How did the polls perform in the 2022 midterm elections? Were they accurate in predicting winners in the major races, or did they miss the mark? How important was the issue you’re assessing for the QU Poll Trends Project in the 2022 midterms and was it reflected in the polls? Finally, what are some of the challenges/issues with polling that were highlighted by this latest election cycle? To support your response, you need to use one of the readings posted on Blackboard, one article you find on your own, and a third article that is either one of the Blackboard readings or from an outside source.
Overall, the polls regarding the Senate race in Pennsylvania turned out to be a mixed bag. A brief article titled, “How Accurate Were The Polls In The 2022 Election?”, by Ed Kilgore compared the polls to the actual outcome of the election. In Pennsylvania’s case, the nail-biting margin, reporting that Mehmet Oz was up 0.4% in the polls did not really reflect in the actual results where John Fetterman had a significant 3.4% lead with most of the votes counted. We can see here that the polls completely missed the mark in gauging how close the election was.
ReplyDeleteThe issue I am covering in the QU Poll Trends Project is the attack on the Capitol, aka January 6th. I don’t think my topic had a huge impact on the Pennsylvania midterms. In the article, “How the Jan. 6 hearings changed public opinion ahead of the midterms” Alexander Hutzler, states that polls did not change much before and after the hearings. Polling conducted by Monmouth University reported that only 8% of participants changed their minds after the hearing which had increased from 6% before the hearings. Public opinion on whether the Jan 6th attack was a riot, insurrection, or legitimate protest also changed very little after the hearings. 65% percent through Jan 6th was a riot which changed to 64% after the hearings. 50% thought it was an insurrection changed to 52%. 34% thought it was a legitimate protest changed to 35%. As we can see, these changes were minuscule. Jan 6th also wasn’t a topic that Fetterman nor Oz focused on. Fetterman focused more on democracy and election integrity which is somewhat related but did not constantly harp on Jan 6th.
It might have had more of an impact on Oz. Oz was stuck in a rock in a hard place as he was being endorsed by Trump. It makes sense that Jan 6th would be a topic he would avoid, as he would be at risk of losing his biggest endorser. In hindsight, his avoidance of the Jan 6th riots and election fraud could have resulted in people choosing not to vote for him as he was not fully aligning himself with the MAGA Republicans and Trump. However, there is no evidence to support this.
I think the most obvious and basic challenge with polling this election cycle was taking poll results at face value. What I mean by this is that issues that the media took the info and ran with it. This is where we see narratives like the “Red Wave” in which Republicans would dominate the election. However it was more of a splash and not the blowout most Republicans predicted. Assistant Research Professor and Director, Ashely Koning, of Rutgers’ Eagleton Center for Public Interest Polling, when into detail on why the “Red Wave” did not come to fruition. As most of us know by now, polls are not supposed to be used to make predictions. Koning states that polls should be used to “asses” public opinion. She stresses the polls should stop being used by the media to develop agendas for their parties and that polls should be used as scientific data within certain margins of error (as is the nature of polling).
In regards to the Georgia senate race, I think that the polls were pretty accurate in the 2022 midterm elections. Over the past few months, the majority of the polls that I’ve seen have shown Raphael Warnock and Herschel Walker basically being tied, with one candidate leading by a very small margin most of the time. For example, in the Quinnipiac University poll from October 12, 2022, Warnock led Walker 52 to 45 percent. In another poll done by the same university released on September 14, Warnock led Walker 52 to 46 percent. This prediction of a tight race rang true, as the Georgia senate race will be headed to a run-off in a few weeks. In Georgia, a candidate must win by 50% plus one, and neither candidate could reach that mark on Election Day.
