David Moore – The Opinion Makers (Book Review)
When I first wrote this in 2012, polls showed an incredibly tight US presidential race between President Barack Obama and challenger Mitt Romney. The Realclearpolitics (RCP) average showed Romney leading by less than a percentage point, while the projected delegate total shows a slender lead for Obama. This was on the day of the final debate. Essentially, we were counting down the days to a nail-biting finish on Nov 6.
At least, that’s what the opinion polls told us. Do they actually reflect voter reality?
No, says David W. Moore, a former insider at Gallup. In The Opinion Makers, he makes a persuasive argument that polls today are more convenient ‘news’ fodder than communication tool between the government and the electorate.
Omitted Options/Results
Moore highlights the inherent flaws of such polls right away. First, these polls press for an Agree/Disagree response. This forces voters to adopt a position when they may not have thought about, or even heard about, a national topic, policy or the candidates of an election. Second, when releasing poll data, pollsters do not elaborate on the results presented. In cases when pollsters have additional data from giving ‘undecided’ as an option, they have chosen not to share them to put the lack of understanding of voters in perspective.
How much do these flaws matter? Just pause for a moment and think about these 2 trends:
1) Consistently low single-digit percentages of undecided voters
2) High fluctuations of polling figures
How can we have the vast majority decided about something, and then see them suddenly change their decisions within a week? In the 2012 Republican nomination contest, we saw the likes of Michelle Bachmann and Rick Perry catapulting into the ‘lead’ and then plummeting in support, forcing their withdrawals soon after. This can only mean that so often, we aren’t seeing figures that reflect the whole truth.
Overly Repetitive
The argument Moore has, that polls are failing at their jobs, even manipulating democracy, is a convincing and meaningful one. Those who are concerned about American politics will surely take some insight from reading this book. However, there is too much repetition in much of the book. The first 5 of 8 chapters revolved around the same argument. If you had no impression of or no interest in the multiple examples he brought up, you should get the point and skip the rest.
It gets better towards the end, though, as Moore begins to apply the argument further, exploring government’s role in public policy polling, the significance of new trends such as cellphones and emails on polling accuracy, and what he envisions as the way forward for the polling industry.
Still A Problem?
The steps that this 2009-released book suggested may not have been applied in 2012. As now known, President Obama winded up with a convincing delegate total of 332 to Romney’s 206. Couldn’t October polls be more indicative of a Nov 6 election?
It’s 2016 now, and as I write this in late June, the RCP average projects Hillary Clinton with 46.4% to Donald Trump’s 39.6%. It feels refreshing that 14% are reflected as undecided, which is certainly the most plausible 3rd option for swing voters. I greatly suspect, though, that this indicates the candidates’ deep unpopularity, rather than a methodological acceptance of the ‘Undecided’ option.
Verdict: ★★☆☆☆ (passable)
This book is for those who take an interest in polling or American politics. It will be a revelation for those who have always believed in polls. The quality of the insight makes up for the tedious repetition in the first half of the book. Overall, a decent book that informs us that we are often limited in our understanding of various policies and political issues, and reminds us that the math may not be as straightforward as we think.
