Revisiting the Gender Gap in 2020: The Gender Gap in Vote Choice

This blog post is part of our “Revisiting the Gender Gap in 2020” symposium. In this blog series we are interviewing experts about the gender gap in American politics in an effort to shed light on how these gaps may impact the 2020 presidential election. These posts are curated conversations, moderated by email by CAWP faculty and staff, with gender and politics experts intended to illuminate the meaning of the gender gap, to explore what we understand and do not understand about women’s political behavior, and most importantly to address questions posed in the media about gender and the presidential election. The interviews have been lightly edited for length and clarity.

We are likely to see a sizable gender gap—and perhaps one of unprecedented magnitude—in voting choices in the upcoming 2020 presidential election.  In an attempt to better understand the dynamics underlying the gender gap historically and in 2020, I asked three experts – Professor Christina Bejarano (Texas Woman’s University), Professor Mary-Kate Lizotte (Augusta University), and Pollster Christine Matthews (Bellwether Research & Consulting) -- to reflect on factors shaping the gender gap, the subgroups of women who may be most critical to 2020’s presidential election outcome, and the strategic advice they would offer to the presidential campaigns about how to maximize their support among women voters. Their responses (and a couple of my own) provide a guide to watching and analyzing the role and influence of women voters in this year’s election.  

Carroll: We have seen a gender gap every presidential election since 1980 with women more likely than men to back Democratic candidates. Recent polls indicate that a gender gap is also likely to be evident in voting in the upcoming presidential election. What do you see as the most important factor(s) shaping the gender gap in presidential vote choice both historically and in 2020?

Matthews: One of the reasons we saw the emergence of the gender gap in 1980 is that the Republican platform changed in significant ways to reflect the views of Ronald Reagan: the party called for a dramatic increase in defense spending and the military budget, backed away from support for the Equal Rights Amendment, and called for a constitutional ban on all abortions. Prior to that, Republicans and Democrats generally had similar attitudes on abortion – something that is hard to believe now.

In the 2016 election between Donald Trump and Hillary Clinton, the gender gap, defined as the difference in the proportion of women and men supporting the winning candidate, was 11 points. One would think this contest would have blown the lid off the gender gap, but it only tied the 11-point gender gap from the 1996 election between Bill Clinton and Bob Dole. Polling pointed to a more significant gender gap in 2016 until the late-October release of the Comey letter saying the FBI was re-opening its investigation into Hillary Clinton’s emails. Late undecideds – men and women – broke toward Trump.

In 2020, national polling indicates a higher-end gender gap – somewhere between 10 and 12 points, so it is possible it could be record-setting. But it is also possible it could be at a level we have seen before. College-educated white women have moved dramatically toward Democrats in the past few cycles, but this year, white college-educated men are joining them in moving to Biden, just in less dramatic fashion. However, these men supported Trump by a 14-point margin in 2016, so their blue move in 2020 may lessen what could have been a more significant gender gap. Non-college educated white women are less enthusiastic about Donald Trump than they were in 2016, while non-college educated white men have remained steady. How much of a gap there will be between them really depends on how far non-college educated white women move in 2020.

Lizotte: Attitudes on social welfare issues are likely a top reason why women are more likely to identify as Democrats and more likely to vote for Democratic presidential candidates. Women have been consistently more likely than men to support greater government spending and the greater provision of government services to citizens including spending on aid to the poor, public schools, social security, and childcare as well as a greater role for government in providing healthcare and general services. My analysis of the American National Election Study from 1980 to 2016 shows attitudes on social welfare issues to be a significant explanation of gender gaps in both party identification and presidential vote choice and to account for more of these two gender gaps than income, motherhood, and attitudinal differences on other issue areas. My analysis also shows that women across various subgroups are more likely than men of the same subgroup to have liberal views on social welfare issues. White women are more liberal than white men, and Black women are more liberal than Black men. This is also true among different age cohorts with Boomer women and post-Boomer women more supportive of social welfare spending than Boomer men and post-Boomer men, respectively. This is also true among those with and without a college degree and among lower and higher income individuals. A separate analysis also finds Republican women and Independent women to be more supportive of government social welfare spending and higher levels of government services than Republican men and Independent men, respectively.

Because of the covid-19 pandemic, many social welfare issues have garnered attention in the media and have likely been salient among the public. Healthcare is an obviously prominent issue during the pandemic because of the burdens on healthcare workers, the cost of healthcare services, and the inequalities in healthcare access. Increased levels of unemployment have also made policies such as the supplemental nutrition assistance program and unemployment benefits salient to the public. Many Americans are financially struggling right now. In addition, the impact of the pandemic on public schools has likely made the issue of government spending on public education more prominent in the minds of voters. Thus, these issues of government spending to provide services to the public and to provide for the well-being of Americans will likely influence voters this election.

