The 2020 presidential election between Joe Biden and Donald Trump was shaped by several intersecting demographic forces, from a widening gap between college-educated and non-college-educated voters to a stark urban-rural divide that grew even deeper than in 2016. Record turnout, driven partly by the COVID-19 pandemic’s expansion of mail voting, brought new voters into the electorate while reshuffling how and where existing voters cast their ballots. Understanding these patterns requires looking beyond simple red-versus-blue maps and into the specific communities whose shifting allegiances decided the outcome.
The Diploma Divide
If one demographic trend defined the 2020 election more than any other, it was the growing gap between voters with and without college degrees. Research examining county-level data from 2000 through 2020 found that the percentage of a county’s population holding a bachelor’s degree was one of the strongest predictors of how its vote share changed over that period. Highly educated counties trended sharply toward Democrats, while counties with lower levels of educational attainment moved toward Republicans.1Political Research Quarterly. TRENDS: Diploma divide: Educational attainment and the realignment of the American electorate This wasn’t a sudden 2020 phenomenon. The education-based realignment had been building for two decades, but 2020 represented its most pronounced expression yet.
The practical effect was visible everywhere. Suburbs around major cities, which tend to have higher concentrations of college graduates, swung toward Biden compared to previous Democratic nominees. Meanwhile, small towns and rural communities without large college-educated populations continued their drift toward Trump. This created a feedback loop in campaign strategy: Democrats increasingly tailored their messaging to suburban professionals, while Republicans leaned harder into appeals to working-class voters without four-year degrees. The diploma divide cut across traditional racial and regional lines, reshaping coalitions in both parties.
Urban Turnout Versus Rural Margins
The rural-urban split in 2020 was a study in competing forces. Trump actually increased his margin of victory in rural America compared to 2016, winning an even larger share of the rural vote the second time around. But that wasn’t enough. Voter turnout in urban areas surged, an even larger share of those urban voters chose Biden, and that combination flipped the result.2American Behavioral Scientist. Donald Trump and changing rural/urban voting patterns
This dynamic played out in state after state. In Georgia, massive turnout in the Atlanta metro area overwhelmed Trump’s gains in the state’s rural counties. In Arizona, the Phoenix and Tucson metro areas delivered enough votes to offset deep-red margins elsewhere. The lesson of 2020 was that winning rural voters by bigger margins doesn’t help if the other side is pulling more voters to the polls in cities. It also highlighted a structural challenge for Republicans going forward: rural areas have finite populations, so there is a ceiling on how many additional votes can be squeezed out of them, while urban and suburban areas continue to grow.
Race and Ethnicity Across the Electorate
Racial and ethnic demographics in 2020 told a more complicated story than simple narratives about “the minority vote” would suggest. Different communities moved in different directions, and even within single racial categories, geography and age produced wildly different voting patterns.
Asian Americans drew particular attention as the fastest-growing racial group in the country and an increasingly consequential voting bloc. Research into Asian American turnout and vote choice in 2020 found that racial considerations informed the politics of this group in ways that depended heavily on context, including experiences with anti-Asian sentiment during the pandemic and engagement with specific policy issues like immigration and healthcare.3PubMed Central. The Asian American Vote in 2020: Indicators of Turnout and Vote Choice The bloc leaned heavily Democratic overall, but treating Asian Americans as a monolith missed significant variation among ethnic subgroups with different immigration histories and political traditions.
Native Americans played a pivotal role in at least one battleground state. In Arizona, which voted Democratic for the first time in decades, Native Americans turned out in record numbers and were credited as one of the key demographic groups that flipped the state to Biden. These voters were predominantly young adults, and their mobilization reflected years of grassroots organizing on and near reservations.4American Behavioral Scientist. “I’m Going to Vote for the First Time in My Life this Year”: Native American Political Engagement in the Trump Era Native voter turnout is often overlooked in national analyses because the population is small in most states, but in a contest decided by roughly ten thousand votes in Arizona, it was decisive.
Latino voting patterns generated heated debate after the election. While Biden won the Latino vote overall, Trump made gains among Latino voters compared to 2016, particularly in South Florida and along the Texas-Mexico border. The reasons varied by community: Cuban Americans in Miami responded to messaging about socialism, while some Mexican American voters in the Rio Grande Valley cited economic concerns and cultural conservatism. The simplistic assumption that demographic growth among Latinos would automatically benefit Democrats took a serious hit in 2020, forcing a rethinking of outreach strategies.
The Generation Gap and Gen Z Independents
Age remained one of the most reliable predictors of vote choice. Younger voters broke for Biden by wide margins, while older voters, particularly white voters over 65, leaned toward Trump, though Biden improved Democratic performance among seniors compared to 2016, partly due to concerns about the pandemic’s impact on older adults.
