Why Hillary Clinton Won the Popular Vote

An evolutionary perspective on the two surprises of the 2016 election

Posted Jan 03, 2017

There were actually two surprises in the 2016 election.  On the one hand, the biggest shock to almost everyone I know in the academic world is the fact that the office of president of the United States will soon be occupied by a fellow who educated people widely perceived as a buffoonish reality TV star (whose only political experience was ranting about President Obama being a secret Muslim).  His opponent, after all, had been a senator, secretary of state, and first lady during a period of American prosperity.

But, considered in the light of the history of our country, the other surprise is that the first woman to run for president actually won the popular election.  And not by a small margin, but by 2,865,075 votes, according to the latest count I’ve seen.  Indeed, many of the people I know are even more distressed by the fact that, in any other democracy, Clinton would be president, and Trump wins because of a historically unfortunate technicality which makes the U.S. presidential election imperfectly representative of the will of the majority. 

Considering the first surprise, how could Donald Trump have won a presidential election, technicality or no technicality?  Much of the reaction I’ve heard has talked about what social psychologists call counterfactuals, “what if” fantasies about alternative realities that might have happened.  What if the FBI director hadn’t sent out that tainted false alarm a week before the election?  What if pre-election polls hadn’t led some Democratic voters in swing states to feel it was unnecessary to vote?  Besides counterfactuals, there has been a lot of depressed resignation about the state of the electorate, sometimes combined with moral indignation:  How could so many voters be so racist/sexist/naïve/ and self-defeating?

And what of the second fact: how did the first woman to ever run for president win almost 66 million votes (almost 3 million more than her male opponent)?

Why the election turned out two ways

Perhaps there is a simple, naturalistic, explanation of why this election turned out two ways – an explanation that has almost nothing to do with the actual qualifications of Donald Trump or Hillary Clinton.  The outcome may be related to an important, but relatively little-known, research finding: When humans are concerned about threats from other tribal groups, they are more likely to favor a man as a leader.  But when they are concerned with getting along with their neighbors, they strongly favor a woman. 

Mark Van Vugt and Brian Spisak conducted several studies examining what they called the “male warrior hypothesis:” the idea that throughout history, males have typically been the ones to make war.  Anthropologist Richard Wrangham has observed how male groups in hunter-gatherer societies often form fighting gangs to go out and fight with males from other territories.  Women never do this.  The tendency for warmaking to be a predominantly male domain is also found in chimpanzees, and it has persisted into the modern world, from Genghis Khan to Vietnam and Afghanistan. 

Van Vugt and Spisak conducted several studies to examine how this ancestral sex bias might work its way into electoral decisions.  In one study, they told students they would be playing a game in which they could make money, and would either compare their results to other members of their own group, or – in the group conflict condition – would compare their results to those of other teams at other universities.  The study was done at Kent University in Canterbury, and the opponents were at other British universities that were traditional rivals of Kent’s soccer team.  They were asked to choose a leader for their group, either ‘‘Sarah,” who had described herself as a “21-year-old law student whose hobbies are exercise, traveling, and going out with friends.’’ or ‘‘Peter, a 20-year-old university student in English literature. His hobbies are reading, making music, and attending parties.” When there was no conflict with other universities involved, over 90% of students chose Sarah as the leader.  But when there was going to be a conflict with rival schools, the results flipped, with 80% preferring Peter over Sarah.

In another study, the researchers asked students to imagine they were citizens of a hypothetical country called Taminia, during a presidential election.  Some of the subjects were told their country was at war with another country. Others were told they were facing civil unrest within their own borders.  When the problem was for Taminians to get along with another, 76% chose a woman as president.  But when the problem was war, results again flipped, and 91% chose a man (See the figure).    

Van Vugt and Spisak, from the research described here, with permission
Source: Van Vugt and Spisak, from the research described here, with permission

These studies provide a more distal perspective on the results of the 2016 American election.  Trump’s continual focus on the dangers posed by members of groups of “outsiders,” such as illegal Mexican immigrants and Fundamentalist Muslim terrorists, framed the current U.S. situation as one of war.  Trump also argued that trade pacts with other nations were a threat to the jobs of marginally employed lower middle class Americans.  Further, the intellectual elites proposing environmental regulations were also painted as enemies, secretly plotting with wealthy Wall Street millionaires to close factories and coal mines and take poor people’s jobs away, not to mention their guns. 

