported by the Agencia Estatal de Investigación under grant: PID2019-
197445GB)
Published at West European Politics: https://doi.org/10.1080/01402382.2021.1969146
Abstract:
This study confirms the existence of a substantial gender gap in internal political efficacy in
contemporary European democracies with survey evidence from the ESS08 and 09 waves.
This gap is rooted in gendered patterns of socialisation according to which men are more
likely than women to be socialised in ways that emphasise psychological traits such as
assertiveness, predisposition to risk, competition, or self-promotion. Findings show that those
who perceive themselves as ready to take risks are more likely to feel able to play an active
role in politics. Using mediation analysis, this article shows that part of the gender gap in
internal political efficacy is a result of the lower inclination of women to take risks in
comparison to men. These findings confirm the masculine character of the political realm. A
realm that is often perceived by citizens to be more in line with gender stereotypes about
men.
2
2
Introduction
Gender differences in the political realm are ubiquitous. Men tend to be more likely
than women to engage in partisan political participation (Quaranta and Sani 2018). Women
also tend to declare less interest in politics than men (Karp and Banducci 2008; Kittilson and
Schwindt-Bayer 2012), and in most advanced industrialized democracies women appear to be
less knowledgeable about political matters than men (Fortin-Rittberger 2016; Marinova and
Anduiza 2020). While scholars have paid a great deal of attention to gender gaps in these
forms of political involvement, citizens’ orientations and feelings regarding the political
world such as political efficacy, have been scarcely studied.
Political efficacy refers to citizens’ feelings and perceptions of their influence in the
political realm. It has two relevant dimensions: external and internal (Balch 1974). While
external efficacy reflects feelings of being heard by the political system, internal efficacy
refers to ideas of personal competence and the extent to which citizens feel they have the
capacity to understand politics and to act politically.
We focus our attention in this article on internal political efficacy, because it is a key
precursor to many other forms of political participation. Social cognitive theory proposes that
people’s beliefs about their own ability to complete specific tasks is vital to understand why
people engage in different enterprises, how much determination they put towards such
activities, and how long they will persist in the face of obstacles and failures (Bandura 1997).
Similarly, scholars have found that in the political sphere citizens’ perceptions about their
ability to influence political matters (i.e. a sense of internal political efficacy) is key to their
actual participation (Condon and Holleque 2013; Gidengil et al. 2008). Therefore,
3
3
understanding why women are less likely than men to feel politically efficacious, is crucial to
the health of our democracies.
Existing studies have examined the gender gap in internal political efficacy in the
United States and in Canada (Gidengil et al. 2008; Thomas 2012; Wolak 2018), but little is
known beyond these two cases. The first contribution of this paper is to show that the gender
gap in internal political efficacy is an enduring feature in contemporary European
democracies. Women consistently report lower levels of internal political efficacy than men
across a diverse set of 27 European countries, and this gender gap does not disappear when
considering individual-level socio-demographic and attitudinal variables that are typically
associated with political engagement. The second contribution of this study is to explain the
pervasiveness of this gender gap using a goal congruity psychological framework.
Our argument is that differences between men and women in self-perceptions of their
willingness to take risks in life account for part of the gap in feelings of internal political
efficacy. Risk constitutes one key component of the agentic model that is typically attached
to men according to social role theory. Citizens’ inclination to risk is key to understand how
individuals relate to the political world because the media often portray (and citizens
habitually perceive) politics as linked to some risk. Consequently, we expect individuals who
feel ready to take risks in their life to experience more affinity with the political world, which
should make them more likely to feel able to play an active role in politics.
Prior research in psychology and economics shows that men are more inclined to take
risks than women. Since the political realm is often related to risk by both men and women,
men are more likely to feel that the political domain is a welcoming setting, a comfort zone
4
4
where they can use their own capabilities to connect with the activities and topics developed
in this sphere. Women are instead more likely to encounter the political sphere as an
unwelcoming environment for which they will never feel sufficiently prepared.
We test our argument using survey data from the 8th and 9th wave of the European
Social Survey, which are unique because they contain two specific measures of internal
political efficacy, as well as an item measuring the way in which respondents perceive their
willingness to take risks in their life. Our findings show that those who perceive themselves
as ready to take risks are more likely to feel able to play an active role in politics. In addition,
and using mediation analysis, we show that part of the gender gap in internal political
efficacy is a result of the lower inclination of women to take risks in comparison to men.
More specifically, we find that the size of the gender gap in internal political efficacy would
decrease around ten percentage points if women perceived themselves to be as ready to take
risks as men do.
This study adds to an emerging line of research suggesting that the application of
social role theory to the study of gender and politics can contribute to fruitful advances in the
comprehension of the intricate ways in which gender shapes politics around the world
(Schneider and Bos 2019). Our findings have implications for how we socialize girls and
boys given that gendered social roles and stereotypes are developed and nurtured early in life.