ReplyDeleteAlthough I think the polls were accurate in predicting the results of the race that I was following, not everyone believes they are always completely accurate. In an article by Julia Manchester for The Hill titled “Did the polls get it wrong again?,” the concept of the “red wave,” or an overwhelming amount of support for the republican (red) party, is discussed. The article includes a quote from the director of Public Interest Polling at Rutgers University, Ashley Koning, who said that “a lot of polls from those credible, reputable, traditional pollsters were met with skepticism and/or disbelief when they were showing better numbers for Democrats.” This makes me think that the “red wave” narrative, although potentially started by the polls, was hyped up too much by media coverage and fell flat. Democratic pollster Celinda Lake proposed the idea that this narrative was a set-up by the republicans to try to make the party and its members more optimistic and even try to drive donations and voter turnout. This is also a theory that I believe more than the theory that the polls were just wrong. Polls aren’t supposed to tell us exactly what will happen in the elections, they’re just supposed to give us an idea of how people are planning on voting and why, and I think that they did a good job of that during this election cycle.
I think the biggest challenge facing polling is that young people, including myself and my classmates, are hesitant to pick up the phone, especially when we don’t recognize the number. Another big challenge is that opinions can change quickly throughout election cycles, especially due to “October surprises” like Herschel Walker’s multiple abortion scandals. The only potential solutions I can think of for these are an online polling method and/or to poll more frequently. However, I don’t know if these would be effective - online polls can be ignored as easily as phone calls, and it takes time for results to be calculated and released, so polling more often might not be a realistic option.
Jennifer Moglia cont.
DeleteThe issue that I am assessing for the QU Poll Trends Project is abortion. I predicted that this issue would be important in this year’s election, and I think that I was right. The same Quinnipiac poll mentioned earlier had abortion as the second-most important issue to Georgia in October, trailing only inflation. Abortion was also the most important issue among democrats and the third-most important issue among independents. There was a whole separate section of the poll results just on this issue, sharing that “More than 8 in 10 likely voters (84 percent) say when thinking about this year’s midterm elections, it’s either very important (59 percent) or somewhat important (25 percent) that a political candidate shares their view on abortion, while 15 percent say it’s either not so important (9 percent) or not important at all (6 percent).” The Politico article titled “A Predicted ‘Red Wave’ Crashed Into Wall of Abortion Rights Support on Tuesday” by Alice Miranda Ollstein and Megan Messerly explores this in a more in-depth way. The article begins by stating that abortion rights being threatened this past summer caused a surge in voter turnout for the midterms. Young voters of color in particular showed up in record numbers, despite predictions that the issue wasn’t as important to them or that Democrats would not see the victories they did in governor, senate, attorney general, and state legislature races. All of that evidence certainly suggests that abortion was a hot topic during this election cycle. I think that Democrats will be trying to build off of this momentum and Republicans will be trying to find what went wrong for years to come.
Works Cited
Malloy, Tim. “GEORGIA SENATE RACE UNCHANGED: WARNOCK LEADS WALKER 52% – 45%, QUINNIPIAC UNIVERSITY GEORGIA POLL FINDS; GOV RACE BETWEEN KEMP AND ABRAMS REMAINS TOO CLOSE TO CALL.” Quinnipiac University Poll. 12 October, 2022. Accessed November 13, 2022.
Manchester, Julia. “Did the polls get it wrong again?” The Hill. 12 November, 2022. https://thehill.com/homenews/campaign/3731665-did-the-polls-get-it-wrong-again/. Accessed November 13, 2022.
Ollstein, Alice Miranda and Messerly, Megan. “A Predicted ‘Red Wave’ Crashed Into Wall of Abortion Rights Support on Tuesday.” Politico. 9 November, 2022. https://www.politico.com/news/2022/11/09/abortion-votes-2022-election-results-00065983. Accessed November 13, 2022.
Piascik Works Cited
ReplyDelete"Georgia U.S. Senate Election Results." The New York Times, 14 Nov. 2022, www.nytimes.com/
interactive/2022/11/08/us/elections/results-georgia-us-senate.html. Accessed 14 Nov. 2022.
Manchester, Julia. "Did the Polls Get It Wrong Again?" The Hill, Nexstar Media, 12 Nov. 2022,
thehill.com/homenews/campaign/3731665-did-the-polls-get-it-wrong-again/. Accessed 14 Nov. 2022.
