At the Intersection of Race and Gender: Historical Progress & Current Challenges

>> Elizabeth Elder: So we've spent some time now talking about groups in general, and I'd like to turn and spend the rest of our time talking about two groups that are especially important in American politics, race and gender. So, first, let's talk about race. This is a plot of the racial composition of the United States for the last 120 years.

As you can see, in 1900, about 90% of Americans were white, about 10% were black, and there was very little of anything else. And these days, that's very, very different. A little more than half of Americans are white. Still, 10 to 15% of Americans are black. And much, much larger proportions of the United States are Hispanic, Asian American, or other multiple racial groups.

So why does this matter for politics? People of different racial backgrounds feel differently from one another about politics. For example, here's a plot of partisanship over the last 30 years or so among people from different racial backgrounds. You can see that Asian Americans have come increasingly to ally with the Democratic Party.

Black Americans have for quite a while, voted for the Democratic Party. So have Hispanic Americans, although to a much, much lesser extent. And white Americans are a little more likely to vote for Republicans. Now, in this and in every other plot by race or gender that I'll show you, there is huge heterogeneity behind this.

No racial group is completely composed of people with the same positions. And I think a great illustration of this is looking at Hispanic and Asian Americans partisanship based on their national origin group. These are both umbrella categories that include a lot of people who come from really different backgrounds, different sets of norms, and political systems.

And so you see really heterogeneous partisanship. Vietnamese Americans, for example, are much more likely than Korean Americans to be Republicans. And people from Cuban backgrounds are much more likely to be Republican than people from Mexican backgrounds. Some people don't even really identify with these broader umbrella categories, but instead feel more strongly tied to their national origin groups.

So this is just to say there's a lot going on underneath these broad labels, and not everybody feels attached to these groups to the same extent. So, with that in mind, let's focus in a little bit on the differences in opinion between black and white Americans, which are pretty often one of the largest divides that you'll see in any public opinion survey.

So, on average, black and white Americans tend to feel differently from one another on issues related to race. I think this is not terribly surprising to anyone. Very different opinions about the extent of racial progress and prospects for future racial progress, somewhat, although not extremely different, attitudes on affirmative action.

With about half of black Americans and a third of white Americans approving, although this changes a lot based on how you ask the question. There are also racial gaps in public opinion on issues that don't really seem to have anything to do with race. White Americans are much less likely to support raising the minimum wage, for example, and black voters are much more interested in the issue of crime when it comes to voting in elections.

But of course, there are not always divides. People are not always divided on every issue. For example, here's some data from a recent poll on legalizing medication abortion, where black and white Americans opinions are almost identical. And here's the importance that voters assign to a bunch more issues in the recent midterms, you can see that, sometimes black, white, and Hispanic Americans care about different things, but on a lot of these, the dots are very close together.

There's a lot of consensus across racial groups on the issues that are important for the government to face. So why might it be that public opinion differs by race so dramatically, especially in the case of black and white Americans? One possibility is differing self-interest. There are some policies, like affirmative action or reparations, that kind of explicitly target racial groups.

And so maybe it's not really much of a puzzle why people from different racial backgrounds feel differently about those issues. Black and white Americans are also different from one another in ways other than their racial identities. Here, for example, is a plot of the net worth of median family for black and white Americans.

And as you can see, white Americans tend to have considerably more wealth than black Americans. So, maybe this explains why they have different opinions on economic issues. You might think that if black and white Americans had more similar economic means, their opinions on economic issues would be more similar to one another.

But it's actually not particularly common in political science research to find examples of self interest in a kind of personal, bottom line sense affecting people's policy opinions. People are generally a lot more concerned about ideas of the fairness of a policy, or if the recipients of a policy, or deserving whether it fits with their moral view of what the government ought to do.

And it seems like this is probably also a lot of what's going on with differences between racial groups. Someone's background tends to be on average, a lot more important to them. They come from a racial minority group, such as black or Hispanic Americans. And these groups can come with different norms about political values and expectations about the kinds of politics to expect.

And these norms are probably just as important. These norms and social dynamics are probably just as important as these ideas about different self-interest.

>> Elizabeth Elder: So we've talked about differences between black, and white, and Hispanic, and Asian Americans. But there are also differences within racial groups based on racial attitudes.

So here's an example of what I mean by that. Here's a plot of approval of interracial marriage since the 1960s. This number is shockingly low for a shockingly long time, I think. So, if I were running a survey in 2001, I could expect about two-thirds of my respondents to approve of interracial marriage and about a third of my respondents to disapprove.

And if I looked at how those people voted in the presidential election or how they felt about welfare reform, they had very different ideas from one another. These days, thankfully, almost all Americans approve of interracial marriage, and we think about racial attitudes fairly differently. Instead, we think about divides based on whether someone sees differences, like the racial wealth gap, as based on differences in individual effort or differences in structural factors allocating wealth.