A final implication of our findings is connected to how the media tend to portray politics, and
how elections focus on single candidates (often men) and on formats (such as debates) that
tend to favour stereotypically masculine attributes such as belligerence, competitiveness and
riskiness. We discuss these implications in the concluding section.
5
5
The argument: the inclination to take risks and the gender gap in internal political
efficacy
Prior studies have demonstrated the importance of socioeconomic resources --
education, income and participation in the workforce --- to explain why women feel less able
to understand and participate in politics (Gidengil et al. 2008, Thomas 2012). They have also
pointed to implicit gendered socialization rules dictating that women are those responsible for
raising children in the family, which leads them to giving less priority and space to political
issues in their daily lives. The lack of contact with the political realm creates a sense of
distance with respect to the political world and nurtures the sensation n that politics is too
complicated and requires a substantial investment in time, which is usually less available for
women, given their caring duties. Contextual factors also seem to matter to explain the
magnitude of the gender gap in political efficacy. Focusing on the institutional variation
across US states, Wolak 2018 shows that the gender gap declines as women’s representation
rises to the point that the gender gap vanishes in states where women comprise at least 26
percent of the legislature.
We contribute to the literature studying gender gaps in public opinion and political
engagement by incorporating social role theory to the explanation of the existence of a gap in
self perceptions of abilities to play an active role in politics. This theory suggests that existing
segregated social roles of women and men create gender stereotypes and gendered values,
opinions and behaviours. Gendered traits and stereotypes are the product of the differential
roles that women and men occupy in society. These roles are traditionally based on biological
and physical dissimilarities between men and women that are reinforced through the
socialization process: as children develop their gender beliefs are nurtured by families, the
media, schools, the toys market, etc. perpetuating gender roles and the expectations to behave
6
6
in ways traditionally associated with their biological sex (Eagly and Wood 2012): those
adopting their stereotypical traits and roles (agentic for men and communal for women) are
socially rewarded while those deviating from such traits and roles are penalized.
Deeply entrenched to social role theory, the role congruity model suggests that
citizens are stimulated to pursue those goals that better align with their gender roles
(Diekman et al. 2017) so that those attached to their gender role (as men or women) will have
more chances to embrace ideas, motivations and aspirations linked to such gender role. This
theory has been used to explain the existence of prejudice towards female leaders on the
grounds of a lack of consistency between the characteristics associated with the female
gender stereotype and those associated with the typical leadership (Eagly and Karau 2002),
and more recently to explain the gender gap in STEM (Diekman et al. 2017) and in political
ambition (Schneider et al. 2016). We build on scholarship showing the utility of applying
social role theory to the study of gender and politics (Schneider and Bos 2019) by using the
role congruity model to complement existing explanations of the pervasive gender gap in
internal political efficacy.
We posit that women perceive political participation as involving specific charges
requiring traits and abilities that are not stereotypically linked with communality. As a
consequence, women feel that there is not a good fit between what they think are their
abilities and those required to be effective in the realm of politics, and this explains their
lower levels of political efficacy in comparison to men. We focus on self-perception of
people’s willingness to take risks in their life for two reasons. First, because willingness to
take risk in life constitutes one of the traits that are more typically attached to the agentic
model according to the social role theory (Schrock and Schwalbe, 2009). And second,
7
7
because politics in contemporary societies is often portrayed (by the media and by politicians
themselves) as a risky and competitive enterprise, where self-promotion is relevant
(Schneider et al. 2016). These perceptions of the political realm are shared by both men and
women, and are rooted in a long story of male domination of the political culture (Lawless
and Fox 2010) and preserved by the media. In an study based on a representative online
survey of the Spanish population collected in June 2020, we asked to what extent participants
identify the political realm with personal promotion (44% of total respondents), adoption of
laws (42%), achievement of agreements (25%), competition (57%), and service to citizens
(18%). What is more, 75% of respondents agreed with the affirmation that “getting involved
in politics always implies taking some risks”.
1
Prior scholarship has argued that political participation poses a variety of risks
connected to the uncertainty of outcomes: the financial, temporal or psychological resources
citizens invest in politics can be wasted in the end if the political action does not provide the
expected gratification to the participants or when their policy demands are not reached (Kam,
2012: 818). Of course, citizens’ perception of risk will vary depending on the type of
political participation we might look at. Still when asked in a survey about their capacity to
take an active role in a group involved with political issues or their ability to participate in
politics, citizens might think about the most typical kind of participation involving an active
engagement such as being a member of a trade union or a political party, attending
demonstrations or, more generally, protesting. What is more relevant, engaging in politics
entails some vulnerability that is more likely to be felt by women than by men: Conversing
about political events, participating in assemblies, sharing news and information about calls
1
Total sample of 1500 participants. No gender differences were relevant for the distribution of responses to any
of the mentioned survey items except for the case of competition (with 62% of total women versus 52% of total
men respondents).
8
8
for political action opens oneself up to the scrutiny, expectations and the demands of others.