Malloy, Tim, and Doug Shwartz. "GEORGIA SENATE RACE UNCHANGED: WARNOCK LEADS
WALKER 52% – 45%, QUINNIPIAC UNIVERSITY GEORGIA POLL FINDS; GOV RACE BETWEEN KEMP AND ABRAMS REMAINS TOO CLOSE TO CALL." Quinnipiac University Poll, 12 Oct. 2022, poll.qu.edu/images/polling/ga/ga10122022_glmf25.pdf. Accessed 14 Nov. 2022.
Charlie Windels Part 1
ReplyDeleteAfter the 2022 midterm elections immediately I thought about the polls and some of the backlash they would get for “not being accurate.” After talking about this issue in class and reading about it further it was apparent to me how this can be a very controversial topic. I read a few articles about the midterm election results and whether or not people thought they were accurate and I was getting very mixed opinions. One article I read quoted “Pollsters argue that outliers happen and caution that polls are meant to be a snapshot in time, not a prediction” (Manchester). This was something we talked about briefly in class and is something people need to understand. While this year's midterm election results were mostly accurate to the polls there still were some outliers. This is something that shouldn't be of surprise to us as in the past we have had similar outcomes with polls and results. We need to remember that as much as we want to think polls are a prediction for the future, they are not. As different issues arise, polls change even in the days leading up to the election.
Another thing I think that is important to mention in addition to this “snapshot in time” idea is that not all polls are reaching everyone, and not everyone chooses to participate in polls. An article I was reading titled How Accurate Were the Polls in the 2022 Elections talked about how while polls do their best they face many challenges and need to make many adjustments to reach certain demographics. The article explains “This led to a big debate over problems with getting voters (especially Trump voters) to respond to traditional phone-interview polling and the divergence of different pollsters using different methodologies” (Kilgore). I think this is something people often don’t think about when they are looking at the polls. While the polls this year did not reflect the results of the election perfectly, I think for the most part the ones I was looking at, especially in some of these very tight races such as Georgia and Pennsylvania, were pretty close to accurate.
Despite a flurry of media coverage concerned over rumors of a “red wave,” “red tsunami,” or otherwise Republican surge, no such thing materialized. Much of this coverage was driven by polling of key federal Senate and House races, and some gubernatorial contests, such as the Pennsylvania Senate race between Dr. Mehmet Oz and Lt. Governor John Fetterman. As with every election cycle, some polls were right on the money...and others were bizarrely wrong. An Axios analysis of 250 statewide surveys, stored by RealClearPolitics, found that some polls were strikingly accurate while others overcounted or undercounted certain races; polls in Georgia were “on average” off by less than a point, well within any margin of error, while Florida’s Gov. Ron DeSantis performed around seven or eight points more than expected (Markay). Some pollsters, and thus their results are going to be incredibly biased, which results in skewed predictions. Republican firms like Insider Advantage, Rasmussen Reports and Trafalgar Group are known to create more Republican-friendly polls with a fast turn-around, while more mainstream groups like Quinnipiac University or ABC/Washington Post have conducted fewer surveys than in recent years (and thus, generally, less biased polling) (Cohn). A Slate explainer of the Trafalgar Group’s polling illustrates much of the concern with political polling: the group’s latest poll showed Oz crushing Fetterman with 47%-45.5%, while on Election Day Fetterman won with 51%-46.6% (Mathis-Lilley). Plus, Trafalgar was off in Pennsylvania’s gubernatorial election, polling Democrat Josh Shapiro at 49.7% and Republican Doug Mastriano at 45.4%; Shapiro would win the election with 14 points more than Mastriano (56.3%-41.9%) (Mathis-Lilley).
