What do you think?
Rate this book
Lost in a Gallup: Polling Failure in U.S. Presidential Elections
This update of a lively, first-of-its-kind study of polling misfires and fiascoes in U.S. presidential campaigns takes up pollsters’ failure over the decades to offer accurate assessments of the most important of American elections.
Lost in a Gallup tells the story of polling flops and failures in presidential elections since 1936. Polls do go bad, as outcomes in 2020, 2016, 2012, 2004, and 2000 all remind us. This updated edition includes a new chapter and conclusion that address the 2020 polling surprise and considers whether polls will get it right in 2024.
As author W. Joseph Campbell discusses, polling misfires in presidential elections are not all alike. Pollsters have anticipated tight elections when landslides have occurred. They have pointed to the wrong winner in closer elections. Misleading state polls have thrown off expected national outcomes. Polling failure also can lead to media error. Journalists covering presidential races invariably take their lead from polls. When polls go bad, media narratives can be off-target as well. Lost in a Gallup encourages readers to treat election polls with healthy skepticism, recognizing that they could be wrong.
Lost in a Gallup tells the story of polling flops and failures in presidential elections since 1936. Polls do go bad, as outcomes in 2020, 2016, 2012, 2004, and 2000 all remind us. This updated edition includes a new chapter and conclusion that address the 2020 polling surprise and considers whether polls will get it right in 2024.
As author W. Joseph Campbell discusses, polling misfires in presidential elections are not all alike. Pollsters have anticipated tight elections when landslides have occurred. They have pointed to the wrong winner in closer elections. Misleading state polls have thrown off expected national outcomes. Polling failure also can lead to media error. Journalists covering presidential races invariably take their lead from polls. When polls go bad, media narratives can be off-target as well. Lost in a Gallup encourages readers to treat election polls with healthy skepticism, recognizing that they could be wrong.
368 pages, Paperback
Published February 20, 2024
12 people are currently reading
113 people want to read
About the author
W. Joseph Campbell
13 books5 followersW. Joseph Campbell is an American writer, historian, media critic, and blogger who is the author of six other books, including the award-winning Getting It Wrong: Debunking the Greatest Myths in American Journalism.
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 4 of 4 reviews
August 20, 2021
Back when I taught data analysis/economics to graduate journalism students, I would warn them about "The Wilder Effect". Douglas Wilder was the first Black person to serve as Governor of Virginia. Pre-election polls indicated he would win in a landslide. Actually, he narrowly eked out a win against his Republican rival. A more recent analysis was based upon social interactions with women living in Manhattan during the "Giuliani era". They would almost unanimously state their disgust with all that Giuliani stood for (and this was years before ol' Rudy went off the deep end, mind). However, an analysis of Upper West and East Side precincts showed an overwhelming majority vote for Giuliani - almost rivaling the vote in Staten Island - a noted GOP enclave.
This is all background before I visited this work. The author does spend considerable time noting the biases inherent in asking people how they will vote, or in the case of post-polling, how they voted. He does extol the virtues of meta-polling, i.e., aggregating several polls in an attempt to reduce natural variation. However, my sole wish is that he would have written this work in time to cover the 2020 election, as several attempts to aggregate (particularly that of The Economist) predicted a "Blue Wave" which would swamp Trump. The fact that this did not happen gave some credence to Trump's outlandish claims of voter fraud- claims which would be laughed at were the electoral vote not as close. This book does not (and should not) attempt to address the inherent inequality of our electoral system. To his credit, he does look at "woke" (and not truthful) tendencies of respondents to say one thing, and do another. The other facet, to his credit, is that he examines the methodological journey from in-person interviewing toward internet panels - and the difficulties of making projections based upon these all-too-common characteristics of human behavior. His range of polling inaccuracies covers from 1936 Willkie v. FDR polls to 2016's Clinton v. Trump. Highly recommended.