Scholarship has shown that in all those situations women tend to be more silent than men,
probably because they feel their voices are less authoritative (Karpowitz and Mendelberg
2014).
We expect individuals who perceive themselves as inclined to take risks to feel more
affinity with the political world, which should make them more likely to feel able to
participate in politics. In turn, we argue that this link between perceptions of risk and internal
political efficacy helps explain why women consistently feel less qualified to participate in
politics than men. This is because women are socialised to perceive themselves to be less
willing to take risks in life than men.
Scholarship has showed that women and men differ significantly in how they perceive
themselves in terms of their general capacities and attributes. For example, recent
experimental evidence shows that boys and girls present different prejudices and expectations
regarding the areas they think they are best skilled at (Lian and Cimpian, 2017). This study
shows that the acquisition of gender stereotypes associating higher levels of cognitive ability
with boys happens quite early (at the age of 6), indicating that the social norms and
stereotypes attached to gender are learned from early childhood. But more pointedly for the
purpose of our argument, adult men and women differ in how they see themselves in terms of
willingness to take risks. Several studies within the psychology and economics literature have
shown that self-perceptions of readiness to take risks vary by gender, and that these
differences explain a variety of social and economic behavior (Weber et al. 2002; Byrnes et
al. 1999). Similarly, there is political science literature showing that women and men
9
9
differently perceive their personal attributes and capacities. For example, Kam (2012) shows
that women perceive themselves as less willing to take risks than men.
To recapitulate, social role theory suggests that men are more likely than women to be
socialized in ways that emphasize traits linked to the agentic model such as assertiveness,
predisposition to risk, competition, self-promotion, focus on the self, etc. As a consequence,
women perceive themselves as being less keen to take risks than men. Since politics is often
related to risk by men and women alike, women will tend to experience less alignment of
their personal qualities with the perceived characteristics of the political domain. This greater
misalignment contributes to explaining part of the gender gap in feelings of efficacy. In other
words, we propose that the risky nature of the political realm is more in line with gender
stereotypes about men: politics is perceived as more welcoming to the attributes that men
appraise themselves to have to a greater extent than women.
We evaluate this expectation using survey data from 27 European countries. Figure 1
summarizes what we call the mediation hypothesis, which suggests that part of the positive
association between being a man and feeling more politically efficacious is due to men’s
higher propensity to self-perceive as ready to take risks in comparison to women.
Figure 1: The mediation hypothesis: Gender differences in self perceptions of risk and
the gender gap in internal political efficacy
Self-perceptions of
willingness to take risks
risks riskbeing risk
loving
Men
Internal Political Efficacy
10
10
Data, measurement, and empirical strategy
We use survey data from the last two waves of the European Social Survey (ESS)
collected between 2016 and 2017 (8th wave) and between 2018 and 2019 (9th wave).
2
These
surveys offer a unique opportunity to explore our argument because they both contain two
items measuring internal political efficacy, as well as the human values scale, which includes
21 items that tap into a variety of personal attributes related to being helpful, taste for risk,
successful, fun etc. Previous waves of the ESS (waves 1 to 7) also asked the items of the
human values scale, but these items were administered through self-completion mode (or
given to the respondent and collected when completed), and consequently response rates for
these items were much lower than for the rest of items. For example, in the 1st ESS wave, the
percentage of missing responses increased around 11 in comparison to the average face-to-
face response. In short, we use the last two available waves of the ESS because they both
collect responses to the human values scale face to face and also contains items measuring
internal political efficacy. The dataset that we analyze includes a total of 27 countries for all
estimations showed in this article (for the list of countries covered, see Table 1 in the
Appendix).
To measure feelings of internal political efficacy, the ESS08/9 includes two items:
“how able do you think you are to take an active role in a group involved with political
issues?” and “how confident are you in your own ability to participate in politics?” Responses
range from 1 = not at all able to 5 = completely able, so higher values indicate greater
feelings of internal political efficacy for both items. In our analysis, we sum responses to the
two items to form a continuous additive scale, ranging from 2 to 10, rescaled from 0 to 8. The
scale reliability coefficient for the two items is 0.81.
2
Data retrieved from: https://www.europeansocialsurvey.org/downloadwizard.
11
11
These two items tap citizens' beliefs about their competence/abilities to play an active
role in politics. However, internal political efficacy also aspires to measure citizens’ beliefs
about their competence to understand the political realm (see for instance Niemi et al. 1991).
Unfortunately, there are no other items tapping respondents’ impressions about their abilities
to understand politics in the abstract. This implies that our findings are limited to the specific
dimension of political efficacy that is linked to political participation. The selection of the
survey items to measure political efficacy in Europe by the ESS research team was justified
on the grounds of cross-cultural equivalence (see Saris and Torcal 2009). We have replicated
the same estimations with evidence coming from the ESS07, where three items were used to
tap internal political efficacy (see Tables 2 and 3 in the Appendix) and results are robust (see
the Discussion and robustness checks section).