Bejarano: Various explanations have been offered for the development and persistence of the gender gap in U.S. politics. However, my analysis reminds us that it is important to expand on these explanations to incorporate the intersecting experiences of racial/ethnic and immigrant groups in this country since they can have gendered experiences of political incorporation and political socialization that can influence their political behavior and attitudes. A close gender analysis of racial/ethnic minority subgroups shows that they are more likely than white voters to support Democratic candidates and issues and less likely to show a wide partisan divide along gender lines. In this politically charged environment leading to the 2020 presidential election, we are likely to again see strong support by racial/ethnic minorities for Democratic candidates since both racial/ethnic minority women and men are likely to be mobilized around issues of racial equality. In addition, there will be signs of a greater depth of support for Democratic candidates by women of color who are also mobilized around issues of gender equality. Social welfare issues will also prove important to racial/ethnic minorities this election especially since they are disproportionately impacted by the current pandemic.

Carroll: As everyone has indicated, evidence suggests that the gender gap is primarily issue-based. Women and men differ in the aggregate in their views on several key policies and issues that align with differences between the Democratic and Republican parties. Most fundamental based on my review of polling data over time is that women and men differ in their attitudes toward the role of government. Men are more likely than women to see government as having too big a role in our lives and to want to cut back on government programs and regulations; women are more concerned with preserving a social safety net and more likely to see a positive role for government in providing for those who are economically or socially disadvantaged.

The gender gap seems less a product of personality or character differences among candidates than issues and values in part because most voters have very little knowledge about candidates below the presidential level. Most voters don’t even know the names of the officials who represent them. But it does seem likely that the larger-than-usual gender gaps we are seeing in the presidential contest in 2020 are in part a product of voters’ evaluations of the character of the contenders, especially Trump’s character. His oversized personality is on display every day although the interpretations of his tweets, statements, and actions vary widely. His supporters, disproportionately men, are wildly enthusiastic about him and embrace or at least tolerate his personal style while his critics, disproportionately women, are equally passionate and view his behavior as unpresidential and often repugnant. If the gender gap in voting in the 2020 presidential election proves to be larger than the gap in previous elections, as recent polls suggest it may, the explanation may lie at least in part with gender differences in evaluations of Trump’s character.

Carroll: Various subgroups of women voters (e.g., Black, young, white college-educated, suburban) have been identified in media coverage as potentially important in influencing the outcome of the 2020 presidential election. Overall, which subgroup(s) of women voters do you think will prove to be the most critical and/or are you most interested in watching?  Why?

Bejarano: The 2016 presidential election included the most diverse electorate in U.S. history.  The majority of this rising American electorate is made up of predominantly young voters, unmarried women, and racial/ethnic minority populations. The most critical subgroup to watch is women of color, especially those who are young and unmarried. Women of color account for the largest increase, at 74%, in eligible women voters since 2000. Not only are women of color a growing segment of eligible voters in the country, they are also highlighting their distinctive political attitudes that help them lead the way in increasing minority political power. In fact, voting differences between men and women in U.S. politics are now reinforced among minority women.

We need to pay close attention to the impact of women of color’s electoral support in the 2020 election. In the 2012 and 2016 presidential elections, women of color overwhelmingly supported the Democratic candidates, President Barack Obama and Hillary Clinton, by a wider margin than their male counterparts. With my research, I highlight this growing importance of women of color in electoral politics, with a particular focus on Latina voters. In examining the political gender gap among minority populations like Latinos, it is also important to include additional breakdowns by subgroups such as national origin, nativity, age, and generational status. When these additional breakdowns are analyzed, we can see that there are significant gender differences in racial/ethnic minority partisan identification and vote choice. For example, my analysis shows that compared to men in their community, Latinas tend to show higher levels of voter registration and turnout, and have a greater depth of support for liberal public policies and Democratic candidates. Latinas and other women of color are likely to continue to vote at higher rates and have the potential to give more electoral support to Democratic candidates, which can have a dramatic impact on the 2020 election. As a result, it is important to expand our analysis and political attention to women of color in U.S. politics. 

