In this conversation, Ruth Agelnik, polling editor at the New York Times, and Elliot Morris, director of Data Analytics, dissect the 2024 election polling landscape. They reveal how polls improved accuracy with only a 2.7 percentage point miss this year, contrasting with previous elections. They tackle challenges like underestimating Trump's support and the evolving methodologies in polling. The duo also discusses innovative techniques such as river sampling and the impact of demographic shifts on voter behavior.
The 2024 election demonstrated a remarkable rebound in polling accuracy, showcasing a national average miss of only 2.7 percentage points.
Emerging polling methodologies like river sampling have proven more effective than traditional methods, indicating a necessary evolution in polling practices.
Deep dives
The State of Election Polling
The 2024 election showcased a significant rebound in polling accuracy, with an average miss of only 2.7 percentage points nationally, marking the best performance for polls in recent history. This improvement comes after notable misses in 2016 and 2020, which had led to a decline in public trust in polling. Despite this progress, challenges remain, particularly regarding the consistent underestimation of support for Trump, which has occurred across three consecutive elections. This raises questions about the methodologies used and the ability of pollsters to reach certain demographics, especially those aligned with Trump's voter base.
Methodological Innovations in Polling
New polling methodologies emerged as the most effective means to gauge public opinion during the 2024 election. Techniques like river sampling, which uses digital ads to recruit respondents, yielded the most accurate results with minimal bias. Traditional methods, particularly live phone polls, struggled to achieve comparable accuracy, underscoring a potential shift away from longstanding polling practices. The discussion around waiting by past vote methods illustrated the complexities of voter psychology, influencing how accurately pollsters could gauge the electorate's sentiments.
Future Challenges for Polling Practices
Despite advancements, there are enduring challenges in the polling landscape, particularly in ensuring representative samples. The low response rates from key demographic groups necessitate a modeling approach to polling, as traditional probability samples become increasingly obsolete. Pollsters must adapt rapidly to changing electoral conditions, particularly as voter behaviors shift and diversify. While optimism remains about understanding public opinion through polling, it is essential to recognize that the methodologies employed must evolve continually to maintain their relevance and accuracy.
Election Day was a moment of truth for pollsters. After high-profile misses in 2016 and 2020, and with a public that has become less trusting of polling, 2024 had the potential to be a make-or-break year for the polls.
Two weeks later, pollsters are, on the whole, breathing sighs of relief. Polls were less error-prone this year than in 2016 and 2020. By one measure, state-level polling was the most accurate it’s been in at least 25 years.
But that’s not the whole story. In this installment of the 538 Politics podcast, we evaluate how the polls did, covering the good, the bad and the statistically insignificant. Hold on to your priors.