One of the main independent variables is gender. Like the majority of large social
surveys, the ESS asks about the biological binary-sex of respondents. We use this
information to operationalise gender, a dichotomous variable with a value of 1 for men and 0
for women.
3
The other key independent variable is self-perceptions of willingness to take
risks in life. We measure it with the following item: “Now I will briefly describe some
people. Please listen to each description and tell me how much each person is or is not like
you: He/She looks for adventures and likes to take risks. He/She wants to have an exciting
life”. The response options range from 1 = not like me at all to 6 =very much like me, so
higher values indicate self-perceptions of greater willingness to take risks. We are confident
that this measure captures personal inclination to take risks because in our data this item is
empirically associated with age (older people are less inclined to take risks: correlation is -
3
We acknowledge that this is a limited way to measure gender given that sex and gender are distinct concepts:
while sex refers to the biological condition linked to genitals and chromosomes, gender refers to comportments
related to belonging to a given sex category (West and Zimmerman 1987). Unfortunately, we had no additional
information to measure the gender of the respondent.
12
12
0.33***) and gender (men are more willing to take risks than women: correlation is 0.13***),
confirming findings from prior research (Kam 2012; Weber et al. 2002).
We also control for a variety of personal, attitudinal and socio-demographic factors.
In addition to our main independent variable of self-perceptions of willingness to take risks,
we include a measure of perceptions of general self-confidence. This measure is also part of
the survey’s human values scale and the specific wording is the following: “it is important to
him/her to make his/her own decisions about what he/she does”. We include this item as a
control because a recent study finds that general self-confidence is associated with various
measures of political engagement in the US case (among which internal political efficacy,
Wolak 2020).
Finally, we control for typical socio-demographics: age (in years), education (in
years), employment status (a dichotomous variable with value 1 for those working and 0
otherwise), and self-reported household income, after tax and compulsory deductions, from
all sources (an ordinal variable ranging from 1 -lowest to 10-highest). Table 1 in the
Appendix provides descriptive information for all variables included in the estimations.
Our empirical strategy consists of two steps. First, we show that there is a significant
gender gap in internal political efficacy across countries and ESS waves (and that it persists
despite usual controls). Having established this gender gap, we then turn to explore its
correlates testing the extent to which self-perceptions of willingness to take risks in life
mediate the association between gender and internal political efficacy. Before conducting
mediation analysis, we first need to show that two conditions hold: first, the inclination to
take risks in life should be greater for men than for women; and second, proclivity to take
13
13
risks (the mediator) should be linked to greater feelings of internal political efficacy
(controlling for gender). If these conditions are met, then we should observe that the size of
the association between gender and internal political efficacy decreases when self-perceptions
of willingness to take risks in life are included in the estimation of internal political efficacy.
4
This last step implies that a percentage of the total association between gender and internal
political efficacy is due to the mediation of self- perceptions of readiness to take risks in life.
Table 1 shows the results of this first set of estimations. All of them include country-level
(not shown) and wave fixed effects to account for any systematic differences in the outcome
variables across countries and wave.
Findings
Figure 2 presents the gender gap in the average responses to the indicator of internal
political efficacy and provides evidence of a consistent gender gap across European
countries: men are more likely than women to feel that they can take an active role in
political groups and/or to trust on their ability to get involved in politics. These differences
are statistically significant across all countries.
5
The size of the gender gap in internal
political efficacy is relevant and varies across countries from 0,4 to 0,8. Since the index used
to measure efficacy ranges from 0 to 8 (see Appendix: descriptive statistics in Table 1) these
gender differences imply between 5 and 10 percentage points of total variation in internal
political efficacy.
4
We do not expect that the effect of gender completely disappears, because that would mean that self-
perceptions of risk explains all the remaining association between gender and internal political efficacy.
Theoretically speaking other psychological factors might be at play, so we expect a partial (instead of complete)
mediation.
5
For those countries included in the two waves we provide mean values of the two waves.
14
14
Figure 2: Internal political efficacy. Differences between mean values for men and mean
values for women across countries. Source: ESS (8th and 9th wave).
Equation 1 in Table 1 further indicates that the gender gap in internal political
efficacy is sizeable and statistically significant even after controlling for typical socio-
demographics associated with internal political efficacy, including personal confidence.
These findings confirm the patterns in previous studies done in Canada and the United States
where the gender gap in internal political efficacy persists despite women’s resource gains
(Gidengil et al. 2008; Thomas 2012, p. 339). It also persists, we may add, in different
European countries and regardless of country-level diversity.
