Then-Republican presidential nominee Donald Trump and Democratic nominee Hillary Clinton speak during a debate on Oct. 9, 2016, at Washington University in St. Louis. (John Locher/AP)

Two years after the 2016 election, there has been no single answer to the question: What happened? In an outcome that saw the popular vote and the electoral college diverge, theories abound, opinions are many and consensus fleeting. Now, a trio of political scientists have come forth with their answer as to why Donald Trump prevailed over Hillary Clinton, summed up in the title of their forthcoming book: “Identity Crisis.”

The co-authors are John Sides of George Washington University, Michael Tesler of the University of California at Irvine and Lynn Vavreck of the University of California at Los Angeles. They have plumbed and analyzed a wealth of polling and voting data, examined surveys of attitudes taken long before, during and after the 2016 campaign. Their conclusion is straightforward. Issues of identity — race, religion, gender and ethnicity — and not economics were the driving forces that determined how people voted, particularly white voters.

In an election decided by fewer than 80,000 votes in Wisconsin, Michigan and Pennsylvania, it’s been common for people to say that anything or everything could have made the difference. The three political scientists beg to differ: “ ‘Everything’ did not ‘matter’ equally,” they argue.

They are believers in the fundamentals of elections — things such as economic conditions, presidential approval, partisanship and the like. Fundamentals predicted that Clinton would win a greater share of the two-party vote, which she did. They dismiss some theories about what happened: If 2016 was really about anger and change, why did Clinton win the popular vote? Clinton’s popular vote victory, they note, was not in line with “casual punditry about voter anger but was in line with the state of the economy and approval of Barack Obama.”

They question other theories: They doubt, for example, that Russian interference determined the outcome of the election. The release of hacked emails in July and October 2016 “did not clearly affect” Clinton’s favorable ratings nor perceptions of her honesty, they write. They also say that, given the billions of tweets and social media postings during the campaign, Russian content was probably only an infinitesimal share of the total. Claims that the Russians turned the election should be greeted “with something between agnosticism and skepticism — and probably leaning toward skepticism,” they say.

But if some things were predictable, based on fundamentals, other things that happened were not. One was the unusual pattern of shifts among the states between 2012 and 2016. Normally, from election to election, states move in the same direction, albeit by different percentages. From 2008 to 2012, “almost every state shifted in the direction of the Republican candidate” due to economic conditions that were unfavorable to Obama.

In 2016, Clinton should have done a bit worse than Obama across the board. Instead, in some states — Arizona, California, Georgia, Massachusetts and Texas — she did better. In others she did about the same. And in some, Ohio and Iowa among them, she did “substantially worse.”

Oddly, she did better, comparatively, in red states — such as Georgia and Texas — than she did in a swing state like Iowa.

The cause for this was a divide among white voters, well documented during and since the election, a division that saw those with college degrees moving one way and those without college degrees the other. Sides, Tesler and Vavreck go step by step through the reasons for what they call the “diploma divide” among white voters.

There were some white supporters of Obama whose views on race and immigration were “out of step” with where the Democratic Party stood on those issues. The fact that the campaign focused on these issues — largely due to Trump’s rhetoric and consistency — voters’ perceptions of where the two candidates stood on identity issues was “further apart . . . than any major-party presidential candidates in 40 years.” Which, in turn, meant that those issues “became more strongly related to how they voted in 2016 than in any recent presidential election.”

It’s not that economic issues didn’t matter. But racial attitudes “shaped the way voters understood economic outcomes.” The authors describe this as “racialized economics” rather than economic anxiety. “Voters’ attitudes on racial issues accounted for the ‘diploma divide’ between less and better educated whites,” they write. “Economic anxiety did not.”

Their conclusion agrees with that of Alan Abramowitz, a political scientist at Emory University, who has long studied the rise of polarization in American politics and who focuses on racial resentment in his recent book, “The Great Alignment: Race, Party Transformation and the Rise of Donald Trump.”

Sides, Tesler and Vavreck reached their conclusion by highlighting data that show how identity issues became more important among white voters than in the past. What gives them confidence in their conclusion is that there were no similar signs in how economic issues affected white voters. They found “generally weak relationships between these measures of economic anxiety and how people voted in 2012 or 2016. Moreover, these relationships were not consistently stronger in 2016 than in 2012.”

Their definition of racialized economics is this: “the belief that undeserving groups are getting ahead while your group is left behind.” They say Trump played on these concerns throughout his campaign and has done so as president. They say “racialized economics” was more significant than economic anxiety in affecting how whites with different levels of education voted.

“These threads tell a straightforward story,” the authors write. The campaign focused on issues of identity more so than in the past. Trump and Clinton differed significantly on these issues, which activated them as important factors in shaping voting decisions. Additionally, there were many Obama voters whose views on these issues were “closer to Trump’s than to Obama’s or Clinton’s,” and many of them resided in battleground states. Their shift gave Trump primacy in the electoral college even as he was losing the popular vote.

The polarization around these issues predated Trump, but his campaign “magnified this polarization.” Sides, Tesler and Vavreck conclude the chapter on what happened by noting that the 2016 election was distinguished not only for the fact that Trump prevailed in the face of so many predictions that he would never be elected president, but also “for how it crystallized the country’s identity crisis: sharp divisions on what America has become and what it should be.”

They argue that the current “identity crisis” in America cannot easily be undone, even though public opinion “contains reservoirs of sentiment that can serve both to unify and divide.” The 2016 election did not produce the divisions in America, but it has embedded them deeper into the politics of the country. Voters and candidates will decide in coming elections whether to move in a different direction.