Kat Bourgault
Anchor Contributor
Leading up to the 2024 Presidential election, we saw a lot of hopeful polls for both sides of the ballot. Both Trump and Harris personally expressed that they were hopeful and confident in their own victories. Pollsters had differing opinions, but the overall consensus is that the race was “too close to call.”
I spoke with Dr. Natalie Rogol from RIC's Political Science Department, who specializes in methodology and polling. Given her expertise, I was curious to hear her take on why these polls ultimately fell short in predicting the actual outcome.
I started by asking Dr. Rogol if she would consider the election polls a failure, especially given their inability to predict the record-breaking turnout for Trump. However, Rogol clarified that she does not view the polls as failures. Instead, she pointed out that almost every reputable poll openly acknowledged that the race was too close to call from the start.
Many people were disappointed with the polling because we expected more definitive answers than the data could realistically provide. I brought up the infamous polling failures of 2016, where nearly every major poll predicted a Clinton landslide, leaving much of the country skeptical of election polling ever since. Dr. Rogol explained that the polling industry has learned a lot since 2016, making changes such as weighting responses more accurately, and recognizing the connection between Trump supporters and low trust in the institutions who are conducting the polls.
Another challenge faced by pollsters is ensuring thoroughness. If a pollster identifies themselves as part of an institution that a respondent distrusts, it might elicit a negative response, especially among Trump supporters. However, because pollsters must remain objective, they can not make assumptions about the respondent’s intentions, which complicates interpreting the data.
Dr. Rogol also expressed her concerns with poll herding, a phenomenon where polling organizations adjust their results to align more closely with the averages of other polls, potentially compromising the accuracy of their findings. No pollster wants to be the outlier predicting a drastically different outcome, as the risk of being wrong could significantly damage their credibility for future elections.
To wrap up, I asked Dr. Rogol if she had any advice or tips for interpreting polls as prediction tools. She emphasized the importance of considering the source, which prompted me to mention the infamous Ann Seltzer poll that predicted Harris leading in Iowa, a traditionally red state. Rogol pointed out that Seltzer, a reporter, gained significant attention and name recognition from this prediction, highlighting why it’s crucial to question not only who is providing the information, but also their motivations behind it.
While polls remain a valuable tool for understanding voter patterns, their flaws are undeniable. As Dr. Rogol emphasized, the source and methodology behind each poll are crucial to making sense of the results. The 2016 election and subsequent polling adjustments show that while progress has been made, issues like poll herding and distrust in institutions continue to complicate polling predictions. Moving forward, we could benefit from a more thoughtful approach to interpreting polls, paying close attention to the context and credibility of the pollsters.
Comments