ReplyDeleteFor the Poll Trends Project, my topic is gun violence. Of course, the most significant event of this nature is the Uvalde school shooting which claimed the lives of 19 students and two teachers after a man invaded the school (Prokupecz, Friedman and Clarke). As a result, while inflation and the economy stole the top slot as the most important issue to voters in July, gun violence was ranked second overall and first among Democrats (Callahan). A June Quinnipiac poll saw 12% of voters rank gun violence as the most important issue (inflation was ranked as most important by 34% of voters), with 22% of Democrats citing gun violence as the most urgent issue; a July Gallup poll also found that gun violence weighed on voters’ minds, with 80% of U.S. adults saying the issue is extremely or very important (Callahan).
A June Quinnipiac poll saw 12% of voters rank gun violence as the most important issue (inflation was ranked as most important by 34% of voters), with 22% of Democrats citing gun violence as the most urgent issue; a July Gallup poll also found that gun violence weighed on voters’ minds, with 80% of U.S. adults saying the issue is extremely or very important (Callahan). However, exit polls show that as time went on gun policy declined in importance to some voters. NBC News exit polling of all voters asked three questions about guns in America, with the most prominent asking voters to pick one of five topics as the issue that mattered most to their vote (crime, abortion, inflation, gun policy and immigration). Just 11% of voters cited gun policy, leaving it tied for 3rd place with crime; a further breakdown showed that 60% of those who answered gun policy are Democrats and 37% are Republicans (“Live Exit Polls 2022”). A question about stricter gun control measures saw 56% of respondents support stricter gun control (76% Democrats and 22% Republicans) and 40% oppose more restrictions (88% Republican and 10% Democrats) (“Live Exit Polls 2022”).
DeleteBottom line, when it comes to the Georgia election, I believe that the polls performed fairly well in the 2022 midterm elections. On Nov. 8, the FiveThirtyEight poll aggregation had Herschel Walker leading Raphael Warnock by a point. However, as the actual votes were counted, Warnock was ahead by 0.9 points. This however did not matter, as both the poll aggregations and the actual results had the election result in a runoff election. In other races, polls were sometimes accurate and sometimes off. In Pennsylvania, John Fetterman trailed Mehmet Oz by 0.4 points near election day, but won by 4.4 points in the end (Kilgore). And in New Hampshire, Maggie Hassan led in the polls by 1.4 points on average, but once the votes were counted, she won by 9.2 points (Kilgore). An interesting aspect of these poll results is how Republican-friendly pollsters have been putting out more results in the 2022 midterms, which may be able to partially explain the “red wave” that the media was discussing during the lead up to election day (Cohn). If the polls ever missed the mark, it never seemed to be by much. They seemed to commonly predict elections to be toss-ups, but in the end, one candidate would win by a few points, which can at times fall in the range of the margin of error.
ReplyDeleteAmong my project group for the Election Eve project, we were able to correctly predict the outcomes of the Florida and Pennsylvania senate elections. However, we are now perplexed by the fact that we have to wait for the results of the Georgia senate runoff election. The polls had the race extremely close to begin with, so now we eagerly await what will happen now that it was forced to go to a runoff. Warnock had the lead once the initial race was called, but anything can happen now as all that matters is what voters decide to return to their polling places.
For the QU Poll Trends project, I am going to be analyzing gun violence, and in the 2022 midterms, this was not an issue at the forefront of voters’ minds. This election cycle, over 50% of voters were focused on the state of inflation in the United States as well as the complete legalization of abortions, with gun violence being thrown to the side (Edwards-Levy & Luhby). I am sure that this is likely to change following the shootings that occurred at both the University of Virginia and the University of Idaho yesterday. However, it begs the question as to why it was not focused on as much by voters for this cycle specifically. I figure it was primarily due to the skyrocketed prices of nearly all goods, as well as the overturning of Roe v. Wade, but with America being in the state that it’s in at the moment, it should probably have been more important to voters.