And people who have those two sets of beliefs tend to have very different vote choices and opinions on a host of issues. Here's, I think, a really cool example of how powerfully race can affect people's voting decisions. And this comes from the 1960s. So this is some Gallup polling data in advance of the 1964 election.

This is from 1963, so the candidates in question are Democratic incumbent John F Kennedy and Republican challenger Barry Goldwater. The blue lines in this plot are opinions about these candidates among non southern white Americans, Californians and New Yorkers, and Montanans. And throughout this time period, from March through November of 1963, non southern whites feel about the same about Kennedy and Goldwater.

But the red lines represent opinions among Southerners, and those change really dramatically. At the beginning, these white Southerners are largely supporting Kennedy. This makes sense, a large majority of white Southerners were Democrats at this time. But then pretty abruptly in the Spring of 1963, the lines flip. These white Southerners start supporting Goldwater and are much less likely to support Kennedy.

So what happened in the Spring of 1963? Kennedy proposed a really large civil rights bill that would have changed a lot about the order of racial politics in the South. And the typical white Southerner really did not care for that civil rights package and came to disfavor Kennedy and approve more of Goldwater, whose racial politics were much more aligned with those of white Southerners at the time.

So again, there are only white voters in this image. This is not a divide between racial groups, but race is still really powerfully shaping the choices that people are making in elections. So I think this is just a great way to wrap up this section, by illustrating that race, both as a group membership and a set of group identities, is really powerful in shaping people's opinions and voting behavior in the United States.

So let's talk now about a case that is really, really different in a number of ways, and that is gender. The most striking difference here, I think, is that public opinion divides by gender are very small. Men and women generally agree with one another on most political issues.

If you think about presidential approval or desire to increase government health insurance spending, or even a gender related issue like abortion, men and women's opinions are very, very similar to one another. This is not always the case. Women tend to be a little bit more oriented towards a larger, more service providing government, tend to have different opinions on a few issues, like gun control.

And just to kind of watch the space, this might be changing. This is a plot of party identification among men and women of different generations. And you can see that among silent generation, baby boomers, and generation X, the gap between men and women is not very large. But there's something really striking happening in the top corner with millennial women, who are dramatically more likely to be Democrats than are millennial men.

So it's a little early to tell whether this is also happening with Gen Z, but it's possible that if I were to give this talk again in five years. I would have to revise my slide, that said, there were not big gender gaps in public opinion, because it seems like this might really be changing.

So, just as in the case of race, there can be divides between racial groups as well as divides based on racial attitudes. The same is true of gender. So maybe there aren't generally very big divides based on men and women, but there can be big divides based on people's gender attitudes.

Here's an example from the 1980s. There were a couple relevant foreign policy issues at the time where men and women had pretty different opinions from one another. And this paper looked at opinions among men and then women who said that they did not consider themselves feminists, and women who did say they considered themselves feminists.

And what you can see here is that the men and the women who are not feminists have quite similar opinions on these issues, whereas the women that are feminists are feeling somewhat differently. So in this case, we're not seeing so much gender gap as a feminism gap. But as I've said, there are generally less reasons to think that gender is a really important factor in public opinion in the United States.

But where there are big differences, in some cases, is participation. So I'll spend the rest of the time on that. So men and women vote at similar rates. Women tend to vote a couple percentage points higher than men, and this has been true for a long time. But men tend to have higher political knowledge, greater political interest, and higher rates of other forms of political participation.

This is some data from a paper from 2019. The first several bars show knowledge by men and women of the partisanship of their representatives, and the partisan control of their representative bodies. Men are much more likely to get these questions right. Men are much more likely to follow politics.

They're voting at slightly higher rates in this data source, but they're much more likely to do other kinds of election related activities, like donating to campaigns or volunteering for campaigns. So why might this be? It's possible that there are psychological explanations for this. Women tend, on average, to be a little bit less self confident than men, a little bit more self monitoring.

So again, no need to read this one super carefully, but this is some data from a study that looked at confidence in teen girls and their participation rates as young women. And it shows that the fact that the teen girls were a little bit less confident than their male classmates, explains part of why they were voting at lower rates and donating to campaigns at lower rates in adulthood.

Women also tend, on average, to be a little bit more conflict averse than men, a little bit less conflict seeking. And so to the extent that electoral politics involves a lot of conflict, that could help explain why men might be on average, a little bit more drawn to it than women are.

But it's also possible that these gaps are just because we measure mostly things that men are more interested in and less things that women are more interested in. So, for example, if you look at knowledge about representatives, when those representatives are women, the gender gaps close a lot.

Women are just as likely to know the names of their female representatives as men are. And they were actually more likely to participate in protests over the last couple of years. So it's possible that if we were measuring more the cases where women felt more involved in the stakes of politics, that these gender and interest gaps might go away.

The biggest place where gender gaps show up in electoral politics, I think is in elected office. In general, women hold right now about a quarter of seats in Congress, Governorships, Mayorships, about a third of seats in State Legislatures, and Municipal offices. So why is this the case? There's been a ton of research trying to address this, and I'm gonna show you some figures from a really great study that was done about 15 years ago.