This is all background before I visited this work. The author does spend considerable time noting the biases inherent in asking people how they will vote, or in the case of post-polling, how they voted. He does extol the virtues of meta-polling, i.e., aggregating several polls in an attempt to reduce natural variation. However, my sole wish is that he would have written this work in time to cover the 2020 election, as several attempts to aggregate (particularly that of The Economist) predicted a "Blue Wave" which would swamp Trump. The fact that this did not happen gave some credence to Trump's outlandish claims of voter fraud- claims which would be laughed at were the electoral vote not as close. This book does not (and should not) attempt to address the inherent inequality of our electoral system. To his credit, he does look at "woke" (and not truthful) tendencies of respondents to say one thing, and do another. The other facet, to his credit, is that he examines the methodological journey from in-person interviewing toward internet panels - and the difficulties of making projections based upon these all-too-common characteristics of human behavior. His range of polling inaccuracies covers from 1936 Willkie v. FDR polls to 2016's Clinton v. Trump. Highly recommended.
November 19, 2020
pretty decent book. Wish they would have gone deeper into the mechanics of pollings. More of a straight-up history of presidential elections and when the polls get it wrong.
Read
June 8, 20243.5 STARS
How this book ended up on my TBR: Front table at the library. Interest was piqued.
If you are like me, you've ever heard polls and news coverage of elections and rolled your eyes all the way back in your head. I feel like I've been bamboozled plenty of times by expectant polls that seem to predict absolutely nothing. They only seem to serve to get my hopes up. (Some pollsters would no doubt argue that prediction isn't the name of the game.) Somewhat to my satisfaction, this book highlights several US presidential elections when the polls were somewhat wrong or very, very wrong. Each chapter focuses on a very specific election, though each case is different, and Campbell provides more context in later chapters. Given my brief perusal of the other reviews, this book stays in the middle ground of too much historical context and too much about the polls themselves. Some say, for example, that they would want more information about the evolutions of polling techniques. That would not be of too much interest to me.
There are some common themes, though. When the polls are wrong, everyone (especially journalists) just piles on the criticism. There's a lot of rhetoric on pollsters being out of touch with the general public, or they avoid or just do not consult specific segments of the audience. I think the book matches the title and summary, and I liked the historical perspective and demonstration that each election is different. I do wish that Campbell had included headings and subheadings for particular topics within each chapter, especially when he briefly wrote about other elections that didn't have their own chapter but sort of bridged the distance between two chapters. I would have liked some guidance on the significance of different points for each chapter. The 3.5 rating mostly comes from my enjoyment of reading the book. That is, it wasn't exactly fun, but it was worth the read.
Though I suppose it was not strictly necessary, given the range of the book, I would have enjoyed reading more about why we still use polls. Campbell addresses this question in the epilogue, and while he does suggest some answers, I suppose I would have enjoyed more elaboration. But maybe we just don't have the research for that. I also would have enjoyed more historical context for each of the earlier chapters, but I'd be fibbing if I said I wasn't more interested for the presidential elections I've lived through.
--
PS. If you all ever go to The StoryGraph, let's be friends there! Here's my profile.
How this book ended up on my TBR: Front table at the library. Interest was piqued.
If you are like me, you've ever heard polls and news coverage of elections and rolled your eyes all the way back in your head. I feel like I've been bamboozled plenty of times by expectant polls that seem to predict absolutely nothing. They only seem to serve to get my hopes up. (Some pollsters would no doubt argue that prediction isn't the name of the game.) Somewhat to my satisfaction, this book highlights several US presidential elections when the polls were somewhat wrong or very, very wrong. Each chapter focuses on a very specific election, though each case is different, and Campbell provides more context in later chapters. Given my brief perusal of the other reviews, this book stays in the middle ground of too much historical context and too much about the polls themselves. Some say, for example, that they would want more information about the evolutions of polling techniques. That would not be of too much interest to me.
There are some common themes, though. When the polls are wrong, everyone (especially journalists) just piles on the criticism. There's a lot of rhetoric on pollsters being out of touch with the general public, or they avoid or just do not consult specific segments of the audience. I think the book matches the title and summary, and I liked the historical perspective and demonstration that each election is different. I do wish that Campbell had included headings and subheadings for particular topics within each chapter, especially when he briefly wrote about other elections that didn't have their own chapter but sort of bridged the distance between two chapters. I would have liked some guidance on the significance of different points for each chapter. The 3.5 rating mostly comes from my enjoyment of reading the book. That is, it wasn't exactly fun, but it was worth the read.