Within the youngest cohort, though, the picture was more nuanced than it appeared. A survey of over 1,300 registered voters in Arizona between the ages of 20 and 30 found that Gen Z voters were decidedly more independent and more Latino than their party-affiliated counterparts, and that they had significantly lower voting participation rates.5Election Law Journal: Rules, Politics, and Policy. Understanding the Gen Z Independent Voter: An Exploratory Study In other words, while young people who did vote tended to favor Democrats, a large share of Gen Z identified as independent rather than aligning with either party, and many didn’t vote at all. The popular image of a uniformly liberal young electorate masked a meaningful chunk of young voters who were disengaged or persuadable.
This matters for interpreting 2020 because mobilizing young voters was a major priority for the Biden campaign. Organizations poured resources into reaching voters under 30 through social media and campus outreach. The result was higher youth turnout than in 2016, but there was clearly still a large pool of young people who stayed home or registered no party preference. Whether those non-voters represent a future Democratic reservoir or a genuinely unaligned group remains an open question.
Religion and the Enduring Divide
The religious landscape of the 2020 electorate showed deep divisions, but they were largely the same divisions that had existed in 2016. White evangelical Protestants remained overwhelmingly loyal to Trump, while religiously unaffiliated voters, Black Protestants, and Jewish voters broke heavily for Biden.6The Forum. Religion and the 2020 Presidential Election: The Enduring Divide The stability of this pattern is itself worth noting. Despite four years of controversies, impeachment, a pandemic, and racial justice protests, the religious coalitions barely budged.
White Catholics were more evenly split, leaning slightly toward Trump in most analyses, while Hispanic Catholics were more likely to favor Biden. The religion gap was as much about race as theology: white Christians of nearly every denomination leaned Republican, while Christians of color leaned Democratic, regardless of how frequently they attended services. Biden’s own Catholicism didn’t dramatically reshape Catholic voting patterns, though it may have given him a marginal advantage with some Catholic moderates who were uncomfortable with Trump’s style.
The most politically significant religious trend wasn’t a shift between 2016 and 2020 but a longer-term change in the electorate’s composition. The share of Americans identifying as religiously unaffiliated has been growing steadily, and these voters favor Democrats by large margins. As the religiously unaffiliated portion of the electorate expands, it puts structural pressure on the Republican coalition’s reliance on regular churchgoers.
How the Pandemic Changed Voting Methods
COVID-19 reshaped the mechanics of the 2020 election in ways that interacted directly with demographics. The expansion of mail-in and early voting options led to the largest use of vote-by-mail in American history, but the shift wasn’t evenly distributed across the population.
Democrats were far more likely to vote by mail than either independents or Republicans, and independents were more likely to vote by mail than Republicans. Older voters across all partisan groups were more likely to use mail ballots than younger voters. The magnitude of the shift was dramatic: the youngest Democrats (ages 18 to 49) were about three times more likely to vote by mail in 2020 than in 2016, while Democrats 65 and older were roughly four and a half times more likely to do so. Republicans also increased their use of mail voting, but the effects were smaller, especially among younger Republican voters.7PLoS ONE. Should I vote-by-mail or in person? The impact of COVID-19 risk factors and partisanship on vote mode decisions in the 2020 presidential election
This partisan asymmetry in voting methods had real consequences on election night. Because many states counted in-person votes first and mail ballots later, early returns skewed Republican before shifting toward Biden as mail ballots were tallied. The so-called “red mirage” and “blue shift” were direct products of this demographic sorting by vote mode. It also created a political flashpoint: Trump’s pre-election rhetoric discouraging mail voting among his supporters likely contributed to the lopsided partisan split in how people cast their ballots, which then fed into post-election disputes about the legitimacy of mail-in votes.
When Mail Ballots Got Rejected
The surge in mail voting also brought new problems. First-time mail voters were learning an unfamiliar process, and not everyone got it right. In Philadelphia, researchers studying the 2020 general election found that voters from racial and ethnic minority backgrounds were more likely to have their mail ballots rejected for procedural errors, such as returning a “naked ballot” (one missing the required inner secrecy envelope). Language barriers contributed to the disparity.8Election Law Journal: Rules, Politics, and Policy. Whose Vote Is Lost by Mail? Evidence from Philadelphia in the 2020 General Election
Interestingly, younger voters were actually less likely to make these errors, possibly because widespread outreach on social media helped educate them about the correct process for completing a mail ballot. Older voters, despite their greater experience with elections, may have been less exposed to the targeted digital campaigns explaining new mail voting procedures. The finding undercuts the assumption that younger voters are inherently less reliable when it comes to the mechanics of voting. In this case, their comfort with social media translated directly into fewer rejected ballots.
