US presidential elections serve as the "Raison D'être" for public opinion polls, significantly increasing interest in the methodologies used to measure public opinion. This heightened focus has led to theoretical contributions in the field of statistics and their practical application, enhancing the ability to use these polls for decision-making and future forecasting in an uncertain electoral environment.
During elections, two main types of public opinion polls are conducted. The first type, pre-election polls, are typically designed to understand voters' attitudes towards the electoral process and candidates. When these polls are constructed with a well-structured sample and an adequate sample size, they provide accurate estimates of the relative weight of candidates among different social segments, including age groups, gender, geographic regions, education levels, and races. Campaign organizers find these results particularly useful in identifying weaknesses, allowing them to reprioritize their efforts to win over the largest number of voters.
The second type, exit polls, are conducted after voters cast their ballots. In these polls, voters are asked about their chosen candidate, as well as demographic information such as age, gender, place of residence, education level, economic status, and race. Additionally, questions relevant to the country's general context may be included. For instance, in the United States, voters are asked about their political leanings (liberal, moderate, conservative) and their stance on important issues where candidates have differing positions, such as immigration policies and abortion. These exit polls, conducted by Edison Research and funded by a consortium of media agencies (CNN, ABC, CBS, and NBC), provide valuable insights into voter behavior and preferences.
Voter Characteristics
A post-election survey was conducted on a sample of 15,553 individuals using a mix of in-person, telephone, and online interviews. The results reveal that the number of female voters was slightly lower than males (47% vs. 53%). Among the voters, the largest proportion were White (71%), followed by those of Latinx origin (12%) and Black (11%).
In terms of religious affiliation, 42% of voters identified as Protestant, while 24% stated they did not belong to any religion. Catholics made up 22% of the voters, with other religions accounting for 14%. Notably, Jews represented only 2% of the voting population.
Educational background played a significant role in the voter demographics, with 57% of voters not having a college education, compared to 43% who did. When it comes to ideological orientation, the largest group identified as moderates (42%), followed by conservatives (34%) and liberals (23%).
Comparing these characteristics to those of voters in the 2020 election, there is no significant change in most aspects. However, the age composition shows a notable shift. Young people (18 to under 30) made up 14% of voters in this election, while voters aged 65 and over represented 28% - twice the proportion of young people. This age pattern differs from the post-election poll conducted after the 2020 election, which indicated that the proportion of young people was 17% and the proportion of seniors was 22% of all voters. Notably, this shift suggests that the voting bloc in 2024 was older compared to that of 2020.
Voting Trends
The following are the main features of the differences in attitudes towards both the Republican and Democratic candidates according to social groups:
1- Males vs. females:
The results show that 53% of female voters cast their ballots for Kamala Harris, while 45% cast their ballots for Donald Trump. Despite this eight-percentage-point difference favoring Harris among women, it did not ultimately benefit her campaign. This is because Trump's strong support among male voters compensated for the gap, with 55% of male voters choosing Trump compared to only 42% for Harris—a substantial thirteen-percentage-point difference.
2- Racial differences:
Support for Harris among Black voters was overwhelming, with her garnering 85% of the Black vote. In contrast, White voters leaned towards Trump, who received 57% of their votes. However, the impact of Black voters' support for Harris was relatively small, as they represent only 11% of the total votes, while White voters account for 71%. Notably, one of the surprises of the 2024 election was that voters of Latinx origin cast their ballots in a pattern very similar to White voters.
3- Educational differences:
Trump received a higher percentage of votes from less educated voters and a lower percentage from those with more education. Specifically, he secured 63% of the votes from non-college voters, while only obtaining 41% of the votes from college-educated voters.
Shifts in Voting Trends
The 2024 election in the United States witnessed significant shifts in voting patterns compared to the 2020 election, particularly in key demographic groups. These changes have reshaped the political landscape, with notable differences observed in the following areas:
1- Women's votes:
Contrary to predictions, Kamala Harris did not secure the anticipated solid base of female voters. While 57% of female voters supported Biden in 2020, only 53% voted for Harris in 2024. Although Harris achieved a remarkable 91% support from Black female voters, their overall impact on the results was limited due to representing only 7% of the electorate. Among White female voters, Harris secured a mere 45% of votes, while only 60% of Latinas supported her candidacy.
2- Latinx voters:
The Latinx vote saw a dramatic shift from 2020 to 2024. In 2020, the gap between those who voted for Biden and those who supported Trump was 33 percentage points (65% vs. 32%). However, this margin narrowed significantly in 2024, with only a 6-point difference (52% vs. 46%) between Harris and Trump. This near-tie represents a substantial change from the two-thirds support the Democratic candidate received in the previous election.
Further analysis of the Latinx vote by gender reveals even more striking trends. The shift among Latinx males was particularly pronounced, with Trump's support increasing from 36% in 2020 to 55% in 2024. This segment experienced the most significant change, though it remains unclear whether this was due to a preference for Trump or a dislike of Harris. Interestingly, while a majority of Latinas still voted for Harris, their support (60%) was lower than what Biden received in 2020 (69%).
3- The less educated:
The voting patterns of those without a college degree also shifted noticeably. In 2020, this demographic was evenly split between the two candidates. However, the 2024 election saw a clear lean towards Trump, with only 42% voting for Harris compared to 56% for Trump. In contrast, the highly educated segment maintained its voting pattern, with a majority (55%) continuing to support the Democratic candidate.
4- The moderates:
Those who describe themselves as moderates tend to support the Democratic Party. In the 2020 election, Biden secured 64% of their votes, a 30-percentage-point difference from Trump. However, this support waned in the recent election, with the gap narrowing to only 17 percentage points (57% for Harris vs. 40% for Trump).
5- Rural voters:
In contrast to moderates, rural voters generally lean towards the Republican candidate. In the 2020 election, Trump secured 57% of their votes, while Biden received 42%. This 15-percentage-point gap further widened in the 2024 election to 30 percentage points (64% for Trump vs. 34% for Harris).
6- Social groups with clear preferences:
As mentioned earlier, voters are more likely to support Trump if they describe themselves as conservatives (90%), if they are Protestant (63%), or if they are White (57%). Within these categories, there are specific subgroups that were particularly supportive of Trump, with about two-thirds of their votes going to him. For instance, 72% of White Protestants voted for Trump. Additionally, 66% of White non-college graduates supported Trump, a percentage that increased to 69% among White males.
Post-election surveys play a crucial role in identifying the characteristics of the political landscape within societies and tracking changes in voter preferences over time. This valuable information serves multiple purposes: it aids in developing strategies for competing political parties and injects vitality into democratic systems facing ongoing challenges.