We now turn to testing whether self-perceptions of willingness to take risks contribute
explaining part of the gender gap in efficacy that we find in Equation 1. Equation 2 in Table 1
estimates respondents’ self-perceptions of their proclivity to take risks in life, to examine the
first condition of the mediation hypothesis. The results support this condition and confirm
what prior scholarship has concluded, that men feel significantly more likely to take risks
than women (Byrnes et al. 1999; Kam 2012; Weber et al. 2002). More specifically, men feel
more ready than women to take risks by 0.34 on average, which implies around a 6.8% of
total variation in risks (ranging from 1 to 6). Equation 2 also corroborates prior findings
0,00
0,10
0,20
0,30
0,40
0,50
0,60
0,70
0,80
0,90
BG
DK
LT
NO
PL
IS
ES
FI
HU
SE
PT
LV
SK
EE
avergae
IE
IT
HR
BE
NL
FR
DE
SI
CZ
GB
AT
CY
CH
15
15
showing that the inclination to take risks decreases with age whereas it increases with
education. Finally, self-confidence is positively associated with readiness to take risks in life
so that the more respondents feel confident of their own abilities, the greater their inclination
to take risks.
Eq 1 (IPE)
Eq2 (Risk)
Eq3 (IPE)
Men
0.51***
0.34***
0.46***
0.01
0.01
0.01
Age(in years)
-0.00***
-0.02***
-0.00***
0.00
0.00
0.00
Education (in years)
0.12***
0.01***
0.12***
0.00
0.00
0.00
Working
0.05**
-0.03**
0.06***
0.02
0.01
0.02
Income
0.07***
0.01***
0.07***
0.00
0.00
0.00
Confidence
0.15***
0.25***
0.11***
0.01
0.00
0.01
Risk
0.15***
0.00
Wave 9
0.06***
-0.03**
0.07***
0.01
0.01
0.01
Constant
0.58***
3.02***
0.13*
0.05
0.04
0.06
country effects
yes
yes
yes
N
66724
66724
66724
Adj R2
0.24
0.17
0.25
Table 1. Estimation of Internal political efficacy and self-perceptions of risk (first step).
Unstandardized OLS coefficients. Source: ESS (8th and 9th pooled waves).
Equation 3 in Table 1 tests the second mediation condition. The results indicate that
people’s self-perceptions of their proclivity to take risks in life are related to their feelings of
internal political efficacy so that the higher their inclination to take risks the greater is their
internal efficacy. More precisely, a one unit increase in tendency to take risks (a variable
16
16
ranging from 1 to 6) is associated with an average increase in internal political efficacy
(ranging from 2 to 10) of 0.15. This implies that a one unit increase in proclivity to take risks
contributes to explain two percentage points of the total variation in internal political
efficacy, and a maximum of nine percentage points if we compare someone with the lowest
inclination to take risks with someone with the highest proclivity to take risks in life.
Equation 3 also shows that the size of the association between inclination to take risks and
internal political efficacy is as relevant as that of typical sociodemographic variables (such as
education, having children, or age). Finally, Equation 3 shows that the size of the gender gap
in efficacy decreases when we include willingness to take risks in the estimation: from
0.51*** to 0.46*** (compare the coefficient corresponding to men in Equation 1 and 3 in
Table 1).
Taken together Table 1 confirms that the conditions of the mediation hypothesis are
met, and that part of the association between gender and internal efficacy is explained by self-
perceptions of proclivity to take risks in life. However, in order to properly test the mediation,
we use Imai et al (2011)’s approach to partition the share of the association between gender
and political efficacy that is conveyed through self-perceptions of inclination towards risk. In
particular, we decompose the total effect of gender on internal political efficacy into direct
and indirect effects––the average direct effect (ADE) and the average causal mediation effect
(ACME), respectively. This provides a substantive measure of the magnitude of this
mediation, and shows that the mediation is statistically significant, something that the OLS
estimations summarised in Table 1 cannot offer.
Table 2 summarizes the findings of the mediation estimation. The average direct
effect-ADE (0.46) denotes the effects of gender on internal political efficacy after controlling
17
17
for the contribution of proclivity to take risks. The average causal mediation effect-ACME
(0.051) represents the change in internal political efficacy resulting from the differences in
proclivity to take risks between men and women. And it is statistically distinguishable from
zero. Finally, and what is more relevant here, the percentage of total effect mediated is 10%,
and indicates how much the total effect of gender on internal political efficacy is mediated by
self-perceptions of inclination to take risks.
Estimate
Lower 95% CI
Upper 95% CI
ACME
0.051
0.046
0.055***
ADE
0.463
0.435
0.49***
Total Effect
0.513
0.486
0.54***
Proportion mediated
0.010
0.093
0.104***
Table 2. Mediation analysis of the effect of gender on internal political efficacy via self-perceptions of
proclivity to take risks (second step). We used the R Package mediation (Tingley et al. 2014). Source:
ESS (8th and 9th pooled waves).
Gender differences in respondents’ inclination to take risks in life are useful to
understand a 10 percentage of the total gender gap in citizens' beliefs about their
competence/abilities to play an active role in politics. This might appear as a modest figure.
Still it is relevant. For one, the significant mediation effect is robust across different
specifications (see the next section below). More importantly, explanations of the persistent
gender gap in internal political efficacy have remained largely incomplete in past studies. Our
findings point to an alternative account suggesting the need to look at the ways in which
women and men perceive their own roles and qualities. These perceptions might be formed
early in life and influenced by the social norms and stereotypes attached to gender, and thus
have important implications on how we think about reducing the gender gap, which we
further discuss in the concluding section.