Overall, in the 2022 midterm elections, the polls performed well. According to Axios, late-cycle midterm polls followed the actual election results closely. For example, in the Georgia Senate and Gubernatorial elections, “the polls examined by Axios nailed both the equivalent races: on average, they were off by less than a point,” (Markay). To be that close to reality is extraordinary. The Website 538, a poll aggregator, had Georgia’s Senate election between Rafael Warnock and Herschel Walker with both candidates under, which is exactly what transpired on election day. 538’s Nate Silver declared this election one of the best for pollsters ever. The differences in polls and reality were mostly contained within the margin of error. Being off within that margin is in no way a slight on the polls. Nate Cohn of the New York Times tweeted that the NYT/Siena poll, “had its best year yet: an average error of 1.7 pts with 0 bias,” (Cohn). The 1.7% average error is well within most polls margin of error, making those polls accurate.
ReplyDeleteThe issue of immigration was not a major issue this midterm. CNN’s exit poll found only 10% of voters listed that as their most important issue. Other issues like Inflation (31%) and Abortion (27%) were more important to voters and were the issues that decided this election. Immigration was firmly in the background of the election. In Pennsylvania, Oz and Fetterman rarely touched on the issue, with inflation, abortion, and crime being the most talked about issues. Of the 10% of voters whose top issue was immigration, they heavily broke for Republicans with the GOP winning nearly 80%.
I believe that the biggest challenge regarding polling highlighted by this election cycle was the way we analyze polls. Often times media consumers confuse polls for the media narrative and what pundits are saying as a result of polls. While the media narrative was that a red wave was incoming, the polls were not showing that to be true. The Trafalgar Group, a Republican pollster, was the only organization predicting a Republican surge of that magnitude. Trafalgar was also one of the pollsters that, according to an analysis by USA Polling, fared the worst this election. Pundits running with the red wave narrative led many people to believe that is what the polls were predicting; however, that is not what was playing out in the actual numbers. The races that the polls predicted would be tight were. This is a failure of media consumers and the media itself. In a perfect world, media consumers would be better at decerning when pundits are engaging in gassbagary and when the analysis is based in fact. However, since we do not live in that perfect world this distinction should be made for media consumers so polls don’t continue to be maligned when reality plays out exactly how they expected it to.
I honestly think the polls performed horrendously again this year, but this could also be media swaying my opinion. Coming into this all, I heard that a “Red Wave” was coming through America, the Senate, and into the House of Representatives and even gubernatorial races. Now that many races have been called, we can see that this was barely even a red droplet. Amy Eskind when talking about how the polls were off, even said, “The "red wave" that pollsters were predicting before the midterms turned out to be more of a red trickle. The expectation that Democrats could lose as many as 35 seats in the House of Representatives has been disproven. If Republicans clinch majorities in the House or Senate, it will be by a razor-thin margin.” (Eskind). The Democrates won many important gubernatorial races and EVEN kept control of the Senate. At the time of writing this blog, the Republicans, while likely to take the house, still do not have the majority needed. (Well, at least the races haven't been called in their favor yet). (New York Times). Looking at how the media could have affected this, Fox News and many right-leaning sites pushed this red-wave narrative. Pushing this narrative made it feel all the more real, even if polling wasn't showing a Red Wave, but I believe it did. I think a problem in the polls is one described by Ben Collins in a tweet saying that Gen Z would rather pick up an unpinned grenade than pick up a phone call from a number they don’t know. Furthermore, my QU Poll Trend Analysis issue did play big in the midterms, or at least the media would like you to believe that. Now since my issue is Gun Violence, it was a little hard to track in other polls, so if you put gun violence in with crimes, from my personal viewing experience Fox News was pushing hard the narrative that crime was rampant in new york and all across the country and that republicans were hard on crime. In reality, in a Marist poll, crime didn't even make the top five in October of 2022 but did in November. The November number was low, coming in at 7% behind Health Care, Immigration, Abortion, Preserving Democracy, and Inflation. (Marist Poll).
ReplyDeleteI think, and this is just a personal thought, reaching the younger generations of voters for polling is going to be extremely hard. I remember seeing on Twitter that younger voters, and Gen Z, voted overwhelmingly for Democrats. This is exactly the problem that Ben Collins was speaking about. If you can't reach this demographic to poll them, you are missing a huge part of each party's youth reach. If the youth continue to be this influential, without the potential of being polled, this will be a problem that will plague voters for elections to come.
