Republican Dr. Mehmet Oz has taken the lead from Democrat Lt. Gov. John Fetterman in Pennsylvania’s U.S. Senate race, while Republican State Sen. Doug Mastriano is right on Democrat Attorney general Josh Shapiro’s heels in the governor’s race, according to a poll.
The Wick Insights poll published Wednesday shows that 49.1 percent of the 1013 likely voters sampled are backing Oz, while 44.6 percent support Fetterman. Another 3.4 percent plan to vote for someone else, and 2.9 percent are undecided. A Fox 29/Insider Advantage poll from earlier this week showed the pair tied after Oz made significant gains in polls in recent months, and the current survey shows he has momentum in the home stretch.
In the governor’s race, Shapiro leads Mastriano by 2.1 percentage points, at 48.5 percent and 46.4 percent, respectively. Just 1.8 percent of respondents plan to vote for another candidate, and 3.4 percent are undecided. The Fox 29/Insider Advantage poll had the republican 7 points back.
Oz and Mastriano are following a similar trend of Republicans across the country who are surging after facing what appeared to be uphill battles in late summer polls, but Wick Insights asserts that a significant change in public opinion has not coincided with the Republican surge. In fact, the pollster reports that those who are unvaccinated or just once vaccinated for the Chinese coronavirus were generally not responding to polls at the rate of fully-vaccinated voters:
In mid-October, we went into the field not knowing that boosted individuals were answering polls at much higher rates than vaxxed-not-boosted or not-vaxxed. When we recognized this imbalance, we made adjustments and then relied on weighting to correct the rest. Unfortunately, we do not know the exact percentage of likely voters that are boosted, so we just set it to 40% in each state. We may move that number up slightly in our next round and customize it in each state, but whatever we set will be at least 15 points lower than what we think is actually being collected right now by probability pollsters. So at the end of the day, compared to the other polls out there I would give this research top marks, but compared to our next round, in which we will design the polling methodology with this knowledge, I would give it a C+. Think of this article more as evidence proving that there is a selection bias that is impacting polling rather than a perfect example of how to treat that bias.
The poll sampled 1013 likely voters from October 8-14 and has a margin of error of plus or minus 3,1 percentage points.
COMMENTS
Please let us know if you're having issues with commenting.