Though I suppose it was not strictly necessary, given the range of the book, I would have enjoyed reading more about why we still use polls. Campbell addresses this question in the epilogue, and while he does suggest some answers, I suppose I would have enjoyed more elaboration. But maybe we just don't have the research for that. I also would have enjoyed more historical context for each of the earlier chapters, but I'd be fibbing if I said I wasn't more interested for the presidential elections I've lived through.
--
PS. If you all ever go to The StoryGraph, let's be friends there! Here's my profile.
December 26, 2021
Lost in a Gallup is a short, engaging history of Presidential political polling failures in the USA. The book spans the eighty years from the incorrect 1936 Literary Digest prediction that Alf Langdon would wipe the floor with FDR to the widespread 2016 predictions that Donald Trump had as much chance of sitting in the White House as the Cleveland Browns had of winning the Super Bowl that year (or any other year). The "Dewey Beats Truman" 1946 election may be the best remembered polling mishap, but Campbell shows it was not an isolated incident.
Each of the main polling failures is given its own chapter, covering the basic history of the election and where the polls went wrong. The polling failures are explained crisply for the non-expert, spelling out simply the varying issues with non-response bias, predicting likely voters, and timing of polls that have plagued pollsters.
The story of polling failures is not just that polling has failed, but that there is a complex relationship between polling and journalism that arguably does neither any favours. Certain factions of journalists have historically argued for replacing polls with an old-fashioned "shoe-leather" approach of conducting myriad personal interviews, with Campbell correct that this may provide some insight into views but cannot provide the necessary quantitative detail (as proven in the UK where the occasional by-election without polling has resulted in press narratives entirely missing the mark).
Campbell's final reckoning is correct that downplaying polling is a wise move, albeit that he rather lets the polling industry off the hook. While the technical advances of polling have been huge, much of the industry does very little to deal with the public and journalistic misunderstandings highlighted in Lost in a Gallup. Pollsters cannot continue to blame the media for misrepresenting polls while at the same time being only too happy to accept money and publicity from the same companies.
Campbell's work is a valuable reminder of the hard work pollsters have done to provide reasonably accurate polls despite the myriad challenges. Issues with the Presidential polls in 2020 (after the publication of this book) show there are no signs that polling has got any easier. While the history of polling is generally one of fairly accurate results, there will be more chapters on failures for Campbell to write in the future.
Each of the main polling failures is given its own chapter, covering the basic history of the election and where the polls went wrong. The polling failures are explained crisply for the non-expert, spelling out simply the varying issues with non-response bias, predicting likely voters, and timing of polls that have plagued pollsters.
The story of polling failures is not just that polling has failed, but that there is a complex relationship between polling and journalism that arguably does neither any favours. Certain factions of journalists have historically argued for replacing polls with an old-fashioned "shoe-leather" approach of conducting myriad personal interviews, with Campbell correct that this may provide some insight into views but cannot provide the necessary quantitative detail (as proven in the UK where the occasional by-election without polling has resulted in press narratives entirely missing the mark).
Campbell's final reckoning is correct that downplaying polling is a wise move, albeit that he rather lets the polling industry off the hook. While the technical advances of polling have been huge, much of the industry does very little to deal with the public and journalistic misunderstandings highlighted in Lost in a Gallup. Pollsters cannot continue to blame the media for misrepresenting polls while at the same time being only too happy to accept money and publicity from the same companies.
Campbell's work is a valuable reminder of the hard work pollsters have done to provide reasonably accurate polls despite the myriad challenges. Issues with the Presidential polls in 2020 (after the publication of this book) show there are no signs that polling has got any easier. While the history of polling is generally one of fairly accurate results, there will be more chapters on failures for Campbell to write in the future.
Displaying 1 - 4 of 4 reviews