Lizotte: According to my research, college-educated women are consistently liberal across several policy areas including gun control, environmental regulations, LGBTQ rights, and social welfare policy. And 57% of college-educated women have voted Democratic in presidential elections during the period of 1980 to 2016. In the 2018 midterm elections, according to my analysis of the American National Election Study (ANES) data, over 70% of college- educated women voted for Democratic candidates in House races, and college education was also a significant predictor of women voting for Democratic Senate and gubernatorial candidates. With mass shootings in public schools, forest fires and hurricane season worsening because of climate change, the threat to LGBTQ rights because of the passing of Justice Ginsberg, and the demand for government aid during the pandemic, all of these issues are likely to compel college-educated women to continue their historical support of the Democratic Party.

In 2020, the media and political pundits have noted the importance of white college-educated women to the presidential election. Exit poll data from 2016 and 2018 indicated white college-educated women were also more likely to vote for Democrats: 51% for Clinton in 2016 and 59% for House Democratic candidates in 2018. This increase in support for Democratic candidates from 2016 to 2018 could indicate a decrease in favorability toward President Trump, Republicans in Congress, and their policies among white college-educated women, which could prove beneficial for the Biden-Harris ticket as well as down-ticket Democrats in November. Moreover, the college-educated (women and men) turnout at higher rates than those with less than a Bachelor’s degree. Thus, the college-educated in general, college-educated women and white college-educated women in particular, are a likely politically consequential segment of the population for the 2020 election.

Matthews: Non-college-educated white women will be an extremely important cohort to watch; Trump cannot win re-election based on support from non-college-educated white men alone. I will also be watching turnout among Gen Z women, some of whom will be voting for the first time in a presidential race. Overwhelmingly Democratic by partisanship, this group is also liberal, and Joe Biden was not the first choice for many.  While many had committed to casting a vote against Donald Trump, with the passing of their progressive icon, Ruth Bader Ginsburg, many may now be organizing “for Ruth.” 

Carroll: I’m also interested in watching what happens with senior women in some of the critical battleground states. Three of the critical battleground states—Florida, Pennsylvania, and Arizona—are among the top twelve in the country in the proportion of residents who are age 65 or older. Along with white non-college-educated voters, seniors have been among Trump’s strongest supporters both in the 2016 election and throughout much of his term in office. However, recent polls have shown signs of declining support among seniors with women less likely than men to support Trump. The lack of a stronger national response to Covid-19 and fears that he may cut Medicare and/or Social Security may have hurt Trump with this vulnerable age group. Also, this age group, who has also lived long enough to experience several chief executives of both parties, may be more acutely aware of how aberrant Trump’s demeanor is for a president and thus more troubled by it. The level of support the candidates receive from seniors, who turn out to vote at higher rates than other age cohorts, and senior women in particular, may well be a key to understanding election outcomes, especially in some of the critical battleground states.

Carroll: What strategic advice would you give to either or both presidential campaigns about steps they should be taking between now and Election Day to maximize the number of votes they receive from women voters?  What should they be doing to increase turnout among key groups of women voters and/or to persuade women who may still be undecided? 

Lizotte:  To attract the majority of women voters, the Biden-Harris campaign should emphasize a policy agenda that addresses the problems and concerns currently facing the American public during the pandemic. First, the Biden-Harris campaign should state that they will fight for legislation to provide support for those who are struggling financially including, but not limited to, continuing the increases in unemployment benefits which began during the pandemic. Second, the Biden-Harris campaign should provide a plan to address the burden on hospitals and healthcare workers. Third, it would likely be appealing to many women for the campaign to provide assurances about access to healthcare, affordability of care, and reducing the current inequalities in access, treatment, and survival rates. The current administration has at times seemed unconcerned with the racial differences in COVID-19 related deaths, and at other times has blamed higher death rates on pre-existing conditions, rather than acknowledging that racial stereotyping and discrimination is likely a contributing factor.

Prior research indicates that men and women differ in their views of the proper role of government; men are more likely to want a small government while women are more likely to support a government that provides a greater level of programs and services. Pointing out failures of the current administration is likely not enough to motivate turnout among women, who believe in a greater role for government. Thus, the Biden-Harris campaign needs to target and appeal to women with the underlying belief that government can solve public problems and convince them that a Biden administration would work tirelessly to resolve the current economic and health crises.

Bejarano: My advice to the candidates is to not take women voters for granted, especially those of women of color. Both parties likely assume they know how women are going to vote, based on preconceived notions or because of their previous voting trends. However, this pandemic election season is calling attention to various driving political issues that can motivate more women to vote. So, it is in candidates’ best interests to mobilize as many potential voters as possible, rather than only focusing on those ‘likely’ or ‘for sure’ supporters. In addition, the Democratic Party needs to strategically mobilize women of color voters and highlight all of the issues that can appeal to them, rather than assume they only care about a couple of ‘hot button’ issues. The party needs to acknowledge that the pandemic and our social unrest has been especially challenging for women of color. The modern gender gap is likely to be reinforced by women of color voters again this year and they can provide the pivotal votes if the party does not take them for granted.    