Referneces:
DeleteEskind, A. (2022, November 10). Why did polls prepare us for a red wave? experts weigh in on the surprising midterm election results. Peoplemag. Retrieved November 15, 2022, from https://people.com/politics/why-midterm-polls-falsely-predicted-red-wave/
The New York Times. (2022, November 8). Live election results: Top races to watch. The New York Times. Retrieved November 15, 2022, from https://www.nytimes.com/interactive/2022/11/08/us/elections/results-key-races.html?name=styln-2022-midterms®ion=TOP_BANNER&block=storyline_menu_recirc&action=click&pgtype=LegacyCollection&variant=show&is_new=false
Polls, analysis, learning, and more. Home of the Marist Poll. (n.d.). Retrieved November 15, 2022, from https://maristpoll.marist.edu/
The 2022 midterm election polls were more accurate than anyone thought, with Nate Silver of FiveThirtyEight tweeting it was “one of the most accurate years for polling ever” (Markay). The “red wave” projected by pundits and political scientists never came to fruition and the GOP walked away frustrated and feeling defeated. It’s rare that the president’s party holds its own during the midterm cycle. After Catherine Cortez Masto won her reelection bid in Nevada, it was confirmed November 13 that Democrats had maintained their hold in the Senate. The results in the House have not yet been finalized, but it looks like Republicans will win. There’s still a runoff to look forward to in Georgia between Raphael Warnock (D) and Herschel Walker (R) which could secure a 51-49 majority if Warnock wins.
ReplyDeletePolling has seen a continued downward trend in success, facing hurdles from landline graveyards, increased expenses, and difficulty reaching key blocs like young and conservative voters (who have been told not to trust pollsters as a party politic). However, with this cycle’s polling closely matching many of the election outcomes, criticism may rest, at least temporarily (Markay). Climate change was not a motivating factor in this year’s midterms. It’s categorized as a much more pressing issue in reality than politically (Miller). Mindy Romero, a research assistant professor at the USC price School says that “While it is certainly a critical topic for many Americans and our nation, it often does not register in surveys as a top issue for older, higher turnout voting groups that are courted by candidates and campaigns” (Miller). Additionally, climate change lost focus from its primary support group of young people, who were motivated this year by the overturn of Roe v Wade. Unless some major environmental event happens in the near future that directly threatens voters, climate change will likely remain an unimportant issue in election polls.
A challenge faced by every poll this cycle was accurately predicting turnout by age. Generation Z is widely credited as protecting the Democratic hold in Congress and preventing a red nationwide flip. They are a difficult bloc to poll, as Generation Z primarily uses cell phones and the internet as their communication platforms. Additionally, they have less public trust in random callers than previous generations. These factors combined make it difficult for pollsters to reach them and thus accurately predict if and how they will vote. Many political operatives used “likely voters” as their polling population, assuming America’s youngest eligible voting bloc wouldn’t turnout this year (Aylward). It’s been determined that abortion had a much larger influence this year than predicted as well. Young voters are strongly pro-choice and mobilize when reproductive rights are threatened (Aylward). A national poll by URGE in May found that 65% of Generation Z voters reported access to abortion care was why they went to the polls (Aylward). Overall, midterm election polls were fairly accurate, but there’s still work to be done to reach key voter blocs in our evolving society.
Works Cited Page
DeleteAylward, Ashley. “How Gen Z Held off the Red Wave.” Time, Time, 10 Nov. 2022, https://time.com/6232272/generation-z-voters-midterm-elections/.
Markay, Lachlan. “Analysis: How Most Late-Cycle Polls Actually Performed.” Axios, 11 Nov. 2022, https://www.axios.com/2022/11/11/polls-performance-midterms-results.
Miller, Jenesee. “Why Climate Change Has Been Missing in Action from the 2022 Midterm Election.” USC Sol Price School of Public Policy, 4 Nov. 2022, https://priceschool.usc.edu/news/why-climate-change-has-been-missing-in-action-from-the-2022-midterm-election/.
