Mellman: Polls vs. initiative results

People for gun reform hold signs as they attend a rally held by multiple groups outside the U.S. Capitol in Washington, D.C., on Monday, June 6, 2022.
Greg Nash
People for gun reform hold signs as they attend a rally held by multiple groups outside the U.S. Capitol in Washington, D.C., on Monday, June 6, 2022.

Last week, The New York Times’s astute chief political analyst, Nate Cohn, raised serious doubts about polling on background checks for gun purchasers and, by extension, about the efficacy of issue polling generally. 

The heart of his case: “When voters in four Democratic-leaning states got the opportunity to enact expanded gun background checks into law [through ballot initiatives], the overwhelming support suggested by national surveys was nowhere to be found.” 

Cohn nods to the counterargument—“Initiative and referendum results are not a perfect or simple measure of public opinion”—but he seems to repose more faith in the ballot measure results than in the poll data.  

I don’t.  

You wouldn’t know from his article’s text that the background checks measures won in three of the four states, but Cohn’s argument is not about winning and losing.     

A raft of polls find 80 percent to 90 percent of voters favoring background checks for all gun purchasers. Yet, ballot measures to do just that garnered far less support, with “yes” votes ranging from 48 percent to 63 percent. 

Having done the polling for one of these measures (that won narrowly), and for scores of other high stakes initiatives, I conclude that initiative results are frequently poor measures of public opinion on the underlying issue. 

First, ballot language usually transforms simple, clear ideas into complex, off-putting legalese.  

A typical poll question on background checks from the ABC-Washington Post poll asked, “Would you support or oppose requiring background checks on all potential gun buyers, including private sales and gun shows?” 

On the ballot voters saw, “Shall Chapter 202 of the Nevada Revised Statutes be amended to prohibit, except in certain circumstances, a person from selling or transferring a firearm to another person unless a federally licensed dealer first conducts a federal background check on the potential buyer or transferee?”  

Much more complicated. Much less clear.  

We tested several versions of potential ballot language, drafted by lawyers to describe the same measure. Support levels for the identical measure differed by 25 points, depending on the precise wording.  

In studying hundreds of initiative polls on a host of topics, from taxes to gaming, medical research, vaccination mandates, abortion, education, civil rights, building projects and more, I’ve never seen ballot language that fares as well as the plain English version of the idea.  

To offer but one example from another realm (in which we weren’t involved), polls consistently find 89 percent to 95 percent support for requiring labels on genetically modified food. Nonetheless, Washington state voters defeated a measure mandating such labels.  

Public polls employing wording akin to the ballot language found support ranging from 45 percent to 66 percent — far lower than for the concept. 

Second, there is the campaign. Defeating ballot measures is about finding the flaw in the law. The flaw need not be relevant to the core concept. Too often, if opponents can’t find a flaw, they invent one.  

In the Nevada gun-show-loophole fight, opponents argued that lending your gun to your niece while hunting together would make you a criminal. Never mind that the measure specifically exempted such situations, a fact confirmed by law enforcement in ads.  

Despite its falsity, the attack moved votes.  

Opponents of the Washington GMO measure, hostile to any GMO labeling, sponsored ads suggesting the measure was a good idea that didn’t go far enough because it exempted some GMO foods. In essence they argued, “let’s have a stronger law.” Extremely disingenuous, but quite effective.  

Finally, there’s the matter of identity. From the very beginning, we warned our Nevada clients that over half the electorate were gun owners, Republicans or both, and if this vote was transformed from an “issue” to an expression of identity, we would be in trouble.  

Our opponents knew this too and worked toward that goal with important success, as a solid phalanx of GOP leaders and gun groups opposed the measure.   

Complex ballot language, dishonest or disingenuous campaigning and identity politics combine to divorce initiative results from public opinion on background checks, and lots of other issues. 

Mellman is president of The Mellman Group and has helped elect 30 U.S. senators, 12 governors and dozens of House members. Mellman served as pollster to Senate Democratic leaders for over 20 years, as president of the American Association of Political Consultants, and is president of Democratic Majority for Israel.   

Tags Ballot initiatives Election Gun control opinion polling Polling voters

The Hill has removed its comment section, as there are many other forums for readers to participate in the conversation. We invite you to join the discussion on Facebook and Twitter.

See all Hill.TV See all Video

Most Popular

Load more


See all Video