Citizen support for the pursuit of gender equality in US foreign policy

Abstract

We ask two questions in this article: First, what is the level of public support for the pursuit of gender equality in foreign policy? Second, what are the most significant correlates of that support? We report the results of the first national opinion survey that queried citizens about their support for policies to increase global gender equality. We find that an average of 60 to 90 percent approve of pursuing gender equality and specific programmatic initiatives. Americans also strongly agree with one rationale for pursuing global gender equality: 65 percent agree that “The world would be a more peaceful place if more women were involved in making decisions.” Second, we find that a number of individual characteristics and personal values are strongly correlated with support for gender initiatives. Our results also suggest that women's greater endorsement of universalism values explains their higher levels of support for global gender equality.

Notes

Acknowledgments

We are grateful to Peter Levine and Kei Kawashima-Ginsberg, who participated in the design of the opinion survey reported here, and to the Office of the Provost at Tufts University for funding. Thanks also to Jason Reifler and Thomas Scotto for advice on survey design, and to Anna Weissman for research assistance.

Notes

1 Press release, Foreign Affairs Committee, US House of Representatives, September 28, 2017. The bill was signed into law by President Trump on October 6, 2017. The press release is available on the website of the Women’s International League of Peace and Freedom (2017).

2 For additional reporting on Trump administration policies on gender and the administration’s lesser commitment to gender equality, see Toosi (2018) and Matfess (2018).

3 Surveys conducted for the Kaiser Family Foundation (February 2–12, 2012, August 6–20, 2013, and March 1–26, 2016) and Chicago Council on Global Affairs (June 10–June 27, 2016).

4 The Online Supplemental Appendix contains complete details on YouGov’s sample matching and weighting procedures as well as complete question wording and response codes for all questions we employ in our analysis.

5 For detailed reviews of this literature, see Hudson and Leidl (2015: 68–73) and Barnhart et al. 2019).

6 The Schwartz value types measure ten cross-culturally consistent value types—achievement, power, stimulation, hedonism, self-direction, benevolence, universalism, conformity, tradition, and security—that are composed of sets of conceptually similar personal values.

7 The following paragraphs draw on our earlier research on gender difference in attitudes toward war (Eichenberg 2019).

8 We draw this and other quotations in this paragraph from Goldstein (2001: 251–321, 410–411).

9 In the case of commitment, the dummy variable for the ranking of the rights of women and girls takes the value of 1 if respondents ranked this item first from among a list of five human rights. In addition, the dummy variable for the second spending item in Table 1 takes the value of 1 if the respondent chooses “increase” from three alternatives (increase, keep spending same, decrease). In the case of the voting item (the last variable listed under commitment), the dummy variable takes the value of 1 if the respondents chooses “more likely to vote for the candidate” in the 2018 midterm election. Full wording for these and all questions in Table 1 appears in the Supplemental Online Appendix (provided for review purposes).

10 In response to a reader’s suggestion, we also analyzed versions of the endorsement and commitment dependent variables that include the sum of the full range of response codes from the component variables. The results, displayed in Online Supplementary Appendix Table A1, are generally the same as those reported here. There is one important exception: The influence on parents of having daughters is highly significant in the analysis of commitment. We pursue the analysis of parents and children in a separate body of research.

11 We employ OLS for endorsement and commitment, which are additive indices. We employ logistic regression for women and peace, which is a binary-dependent variable. The regressions included a number of control variables that proved insignificant, so they are excluded from the following discussion. The full results are displayed in Online Supplemental Appendix Table C1. Data are weighted using the survey weights provided by YouGov.

12 These preliminary analyses are presented in Online Supplementary Appendix Tables B1–B3.

13 We also found that several additional variables unrelated to our theory were largely insignificant and are thus omitted from this discussion, although they appear in Online Supplementary Appendix Tables B1–B3. Specifically, ethnocentrism shows no relationship to any dependent variable, and perhaps most surprising, a measure of traditionalism (respect for authority) shows no relationship to any of our dependent variables. Despite the fact that realization of gender equality represents a direct challenge to traditional authority, our measure of respect for traditional authority is not significant in any model.

14 We are unable to report the percentage mediation according to the Hicks and Tingley (2011) mediation package (based on the new mediation procedure from Imai, Keele, Tingley, and Yamamoto, 2011), because it is not compatible with the weighted survey commands in Stata.

15 On “feminist hawks,” see Hudson and Leidl (2015: 34–60).

Additional information

Notes on contributors

Mary-Kate Lizotte

Mary-Kate Lizotte is an associate professor of political science at Augusta University. Her research interests include political behavior, public opinion, and gender. Much of her work is concerned with the origins and implications of gender difference in public opinion.

Richard C. Eichenberg

Richard C. Eichenberg is an associate professor of political science at Tufts University. He has published on public opinion concerning foreign policy, national security, and European integration. His most recent work focuses on gender difference in national security attitudes.

Richard J. Stoll

Richard J. Stoll is the Albert Thomas Professor of Political Science at Rice University. His research interests are in the study of international conflict, public opinion on foreign and defense policy and American national security policy.