Van Vugt and Spisak’s findings help us understand why Trump’s message resonated with less educated poorer White people living in the Deep South and Midwest, but not with educated White voters living in urban areas.  Educated urbanites are less concerned with immigrants taking their jobs away, and more concerned with getting along with their neighbors and coworkers, who are likely to come from diverse religious and cultural backgrounds.  If the problem is getting along with our diverse neighbors, then the research suggests that we strongly prefer a woman leader.  Indeed, the strength of that preference was sufficient that Hillary Clinton was able to trump Trump by 2,564,000 votes in the popular count.  

In another series of studies, Van Vugt and Spisak found that, even when comparing two men or two women, the more masculine of the pair was chosen as president when the context was a war.  This suggests another twist on the election, one that is counterintuitive at first glance.  The features of Trump that educated urbanites found most reprehensible – his history of sexual harassment and assault, and his prancing menacingly around the stage during a televised debate with Hillary Trump – may have ironically helped him win the election amongst those portions of the population most concerned about attacks from outsiders.  How so? These behaviors made him appear even more masculine.  Conversely, Hillary Clinton’s history of sensitive and charitable behavior toward children and downtrodden minorities, made her appear more feminine.  (click here for a follow-up containing a video link discussing Van Vugt's research on gender and leadership in more detail). 

Deep rationality and the naturalistic fallacy

Vlad Griskevicius is a professor of marketing at the University of Minnesota, whose research delves into the evolutionary bases of human decision-making. In the book The Rational Animal, he and I discuss a number of ways in which human decision-making falls short of the classical ideal of rationality, and instead often falls prey to simplistic biases left over from the caveman era.  Those biases often served our ancestors well, but they are often mismatched to the demands of the modern world.  In this light, Trump’s election may be just another example of human nature in action. Is this more or less comforting than simply wringing our hands about ignorance, irrationality, and chaos?  It depends.  It is useful to remind ourselves of the “naturalistic fallacy” – the misconception that calling something “natural” is also calling it “good.”  Some natural phenomena, such as the song of mockingbird, the scent of a flower, or the taste of a mango, certainly seem good from the perspective of our human sensibilities.  Others, such as wasps who paralyze their prey, and then lay their eggs inside the victims’ bodies, or the amazing success of some disease organisms at defeating our immune systems, do not seem so good at all.  The general tendency to choose belligerent males as leaders during wartime might in fact prolong wars, as other research suggests that opponents tend to respond to threats by doubling down on their own authoritarian and militant tendencies.  Not every product of our ancestor’s natural history is a good thing in the modern world.  

Douglas T. Kenrick is author of:

The Rational Animal: How evolution made us smarter than we think, and of: 

Sex, Murder, and the Meaning of Life: A psychologist investigates how evolution, cognition, and complexity are revolutionizing our view of human nature.

Related blogs:

Why Hillary won the popular vote, Part 2: This includes a 3 minute video in which Van Vugt describes his research, as well as my response to some of the angry commentators, explaining why it's not time to stop "whining" quite yet.  

Does Trump’s election disprove the existence of God?  Political psychology meets the psychology of religion.

Would you wear a Great/Together t-shirt

Why are THEIR political views so blatantly self-interested? Revealing the hidden agenda of the political mind.

Atheistic liberals ARE smarter, but for a funny reason.


Van Vugt, M. & Spisak, B.R. (2008). Sex differences in the emergence of leadership during competitions within and between groups. Psychological Science, 19, 854-858.

Spisak, B.R., Dekker, P.H., Krüger, M., & van Vugt, M. (2012).  Warriors and Peacekeepers: Testing a Biosocial Implicit Leadership Hypothesis of Intergroup Relations Using Masculine and Feminine Faces.  PLoS ONE, 7, e30399.