In which some political scientists interviewed the kinds of people who are likely to run for office at some point in their lives. These are people in their twenties and thirties, with backgrounds in business and law, and education, and healthcare. And they asked them various questions about the prospect of running for office at the point when they hadn't really started to take any steps toward that yet.

So one possibility for differences between men and women in elected office is differences in ambition. Women are just less likely to make it a goal of theirs to run for office. So the authors asked men and women whether they had considered running for office or taken any of a number of steps to make that happen, and the women were less likely to say they had done so.

One possible reason related to this is that women have less of a taste for the kinds of activities that running for office involves. So the authors showed them this list of things that people have to do when they run for office, and asked them, how likely is it that this is so negative it would deter you or prevent you from running for office?

And so these are things like calling people and asking for money, or running attack ads on your opponents, but also things like missing out on your career prospects and losing your personal privacy. And in every case, women are just much more reacting to these, much more negatively than are men.

So if you think about the numbers we saw earlier on conflict seeking versus conflict aversion, this might be related to that. Women are also less likely to say that they are qualified to run for office. And this is especially interesting because the authors looked at the characteristics of the men and women in their study, and they found that they had attended identical numbers of public meetings.

And had identical previous amounts of public speaking experience and policy research experience. By any objective measure, the men and women in the sample were equally qualified. But their feelings that they had about how qualified they were were quite different. So this, again, might have to do with the self confidence explanation that we noted earlier.

But there are a number of other explanations that they uncovered which are kind of moving away from the psychological and more towards what's going on in the world around these potential candidates. For example, most people do not decide one day to run for office and do everything by themselves.

Usually what happens is people get an invitation, they get a conversation from their local party chair, a member of their school board, and that person says, have you ever thought about running for office? And that might seem like kind of a simple set of words, but that also signals a connection to someone who knows who your first calls should be to get donors.

Who knows whom you should talk to about getting your name on the ballot. Who knows who to talk to about running at the next level of office. So these connections to kind of local gatekeepers for elected office are really important. And men were noticeably more likely than women to say that they had been approached by these people and asked about running for office.

Women and men also have fairly different sets of resources and structures in their day to day lives. They have different things going on in their household. The women, again, this is a sample of pretty professional, qualified men and women. The women are a little bit less likely to be married, more likely to be divorced or single, less likely to have kids.

Dramatically more likely to have sole responsibility for taking care of their children, or for taking care of their household, and spend many more hours a week on this. And again, these are people from similar occupational backgrounds. So this suggests that the women in this sample, even if they have the same kind of professional qualifications, are less likely to have a two income household that might be able to shoulder the burden of funding a campaign or time off from work for a campaign.

And less likely to have kind of extra hours in the day to devote to the really full job of running for office. So, this is not really about anyone's attitudes or about anyone's bias. It's just kind of the patterns to men and women's lives are different in ways that can shape their ability to run for office.

And you see this echoed in this study that looks at things a little bit further down the pipeline. So this is a study of State Legislators looking at whether they make the leap then to run for Congress. And you see some similar patterns to what we saw in the previous table.

So by the time women get to the State House, they're on average older than men. They're less likely to be in this core kind of 35 to 50 age group where people often make the leap to national politics. They're less likely to be married. And so these are all ways in which they might have a little bit less kind of an opportunity to make it very far down the pipeline and have a little bit less support in doing so.

And they're also, if you look at the second to last line, differently likely, even at the level of the State Legislature, to come from these backgrounds of business or law that are really desirable in running for national politics especially. So these are just more examples of the ways in which the other kinds of life choices that men and women make structure their ability to run for office.

You'll notice one thing I have not mentioned here is voter bias, whether women are discriminated against at the ballot box. It's kind of hard to say whether that's the case, but the best evidence suggests it's not. Voters are happy to vote for a woman just as much as a man as a candidate.

So when we're thinking about increasing the number of women in elected office, it's more about addressing these kind of resource explanations. The confidence and time and money that women need to run for office at rates. And of course, this has been getting better. There have been a number of changes in society, as well as targeted programs encouraging and providing resources to women candidates.

There are more women in elected office than ever, and this is continuing to increase. And hopefully, if I gave this talk again in five years, some of you would be in this chart and the numbers would look even better. So to summarize what we've talked about today, we started with the notion that most people do not pay attention to politics.

Those that do generally don't have very strong or meaningful issue opinions or coherent ideologies. People do tend to know a lot about social groups and to feel strongly about those social groups. And those groups give people both affective ties to and information that they can use to make sense of politics.

We talked about race and racial attitudes and how these things can structure divides in public opinion both within and between racial groups. And finally, we talked about gender and the way that it changes really deeply how people engage with politics, and in particular how and whether they run for office.

So with that, I'll open it up to Q&A, thank you all so much for your attention.