The larger question behind a predictable ‘social experiment’

In what may be the irony of the decade, a reputed polling company called Media Strategies posted this statement on social media earlier this month: “Polling has a credibility problem. Ours is the fix.”

According to multiple news sources, the organization later reported poll results related to primary elections in three states, including the races for Los Angeles mayor and Wisconsin governor. On Monday, the company shuttered its web site and acknowledged it had fabricated those results as part of a “short-term social experiment examining how purported polling information could enter and spread through the political information ecosystem without independent verification.”

The limited degree to which a “social experiment” was needed to demonstrate the potential harm of publishing unverified polling results is hardly debatable. Any thinking adult could immediately recognize the danger inherent in an agency founded on trust wantonly ignoring the actions required to justify it. Indeed, Media Strategies’ failure to include standard descriptions of how its “poll” was conducted warned legitimate pollsters such as The New York Times, Associated Press and others away from including its results in their poll reporting.

So, in that context, Media Strategies’ stunt was more an object lesson than a social experiment. It demonstrated consequences that people already could predict — notably that voter behavior could be affected by untrustworthy polling and that campaigns could, even if unwittingly, try to use incorrect results to their advantage.

In essence, the company’s action didn’t “fix” any presumed “credibility problem” of polling. It made it worse.

That is unnerving enough. But, with the midterm election campaigns just weeks away from shifting into post-Labor Day high gear, the greater concern for voters isn’t whether they should trust this poll or that one. It’s how they use these surveys in the first place.

I have always urged people to be wary of even legitimate, responsible polling. Its primary function seems to be merely to serve as an announcer describing the state of “the horse race.” If you care about the issues that affect your voting decisions, you should be paying attention to candidates’ positions on those issues, not their positions in the field as they round the track going into the final stretch.

Polling, to the extent that it is reliable, may help spur your commitment to making it to the voting booth, and it may help you and anyone inclined toward punditry to analyze what an election says about the public will. But it ought not affect your vote itself.

The Media Strategies case certainly reminds us of all sorts of themes related to credibility these days. Be skeptical, for instance. Know the source of the data you are getting. Know how much you can trust the provider relaying it to you. Be open to examining multiple sources.

It begins to sound like the admonitions related to evaluating print, broadcast and online news, and indeed similar warnings apply to all information providers. But as the issue applies to polling, there is another worthwhile message to consider. Don’t just question the extent to which you can trust the information you’re being given. Also understand how you are using it. If the results of a poll are influencing how you vote, you are probably using the wrong criteria.

• Jim Slusher, jslusher@dailyherald.com, is managing editor for news and opinion at the Daily Herald. His book “To Nudge The World: Conversations, community and the role of the local newspaper,” a 2025 Chicago Writers Association Book of the Year, is available at eckhartzpress.com.