I thought overall, the polls for the midterm elections were relatively accurate, or at least within the margin of error. Over the past few years, polls have been used as scapegoats for candidates and supporters to explain why they lost a race. Though polls are scientifically supposed to be an accurate, representative sample of the electorate, they truly are only a small sample of the larger population. I recall hearing from the representative of the QU poll that they take samples of only about 2,000 people. While the QU poll is one of the most highly regarded polls in the country, it is hard to make definitive predictions based on such a small sample size. Often times, people forget that polls don’t vote, and they still have to get out and submit a ballot of their own.
ReplyDeleteThis time around, the polls were relatively accurate. According to an analysis by Axios of 250 statewide polls, they found that the average number of points a poll was off was 3 percent, which is usually within the margin of error (Axios). "Averaged polls in some races were more accurate than others. Republicans performed between seven and eight points better in Florida's gubernatorial and Senate races than the surveys predicted” (Axios). In Georgia, on the other hand, the race ended up going to a runoff, which is very much so in the margin of error, as most polls were only off by less than one percent (Axios). Nate Silver, founder of FiveThirtyEight said that this was one of the most accurate years for polling ever (Twitter).
In terms of January 6th, which is the issue I am examining for the QU poll trends analysis project, that issue in particular was not very pressing to voters, but state of Democracy in the US was asked in many exit polls. According to a CNN exit poll, 68% of respondents said Democracy is threatened in some way, shape, or form (CNN). Thought this is not directly a question about January 6th, I don’t think this question can be answered or asked without taking the insurrection into account.
The biggest issues with polling this year, was probably getting an accurate sample size. Making sure there are respondents from every voting generation. This is especially important now as Gen Z is voting in higher numbers than ever before, but also picking up phone calls from random numbers less than ever before. This is a relatively small issue, especially looking at the midterms, but it is something pollsters will need to take into account as more and more young people are voting.
DeleteOverall, the polls for the midterms were relatively accurate – about as accurate as a poll can be with such close races across the country. It will be interesting to watch the next couple of election cycles and see whether or not polls are able to remain as accurate as they were this time around.
“Exit Polls for Midterm Election Results 2022 | CNN Politics.” CNN, https://www.cnn.com/election/2022/exit-polls. Accessed 15 Nov. 2022.
Markay, Lachlan. “Analysis: How Most Late-Cycle Polls Actually Performed.” Axios, 11 Nov. 2022, https://www.axios.com/2022/11/11/polls-performance-midterms-results.
Nate Silver [@NateSilver538]. “This Was One of the Most Accurate Years for Polling Ever. You Have Absolutely No Clue What You’re Talking about. Https://T.Co/H41uviGS1n.” Twitter, 9 Nov. 2022, https://twitter.com/NateSilver538/status/1590461764516610048.
Despite claims of a “red wave” and distrust towards the polling industry in the aftermath of the 2016 election, the polls in the 2022 midterm elections actually didn’t perform that poorly. Markay (2022) writes that “On the whole, pollsters' predictions aligned better with election results than punditry that foretold a Republican wave” (para. 6.) When it comes to my own election eve candidate, Val Demings, Demings was not set to win the election. In the days leading up to the election, Marco Rubio was ahead by 11%, according to a University of North Florida Poll from October 26 (2022.) The reality was pretty spot on; Rubio won by around 16% (Florida U.S. Senate Election Results, 2022.)
ReplyDeleteHowever, this race didn’t have as close of a margin as many others because we pretty much knew Rubio was going to win in the final days. Other races, like in the case of Herschel Walker-Raphael Warnock, proved harder to call by the pollsters. The polling averages had Walker ahead by 1.4%, but neither candidate ended up securing a majority of the vote, so they will have to go to a runoff (Kilgore, 2022, para. 5-6.)
My issue for the polling project is immigration. In the Demings-Rubio race, immigration wasn’t a huge talking point for either candidate. In the November 2 Quinnipiac Poll (2022), around 6% of Americans said that immigration was the biggest issue that faces the nation today. Though this number is a lot smaller, according to the QU poll (2022), than issues like inflation (36%) and abortion (10%), immigration was still a significant issue in some races.
