Emerson College Polling on Wednesday released a survey of Tennessee’s 7th Congressional District that appeared to show a very tight race, with the Republican nominee Matt Van Epps only leading State Representative Aftyn Behn (D-Nashville) by 2 percent as the election approaches in eight days.
However, the pollsters did not immediately respond to a press inquiry from The Tennessee Star that sought an explanation for their methodology, and specifically, why their sample size included a disproportionate number of Democrat respondents when compared to the recent vote returns within the district.
According to the poll, 48 percent of respondents said they intend to vote for Van Epps, the U.S. Army veteran and former Tennessee state official who Republicans nominated by a comfortable majority in October.
Emerson’s pollsters showed Behn trailing by just 2 percent, at 46 percent. An additional 2 percent of respondents said they planned to vote for one of the other candidates, while 5 percent were undecided.
However, in the breakdown of their polling questions, the pollsters revealed that only 53.6 percent of respondents said they voted for President Donald Trump during the 2024 presidential elections, compared to 40.9 percent who said they voted for former Vice President Kamala Harris. This means the poll’s sample size included only about 12.7 percent more Republican voters than Democratic.
The number of Democratic and Republican voters included in the sample size differs wildly from the most recent election results in the district, as last year, both Trump and former U.S. Representative Mark Green won the district with commanding victories.
Trump carried the district by 22 percent, and Green similarly received 59.5 percent of the vote, dwarfing the 38 percent of the vote that went to former Nashville Mayor Megan Barry, and 2.4 percent drawn by independent candidate Shaun Green.
The Star asked the pollsters why their sample size did not match the most recent election results, whether methodology was used to account for this apparent discrepancy, and whether the pollsters were concerned their resulting survey could be construed as biased in favor of Democrats, but did not receive an immediate response.
Early voting in Tennessee’s 7th Congressional District concludes on Wednesday. The election will be held on December 2.
Trump endorsed Van Epps ahead of his primary election victory in October, and on Tuesday urged Republicans in the district to surge to the polls during the final hours of early voting.
– – –
Tom Pappert is the lead reporter for The Tennessee Star, and also reports for The Pennsylvania Daily Star and The Arizona Sun Times. Follow Tom on X/Twitter. Email tips to [email protected].
I doesn’t matter if you disproportionately sample some groups more than others AS LONG AS you normalize the results by weighting the results according to the known proportion of those voters in the electorate! That was the very reason they initially started collecting demographic and voter registration data in the first place! For instance, lets say the raw results predict a Behn win 55% to 45%, but Dems and independents are oversampled by 10%. So the pollster divides their tallies by the true proportion in the electorate, and her advantage disappears and the prediction reported out is 45% to 55% GOP wins. The reason the Emerson poll didn’t reflect the final results is very simple: the poll was performed before her gaffe about country music and tourism hit the news cycle. All polling afterward reflected her slide into the final results.
polling doesn’t start out with “we need to show this many from this side, and that many from that side” it’s supposed to do it blindly
but you probably knew that already
Nothing is unbiased at Emerson.