18
18
Discussion and robustness checks
The analyses presented so far support our argument that politics is perceived by
citizens as more welcoming to the attributes that men appraise themselves to have to a greater
extent than women. We have also found a statistically significant partial mediation of
proclivity to take risks in life. However, some of the assumptions and mechanisms that we
posit in our argument are hard to fully test with the data at hand, and thus justify a more
detailed discussion here.
First, our argument implies that people’s self-perceptions of their inclination to take
risks in life precede feelings of internal political efficacy. We recognize that our research
design (based on observational evidence from a cross-national comparative survey) has
limitations in terms of fully demonstrating this causal link. Moreover, while we are using the
causality language to describe the mediation test summarized in Table 2, we acknowledge
that the evidence provided is correlational (or associational) rather than causal. However, we
have confidence in the causal direction of our argument for two reasons. First, the items in
the survey that measure self- perceptions of proclivity to take risks in life were asked at the
very end of the ESS08/9 questionnaires, immediately after a group of questions related to the
personal/family socioeconomic situation of the respondent. The location of these questions in
the questionnaire thus minimizes the possibility that political issue questions contaminated
the responses to perceptions of risk. Second, at a theoretical level it seems reasonable to
assume that self-perceptions of personal attributes (and of gendered perceptions more
generally) are prior to political attitudes. We have previously mentioned experimental
evidence showing that gender stereotypes associating higher levels of cognitive ability with
boys are acquired quite early in life (Lian and Cimpian 2017).
19
19
We also conducted additional robustness tests to address (i) alternative mechanisms at
work to account for the gender gap in efficacy; and (ii) survey items used to measure internal
political efficacy. These robustness tests provide further empirical support for our argument.
A potential alternative mechanism accounting for the gender gap in political efficacy is that
self-perceptions of proclivity to take risks have different effects for women than for men.
Prior scholarship on political ambition at the elite level (Lawless and Fox 2010) has
documented that attitudes towards one’s own personal qualifications are (together with
family obligations and career pathways) especially relevant in explaining women’s decisions
to run for elections, but not men’s decisions. At the level of the mass public, the factors
associated with political ambition are also different for men and for women. According to
Crowder-Meyer (2018), ordinary women’s ambition appears especially shaped by the support
of personal and political sources that can facilitate them to oversee the requests of candidacy.
Similarly, one could argue that the alignment between proclivity to take risks and
insights on the political sphere is more relevant to women than to men in forming their
judgements about their effectiveness in the realm of politics. If this is the case, then the
association between self- perceptions of disposition to take risks and internal political
efficacy should be larger for women than for men, which would contribute to the gender gap
in internal political efficacy. We test this possibility by replicating the estimation of equation
3 in Table 1 and adding an interaction term between gender and self-perceptions of proclivity
to take risks. Figure 3 summarizes the findings of this estimation and shows that the size of
the association between readiness to take risks and internal political efficacy is not
statistically distinguishable between men and women.
6
For example, the predicted value of
6
The coefficient corresponding to the interaction term between gender and self-perceptions of proclivity to take
risks is = -0.017 (0.009) with corresponding p = 0.06.
20
20
internal political efficacy of someone who presents the lowest level of proclivity to take risks
in life is 1.7 for women and 2.2 for men (implying a gender difference of 0.5 that is
statistically different from zero). This gender difference is similar for respondents presenting
the highest level of proclivity to take risks in life (2.5 for women and 2.93 for men, involving
a gender gap of 0.43 that is statistically significant).
7
Figure 3. Predicted Internal Efficacy by self-perceptions of readiness to take risks in life for
women and men (moderation alternative mechanism). Source: ESS (8th and 9th pooled
waves).
7
Put it another way: the predicted value of efficacy is on average 0.70 greater for someone with the highest
proclivity to take risks (and in comparison to someone with the lowest) for women and 0.71 for men.
21
21
Finally, we have replicated the estimations presented in this article with data from the
prior wave (ESS07) containing slightly different items to measure internal political efficacy.
8
Findings for the replication of Table 1 and 2 in the main text are summarized in the
Appendix, Tables 2 and 3. Confirming our results, we find that 9% of the total association
between gender and internal political efficacy is mediated by self-perceptions of willingness
to take risks. A mediation that is statistically significant in both cases (see Table 4 in the
Appendix).
Yet, we recognize that our findings are limited to the explicit component of internal
political efficacy that is linked to political participation but cannot be extrapolated to its
second dimension: the one tapping citizens’ impressions about their abilities to understand
politics in the abstract. Future research involving the collection of surveys including enough
items to measure all dimensions of internal political efficacy would contribute advancing the
study of the sources of inequalities of this attitude.