Returning to the Walker-Warnock race, by aligning himself with Donald Trump, Walker took an extreme stance on immigration and the Mexican border. His website states, “the border crisis isn’t just a border state problem, it is a Georgia problem that is bringing drugs, crime, and human trafficking to our state” (Secure Our Border to Protect Americans, 2022.) In comparison, Walker’s website focuses on fixing immigration laws and providing a path to legal citizenship and states that “the United States has always been a nation of immigrants” (Immigration, 2022.) Since this race is still so close and is approaching a runoff, it’s possible that this contentious issue will play a role in the election.
Personally, I think that some of the challenges polling faced during this election was mistrust of polls. As a journalist, I still believe that polls have a role in the media and in accurately depicting what Americans think and cutting through the narrative of political pundits. Going forward, I think that polling organizations need to increase their reach by using text and email polls, because I know that my generation is not keen to pick up the phone when a pollster calls.
In examining the accuracy of polls, I think it would be worthwhile to split the sample into two camps: pre- and post-November 1st. Though certainly not a poll, the FiveThirtyEight’s forecast had Republicans winning in 59 of 100 models ran (Silver). Other projectors, such as Cook, predicted a Republican gain between 15 and 30 seats in the House (Kilgore).
ReplyDeleteThe most noise was caused around Republican losses in the Senate, as none of the new Republican candidates won aside from J.D. Vance in Ohio and Ted Budd in North Carolina. In a short rundown, Oz lost in Pennsylvania, Masters lost in Arizona, Laxalt lost in Nevada and Bullock lost in New Hampshire. Georgia is headed to a runoff, though Warnock (D) is the odds-on favorite to win.
The only election night victory the Republicans scored was in Florida, where Rubio gained his third term as Senator and DeSantis won his second term as Governor by double-digit margins. So wide, in fact, that they even flipped Miami-Dade county red, a feat not done in decades. So wide, in fact, that they even flipped Miami-Dade county red, a feat not done in decades. Rubio’s 16-point and DeSantis’ 19-point win vastly outperformed polls, which predicted a roughly 9-point and 12-point victory respectively. In Miami-Dade alone, there was a roughly 30-point flip from DeSantis’ previous election, where he “lost the state's most populous, majority-Hispanic county by roughly 20 points” and turned the ship around by winning it by 11 points this year (Vassolo).
As always, the smart move is to follow the money. Across the board, Democrat candidates out-raised Republicans by two-to-one, three-to-one, even six-to-one margins (Breuninger and Cortés). Save for Fetterman, the majority of these funds for Democrat campaigns came from large individual contributions (ibid.). Democrats even commanded more money from PACs, often eschewed by the party as an illegitimate way that Republicans sneak dark money into campaigns (ibid.).
The lesson to be learned is that polls are yet to adjust to the way elections have been remade. The mail-in ballot phenomenon was thought to be a black swan event contained to the 2020 election cycle. But even as covid subsided, the “temporary” voting health safety measures remained. As a result, there is no longer “Election Day” but instead election season.
Sorry, forgot to put my name: Will Dean
DeleteCitations:
Breuninger, Kevin, and Gabriel Cortés. “Here Are the Top Senate Races to Watch in the 2022 Midterms.” CNBC, 9 Nov. 2022, https://www.cnbc.com/guide/midterms-2022-top-senate-races-oz-walker-vance-kelly/.
Kilgore, Ed. “How Accurate Were the Polls in the 2022 Elections?” Intelligencer, 9 Nov. 2022, https://nymag.com/intelligencer/article/election-results-midterms-polls-right.html.
Vassolo, Deirdra Funcheon, Martin. “Miami-Dade Flips Red: Midterm Lessons for Republicans and Democrats.” Axios, 14 Nov. 2022, https://www.axios.com/local/miami/2022/11/14/miami-dade-midterm-election-republicans.