Thursday, September 27, 2018

Interview from the Archive: Allan Lichtman warns that polls "are abused and misused as predictors"

Editor's note: This article was originally published on April 9, 2017. 

This is the second of four articles spanning my discussion with Dr. Allan Lichtman. Read the first piece here. 
Story by Joseph Ford Cotto
"Like I've said before .... polls are only good for strippers and cross-country skiers," Former Alaska Governor Sarah Palin told Fox News host Bill O'Reilly two days before last year’s presidential election.
While I have criticized her more than a fair bit during years gone by, it is undeniable that she turned out to be a sage among pundits.
When Palin made her remark, virtually all national opinion surveys – save two highly important yet conspicuously underreported ones – indicated an impending win for Hillary Clinton. Reuters predicted she was set to win 247 electoral votes outright and favored to seize so many more that her chance of victory hovered at 90 percent.
At the Princeton Election Consortium, Dr. Sam Wang – a neuroscientist and prolific author – declared that Clinton enjoyed a 99 percent probability of winning. The platinum-grade forecaster Moody's Analytics also claimed she would triumph in the Electoral College.

United Press International, in conjunction with the polling group CVoter, reported that her Electoral College lead over Trump was substantial: 259 to 209.
After the race was called, Clinton barely eked out 232 votes. Her vaunted 'firewall' of Michigan, Pennsylvania, and Wisconsin crumbled into ash. Only Minnesota and New Hampshire stood -- and not by any means tall. Clinton struggled to win either, despite the former having gone to every Democratic nominee since Richard Nixon's 1972 earth-slide over George McGovern.
Just a few hours earlier, Clinton fans were jubilant over her seemingly assured victory. By the morning after, perhaps more tears were shed than at any other time in twenty-first century America.
What is the moral to this story?
Above and beyond all other factors, something is not impossible simply because the smartest guys and gals in the room – or at least those who perceive themselves as such – claim it is. While the self-appointed experts are accustomed to seeing a handful of their predictions go south, such as who will win a House seat or state legislature majority, they were not prepared to flub on so epic a scale as the U.S. presidency.
Unlike virtually all other professional election watchers, Dr. Allan Lichtman did not forecast a Clinton victory. Instead, he suggested that Trump would win, which caused a considerable stir among the echo chambers within D.C.'s beltway. Using historical rather than polling data, Lichtman turned out to be correct and many of his detractors were left eating their hats, so to speak.
"Lichtman received his PhD from Harvard University in 1973 with a specialty in modern American history and quantitative methods," his biography at American University reads. "He became an Assistant Professor of History at American University in 1973 and a Full Professor in 1980. He was the recipient of the Scholar/Teacher of the year award for 1992-93. He has published seven books and several hundred popular and scholarly articles. He has lectured in the US and internationally and provided commentary for major US and foreign networks and leading newspapers and magazines across the world. He has been an expert witness in more than 75 civil and voting rights cases. His book, White Protestant Nation: The Rise of the American Conservative Movement was a finalist for the National Book Critics Circle Award in nonfiction. His prediction system, the Keys to the White House, has correctly predicted the outcomes of all US presidential elections since 1984."
Lichtman recently spoke with me about many issues relative to political predictions. Some of our conversation is included below.


Joseph Ford Cotto: Many different polls agreed with each other on the presidential election's anticipated outcome, yet were rendered false when all was said and done. Your single prediction, on the other hand, was more solid than the overwhelming majority of opinion surveys. Why did untold sums of money spent on gaging public sentiment pale in comparison to your analysis of historical trends?

Dr. Allan Lichtman: Polls are not predictors. They are abused and misused as predictors, rendering horserace coverage not just meaningless, but deceptive. Polls are snapshots that simulate an election at a single point in time with no guarantee of future results. The truth is that no one understands the relationship between polls and subsequent elections. The polls also screen for so-called “likely voters,” even though no pollster knows who will actually vote or has accurately reported an early vote. The so-called margins of error reported with polls (e. g., plus or minus 3.5 percent) are misleading because they represent only sampling error: the error deriving from the failure to study the entire universe of voters. Such margins of error presume that the poll is otherwise 100 percent accurate, including in its screen for likely voters or its procedures for coping with non-responses. Actual error will always be larger than sampling error and may well be systematic rather than random.

The compilations of polls, but by “experts” like Nate Silver, that produce so-called probability of victories and defeat for candidates produced by “experts” like Nate Silver are equally misleading. These probabilities look precise and scientific, with probabilities of a Clinton ranging from more than 70 percent to more than 99 percent likely. However, these forecasts fall victim to the fallacy of “false precision.” Nate Silver and other such purveyors of win-loss probabilities do not conduct independent scientific analysis. Nate Silver is a clerk. He compiles poll data and his seemingly precise probabilities are no more accurate than the underlying polls. Silver and other compilers cannot know whether the polls are right or wrong, of if wrong, in which direction.

Cotto: Considering how historical data once again established itself as more reliable than opinion polls, is it likely that, during 2020, polls will find less credibility among the media and general public?

Lichtman: The pollsters have recovered from serious errors before, for example, miscalling the Harry Truman’s victory in 1948 and missing the extent of Ronald Reagan’s victory in  1980. There is a symbiotic relationship between the pollsters, the media, and the political consultants. Everybody makes money from the focus on the polls which drives coverage and campaign strategy. I hope, however, that the errors of this cycle and the success of the Keys will lead to more serious, substantive coverage of the next election.

The pundits twisted themselves into pretzels to justify why what they assured us would happen – a Clinton victory -- had not happened. It was much too easy to blame Hillary Clinton or credit the Donald Trump campaign. .Just 24 hours earlier, these same pundits and Democratic operatives had informed the world that Hillary Clinton was poised to complete a historic victory as the first woman to be elected president of the United States. The Clinton and Trump campaigns did not suddenly change overnight. It was the same candidates and the same campaigns as the day before.

No comments:

Post a Comment