Matthews: With President Trump’s positive diagnosis of COVID-19, everything feels – again – disrupted. It is difficult to imagine what the next 30 days will bring, but one hallmark of this contest so far has been its stability.

There is now no question that COVID-19 will dominate the remaining days of the campaign in a way that Trump tried hard to prevent. There are so many ways this could play out depending on the president’s health and the extent of the spread within the administration and among elected officials, but taking it from where we are now, here are some data points among women on this second day of October:

  • Joe Biden has been on track to win the votes of college-educated white women by 25-30 points; Hillary Clinton won them by seven points. This group is already enthusiastic, as they were in 2018, to cast a vote against this administration.
  • Non-college-educated white women in the Upper Midwest have soured on Donald Trump and Biden is leading in the key states of Wisconsin and Michigan, in part, because of their movement. The race is much closer south of the Mason-Dixon line. In North Carolina and Georgia where the candidates are effectively tied, both college and non-college-educated white women are more likely to be evangelical and are more supportive of Trump. If Amy Coney Barrett’s Supreme Court nomination is confirmed prior to the election, one question is whether evangelical women, who may have serious reservations about Trump outside of his pro-life stance and the Supreme Court, would feel that they have what they need and refrain from voting for the president.
  • Another very important cohort for the Biden campaign to mobilize – rather than persuade – is black women. In the toss-up state of North Carolina, for example, Black turnout was down nearly 9% in 2016, in part due to difficulties with voting accessibility (some would say voter suppression efforts). In fact, Black turnout was down in many states, and ensuring that Black women – the most loyal Democratic cohort – turn out and are able to cast a ballot this year will be critical to Joe Biden’s success.
  • Finally, let’s consider senior women. In national polls, Biden has a small edge over Donald Trump with seniors, a group that went for the Republican candidate by seven-points according to 2016 exit polls. In the key state of Florida, as well as Arizona, success among senior women could tip the state. In the latest ABC/Washington Post polls in Florida and Arizona, Trump leads among seniors in these states. Senior women care about COVID-19 and its impact on health, but also on their ability to travel to see grandchildren; they care about prescription drug prices and the solvency of Social Security.    


Christina Bejarano is Professor of Political Science with a joint appointment with the Jane Nelson Institute for Women’s Leadership at Texas Woman’s University. She is the author of The Latino Gender Gap in U.S. Politics (Routledge Press, 2014) and The Latina Advantage: Gender, Race, and Political Success (University of Texas Press, 2013).

Mary-Kate Lizotte is Associate Professor of Political Science at Augusta University.  She is the author of Gender Differences in Public Opinion: Values and Political Consequences (Temple University Press, 2020).

Christine Matthews, a prominent public opinion pollster, is president and co-founder of Bellwether Research & Consulting. She has 25 years of experience in survey research and is a leading expert on the attitudes of women voters.


Susan J. Carroll

Susan J. Carroll is professor emerita of political science and women’s and gender studies at Rutgers University and senior scholar at the Center for American Women and Politics (CAWP) of the Eagleton Institute of Politics. Her books include: A Seat at the Table: Congresswomen’s Perspectives on Why Their Presence Matters (Oxford forthcoming 2018, with Kelly Dittmar and Kira Sanbonmatsu); Gender and Elections: Shaping the Future of American Politics (Fourth Edition, Cambridge 2018, with Richard L. Fox); More Women Can Run: Gender and Pathways to State Legislatures (Oxford 2013, with Kira Sanbonmatsu); Women and American Politics: New Questions, New Directions (Oxford 2003); The Impact of Women in Public Office (Indiana 2001); and Women as Candidates in American Politics (Second Edition, Indiana 1994). Carroll also has published numerous journal articles and book chapters focusing on women candidates, voters, elected officials, and political appointees in the United States. Carroll is a founder and former president of the Organized Section for Women and Politics Research of the American Political Science Association, and she currently co-edits the CAWP Series in Gender and American Politics, a book series published by the University of Michigan Press. She is the recipient of the Lifetime Contribution to Political Studies Award of the Political Studies Association in the United Kingdom as well as the Outstanding Professional Achievement Award from the Women’s Caucus of the Midwest Political Science Association. As a nationally recognized expert on women’s political participation, Carroll is frequently called upon for media commentary.