The evidence and messaging are clear – one of the most important ways that people can protect themselves and others from the spread of Covid-19 is to frequently wash their hands. And yet, according to survey data, men are still washing their hands less often than women. Using survey data, Dan Cassino finds that men who feel that they have the same chance of dying from Covid-19 as women are less likely to wash their hands, potentially because they are doubling down on their masculine identities in the face of a threat that they can’t control.
In survey after survey, we see that American men just aren’t washing their hands at the same rates as American women. At a moment when health experts are begging Americans to wash their hands frequently in order to help stop the spread of the coronavirus, why are American men lagging? In recent data – collected between March 10th and 16th by USC’s Understanding America Survey – 91 percent of women say that they’re washing their hands more frequently because of the Covid-19 pandemic, compared with only 85 percent of men. At the same time, Democrats are more likely to report washing their hands frequently than Republicans, at rates mirroring the gap between men and women.
So, what’s going on here? Are men washing their hands less frequently out of some misplaced machismo? Or does is only seem like men are less likely to wash their hands because they’re more likely to be Republicans, a group that, until very recently, has been skeptical about the threat posed by the virus? The answer is important: if it’s just about partisanship, we should be able to increase hand washing behavior by changing the messaging from political leaders. If it’s about masculinity, it may be much harder to change.
One clue to the difference comes from the fact that men are, on average, less concerned about the coronavirus. While men and women are about equally likely to think that they’ll get the virus – both put their chance, on average, at about 20 percent – as Figure 1 shows, women are much more likely to think that they’ll die from it. The mean woman in the US thinks that there’s a 15 percent chance that she’ll die from coronavirus, while the mean man thinks that there’s only about a 13 percent chance. That may not sound like a huge difference, but when we dig into the numbers, it turns out that women are much more likely to think that there’s a significant chance of them dying. Twenty-three percent of women say that there’s a 25 percent chance (or greater) that they’ll die, compared with just 18 percent of men.
Figure 1 – Reported perceived likelihood of respondent dying of coronavirus, by gender
There’s a similar pattern based on the partisanship of the respondent (measured in a separate survey, carried out in January, and combined with the March data). Figure 2 illustrates that Democrats are less likely to think that there’s no chance that they’ll die from the Coronavirus than either independents or Republicans.
Figure 2 – Reported likelihood of respondent dying of coronavirus, by partisanship
It turns out that these differences add up. The more scared Americans are of getting the coronavirus, the more likely they are to take preventative measures, like washing their hands. Only 83 percent of Americans who don’t think there’s any chance they’ll get the virus are washing their hands frequently (that’s about 20 percent of the US), but that goes up to 92 percent among those Americans who think there’s a 50 percent chance that they’ll catch it.
The simplest way to sort out the differences is to build a statistical model that controls for all of the relevant factors (the state someone lives in, their income, age, race, education, whether they are working from home, and so on), as well as partisanship, sex, and the perceived likelihood of dying from coronavirus. Such a model will show us which of these effects are directly driving behavior, and which are only doing so indirectly: for instance, if party only seems to matter because men are more likely to be Republicans.
“IMG_20180517_100515” by Tom Page is licensed under CC BY SA 2.0
The results of the model are surprising. It turns out that partisanship doesn’t really matter. Politically, what matters is support for Donald Trump in the 2020 election. Individuals who support Trump – male and female, across the partisan spectrum – are about 10 points less likely to report washing their hands than those who support presumptive Democratic nominee Joe Biden, or haven’t decided. This is rather larger than the expected difference between men and women, which, controlling for all other factors, is about 6 points.
But larger than both of these is the effect of fear of the coronavirus, which has a bigger impact than either electoral politics or sex – but only among men. As Figure 3 shows, men who say that there’s no chance that they’ll get the virus have only a 79 percent chance of saying that they’re washing their hands frequently. This goes up to 91 percent among those men who give themselves a 6 to 25 percent chance of dying from it. What’s especially interesting though is the effect of very high levels of fear. Among men – but, again, not women – thinking that there’s a 25 percent chance of dying or greater leads to a decrease in reported handwashing, down to an 85 percent likelihood of frequent handwashing.
Figure 3 – Expected percentage washing hands, by perceived threat of dying
What’s going on here? It seems like masculinity may literally be killing men. In past research, I’ve shown that conditions of threat can lead men to double-down on behaviors that they perceive as masculine. So, for instance, times of economic threat lead men to be more likely to embrace conservative political views, buy guns, or even engage in sexual harassment in the workplace. In this case, when men see the coronavirus as a serious, one that they can’t fight back against, they look for some way to assert their masculine gender identity, in this case by proclaiming that they don’t need to wash their hands, that they’re tough enough to resist the disease without taking any steps.
The same pattern can be for those saying that they haven’t changed their behavior at all in the face of the Coronavirus (Figure 4). Among women, this percentage falls as their own perceived risk of dying increases, from 40 percent among those women with no perceived risk to 31 percent among those with the greatest. Among men, the percent saying that they’re doing nothing starts higher, drops, then rebounds at higher levels of perceived risk. Men who think that they are very likely to die from the coronavirus are more likely to be doing nothing about it.
Figure 4 – Perceived chance of dying from coronavirus among those who report having done nothing
The good news is that some of this gap between men and women should be fixable. As a failure to wash hands is linked to support for President Trump, the President’s change in messaging may well lead his supporters to wash their hands more. More dangerous, however, is the way in which men are expressing their masculine gender identities in the face of the crisis, by avoiding preventative behaviors as the threat level increases. As the virus continues to spread, the perceived danger is only going to increase, and if men continue to play macho by pretending that they don’t have to do anything to keep themselves and their families safe, it’s only going to get worse. That’s not to say that there aren’t solutions, though: messaging about the coronavirus should focus on other aspects of masculinity: rather than toughness, men should be told that they can be ‘real men’ by protecting their families and communities. If we’re not able to get this message through, this misplaced masculinity is likely to kill them.
Notes: This article gives the views of the author, and not the position of USAPP – American Politics and Policy, nor the London School of Economics.
The project described in this paper relies on data from survey(s) administered by the Understanding America Study, which is maintained by the Center for Economic and Social Research (CESR) at the University of Southern California. The content of this paper is solely the responsibility of the authors and does not necessarily represent the official views of USC or UAS.
Shortened URL for this post: https://bit.ly/2xf93I4
About the author
Dan Cassino – Fairleigh Dickinson University
Dan Cassino is an associate professor of Political Science at Fairleigh Dickinson University in Madison, New Jersey, who studies political psychology and polling.