These disparities in ballot rejection rates raise serious equity questions. If minority voters are disproportionately losing their votes to procedural technicalities, the expansion of mail voting doesn’t necessarily expand access equally. States that adopted or expanded mail voting in 2020 vary widely in how forgiving their rules are: some allow voters to “cure” defective ballots before certification, while others do not. Where you lived determined not just how you voted but how likely your vote was to count.
Union Households and Down-Ballot Behavior
Union membership has long been associated with Democratic voting, but the relationship has grown more complicated. In 2020, Biden performed better with union households than Hillary Clinton had in 2016, partly recapturing some of the working-class voters Trump had peeled away. But the union vote is no longer a reliable Democratic lock, especially among white union members in industries like construction and manufacturing.
Research using data from the Cooperative Election Study across 2012, 2016, and 2020 found that union voters themselves were more likely than non-union voters to cast ballots in down-ballot races for the House, Senate, and governor, suggesting higher overall civic engagement. However, members of union households who were not themselves union members were actually less likely to vote down-ballot than non-union voters.9Labor Studies Journal. Off to the Polls and Then Down the Ballot: Are Union Members Less Likely to Roll-Off? The distinction matters because union household size is much larger than union membership itself. A union steelworker and that steelworker’s spouse may share a household but not the same level of political engagement or party loyalty.
For Democrats, this meant that winning union endorsements didn’t automatically translate into full household support. And for Republicans, the appeal to union households wasn’t just about convincing the union member; it was often about reaching the family members who lacked the mobilizing infrastructure of union political programs. The erosion of union density over decades has also weakened the mechanism by which unions historically turned out voters, making each election cycle’s labor outreach less potent than the last.
Why the Polls Got Demographics Wrong Again
Pre-election polling in 2020 faced significant challenges in accurately capturing the demographic composition of the actual electorate. Polls systematically underestimated Trump’s support for the second consecutive cycle, and the errors were larger in 2020 than in 2016 in many states. Researchers studying pre-election polls across multiple states found that selection bias, meaning that the people who chose to respond to polls were not representative of the people who actually voted, was a persistent problem that standard demographic weighting couldn’t fully correct.10PubMed Central. Evaluating Pre-election Polling Estimates Using a New Measure of Non-ignorable Selection Bias
The demographic roots of the polling miss were tangled. Non-college-educated white voters, a group that broke heavily for Trump, were less likely to respond to surveys. The pandemic made this worse: lockdown conditions may have increased response rates among people who were staying home (disproportionately Democrats) while decreasing them among people going about normal life (disproportionately Republicans in 2020). Pollsters attempted to correct for these imbalances through demographic weighting, but if the non-responders within a demographic category differed politically from the responders, weighting by demographics alone couldn’t fix the problem.
The consequences of this polling miss extended beyond prediction markets and media narratives. Campaign resource allocation, get-out-the-vote targeting, and even voter perceptions of the race’s competitiveness were all influenced by polls that painted an inaccurately rosy picture for Biden. In several states, what polls suggested would be comfortable Biden victories turned out to be nail-biters or losses. The demographic miscalibration of polls is now one of the central methodological challenges facing election research, with no consensus solution yet in place.
The Suburban Realignment in Practice
Much of the 2020 election’s outcome can be traced to suburbs, where the education divide and the urban-rural divide collide. Suburban counties with high concentrations of college-educated residents, particularly women, swung decisively toward Biden. This was especially pronounced in the Sun Belt: the suburbs of Atlanta, Phoenix, and Houston all moved significantly toward Democrats compared to their 2012 or even 2016 baselines.
The suburban shift was driven by a combination of factors that overlapped with other demographic trends. College-educated suburban women were among the most anti-Trump demographics in polling and exit data. Many suburban communities had also become more racially diverse over the preceding decade, as Black, Asian American, and Latino families moved to outer-ring suburbs previously dominated by white homeowners. The result was that “suburban voter” in 2020 meant something different demographically than it had even a decade earlier.
For Republicans, losing ground in the suburbs while gaining in rural areas created a geographic math problem. Suburban counties tend to be high-population, so even modest swings produce large absolute vote changes. Rural gains, by contrast, happen in low-population counties where running up the score yields diminishing returns. The 2020 map illustrated this vividly: Trump’s rural margins were often historically large, but in states like Georgia and Arizona, they simply couldn’t overcome the volume of votes coming out of suburbia. Whether this suburban realignment persists will depend heavily on whether the education-driven and diversity-driven trends that powered it continue, or whether new issues scramble the picture in future cycles.