Conclusions
Women are less likely to feel politically efficacious than men. We expand existing
work on this crucial aspect of political engagement by showing that this gap is a general
pattern across countries in Europe, corroborating what has been found previously in the US
and in Canada. Our other contribution is to incorporate social role theory to the explanation
of the existence of the gender gap in internal political efficacy. While previous studies have
emphasized socioeconomic or contextual explanations, we show in this article that engaging
in politics is more in line with gender stereotypes about men. What is more, political
8
The wording of the three items is: (i) how confident are you in your own ability to participate in politics?”; (ii)
“How easy do you personally find it to take part in politics?” and (iii) how able do you think you are to take an
active role in a group involved with political issues?
22
22
participation is perceived as more welcoming to the attributes that men appraise themselves
to have to a greater extent than women.
Our use of the role congruity model to complement existing explanations of the
pervasive gender gap in internal political efficacy adds to an incipient literature suggesting
the utility of applying social role theory to the study of gender and politics (Schneider and
Bos 2019). Our content is that low levels of internal efficacy are the product of a miss-
alignment between how individuals perceive themselves and the common perceptions about
political participation. This miss-alignment is greater for women, we argue, which is our
explanation of the gender gap in citizens’ feelings about their capacity to play an active role
in politics. This argument opens up a complementary way to think about the diverse roads
through which gender shapes politics around the world.
Using observational survey evidence from the two last waves of the ESS (8th and 9th),
we show that those who perceive themselves as ready to take risks in their life are also more
likely to feel able to play an active role in politics. We also find evidence of a partial
mediation of the association of gender and internal political efficacy through people’s
perceptions of their inclination to take risks, discarding the alternative explanation that the
association of perceptions of risk and efficacy is of a different magnitude for men and
women. Taken together these results suggest the need to integrate other perspectives and
explanations (such as the role congruity model) when assessing the gender gap in public
opinion and political engagement.
Our findings seek to respond to the puzzle of the persistent gender gap in internal
political efficacy, despite significant gains of women in employment, education, and
23
23
economic status in advanced democracies. The implications of these findings are important
for how we think about closing this gender gap. While material and educational advancement
has made a difference, it appears insufficient to close the gender gap. We suggest to pay more
attention to early childhood socialization and to the way in which social norms and
stereotypes attached to gender are learned through socialization agents: family, school,
friends, and specially the traditional and social media. Overcoming these traditional gender
stereotypes, would go a long way in closing the gender gap in people’s feelings about their
capacities to play an active role in politics.
But perhaps more importantly, we think that institutions, politicians, parties and the
media should contribute to change the idea that politics is a fierce world of self-promotion
and competition. Traditional political formats (such as televised debates) and descriptions of
the electoral world using references to stereotypical masculine attributes such as
competitiveness, belligerence, and/or riskiness only contributes to preserve a political culture
dominated by men. And this masculinized ethos works with traditional gendered socialization
to reinforce women’s feelings that they do not belong in the realm of politics. A feeling that
is truly rational, given the masculine stereotypes that surround politics. Making politics
equally inclusive to men and women requires the collective work of all relevant political
actors.
24
24
References
Balch, G. I. 1974. Multiple indicators in survey research: The concept sense of political efficacy.
Political Methodology, 1(2): 1–43.
Bandura, A. 1997. Self-efficacy: The exercise of control. New York: Freeman.
Byrnes, J. P., D. C Miller and W. D. Schafer. 1999. Gender differences in risk-taking: A meta-
analysis. Psychological Bulletin 125(3): 367–383.
Condon, M., and M. Holleque. 2013. Entering politics: General self-efficacy and voting behavior
among young people. Political Psychology 34(2): 167–181.
Crowder-Mayer, M. 2020. Baker, bus driver, babysitter, candidate? Revealing the gendered
development of political ambition among ordinary Americans. Political Behaviour 42: 359–
384.
Diekman, A. B., Steinberg, M., Brown, E. R., Belanger, A. L., and Clark, E. K. 2017. A goal
congruity model of role entry, engagement, and exit: Understanding communal goal
processes in STEM gender gaps. Personality and Social Psychology Review 21(2): 142–175
Eagly, A. H., and Wood, W. 2012. Social role theory. In P. A. M. van Lange, A. W. Kruglanski, and
E. T. Higgens (Eds.) Handbook of theories in social psychology (Chapter 49: 458–476).
Thousand Oaks, CA: Sage.
Eagly A. H. and Karau S. J. 2002. Role congruity theory of prejudice toward female leaders.
Psychological Review 109(3): 573-98.
Fortin-Rittberger, J. 2016. Cross-national gender-gaps in political knowledge: How much is due to
context? Political Research Quarterly 69(3): 391–402.
Gidengil, E., J. Giles, and M. Thomas. 2008. The gender gap in self-perceived understanding of
politics in Canada and the United States. Politics & Gender 4: 535–561.
Imai, K., L. Keele, D. Tingle and T. Yamamoto. 2011. Unpacking the black box of causality:
Learning about causal mechanisms from experimental and observational studies. American
Political Science Review 105: 765–789.
Kam, C. D. 2012. Risk attitudes and political participation. American Journal of Political Science
56(4): 817–836.
Karp, J.A. and Banducci, S.A. (2008). When politics is not just a man’s game: Women’s
representation and political engagement. Electoral Studies 27(1): 105–114.
Karpowitz, C F., and T. Mendelberg. 2014. The Silent Sex: Gender, Deliberation, and Institutions.
Princeton University Press.
25
25
Kittilson, M. C., and L. A. Schwindt-Bayer. 2012. The gendered effects of electoral institutions. New
York: Oxford University Press.
Lawless, J. L., and R. L. Fox. 2010. It still takes a candidate: Why women don’t run for office. New
York: Cambridge University Press.
Lian, B., S-J. Leslie and A. Cimpian. 2017. Gender stereotypes about intellectual ability emerge
early and influence children’s interests. Science 355(1): 389–391.
Marinova, D. and E. Anduiza. 2020. When bad news is good news: Information acquisition in times
of economic crisis. Political Behavior 42(2): 465–486.
Niemi, R. G., SC Craig and F. Mattei 1991. Measuring internal political efficacy in the 1988
National Election Study. American Political Science Review, 85(4), 1407-1413
Quaranta, M., and G. Dotti Sani. 2018. Left behind? Gender gaps in political engagement over the
life course in 27 European countries. Social Politics 25(3): 254–286.
Saris, W. E., and M. Torcal. 2009. Alternative measurement procedures and models for political
efficacy. Retrieved from: http://hdl.handle.net/10230/28300
Schneider, M. C., M. R. Holman, A. B. Diekman and T. McAndrew. 2016. Power, conflict, and
community: How gendered views of political power influence women’s political ambition.
Political Psychology 37(4): 515–531.
Schneider, M.C. and Bos, A.L. 2019. The application of social role theory to the study of gender in
politics. Political Psychology 40(1): 173-213
Schrock, D. and Schwalbe, M. 2009. Men, masculinity, and manhood acts. Annual Review of
Sociology 35(1): 277-95
Thomas, M. 2012. The complexity conundrum: Why hasn’t the gender gap in subjective political
competence closed? Canadian Journal of Political Science 45(2): 337–358.
Tingley, D.,Yamamoto,T.,Hirose, K.,Keele, L. and Imai, K. 2014. Mediation: R package for causal
mediation analysis. Journal of Statistical Software 59(5): 1–38.
Weber, E.U., A-R. Blais and N. E. Betz. 2002. A domain-specific risk-attitude scale: Measuring risk
perceptions and risk behaviors. Journal of Behavioral Decision Making 4(15): 263–290
West, C., and D. Zimmerman. 1987. Doing gender. Gender & Society 1(2): 125–51.
Wolak, J. 2018. Feelings of political efficacy in the fifty states. Political Behaviour 40: 763–784.
Wolak, J. 2020. Self-confidence and gender gaps in political interest, attention, and efficacy. Journal
of Politics 82(4): 1490-1501. !
26
26
APPENDIX
Variable
Observations
Mean
Sd
Min
Max
Internal political efficacy
82613
2,24
1,93
0
8
Men
85667
0,47
0,50
0
1
Age
85330
50,29
18,65
15
96
Education
84624
13,02
3,98
0
26
Working
85676
0,54
0,50
0
1
Income
70017
5,27
2,76
1
10
Personal confidence
84213
4,80
1,10
1
6
Risk
84130
3,07
1,43
1
6
Table 1. Descriptive statistics. Source: ESS (8th and 9th wave). Education is a measure of years of
full-time education completed. Income is the self-reported household income, after tax and
compulsory deductions, from all sources. Working is a dichotomous variable coded by the interviewer
that indicates whether the respondent has paid work (1) or does not have paid work (0). Personal
confidence and Risk tolerance range from 1 = not like me at all to 6 =very much like me
List of countries covered in this study: Austria, Belgium, Bulgaria, Croatia, Cyprus, Czechia,
Denmark, Estonia, Finland, France, Germany, Hungary, Iceland, Ireland, Italy, Latvia,
Lithuania, Netherlands, Norway, Poland, Portugal, Slovakia, Slovenia, Spain, Sweden,
Switzerland and United Kingdom
27
27
Table 2. Replication of Table 1 in the main text with data from the ESS07. Unstandardized
OLS coefficients. Source: ESS (7th wave)
Estimate
Lower 95% CI
Upper 95% CI
ACME
0.056
0.048
0.07***
ADE
0.632
0.579
0.68***
Total Effect
0.688
0.637
0.74***
Proportion mediated
0.089
0.079
0.10***
Table 3. Mediation analysis of the effect of gender on internal political efficacy via self-
perceptions of proclivity to take risks (second step). We used the R Package mediation
(Tingley et al. 2014). Replication of Table 2 in the main text with data from the ESS07.
Source: ESS (7th wave).
