Labor-connected polling firm uses gift cards to entice workers, media don’t question data
 

[March 23, 2015]  By Kathryn Watson | Watchdog.org
 
 ALEXANDRIA, Va. — It took hours for a poll released by Hart Research Associates to circulate the Internet, claiming 87 percent of fast-food workers were injured on the job in the last year.

The poll’s release conveniently comes as the big-labor movement to raise the minimum wage in the U.S. to $15, dubbed the “Fight for 15,” reaches a two-year anniversary, and as McDonald’s workers file 28 health and safety complaints against the world’s largest fast-food chain.

But to get those responses, the Democratic polling firm that’s worked with groups like the Service Employees International Union enticed respondents to the online poll by offering them the chance to win a gift card, according to the Huffington Post. After its original story on the poll, the Huffington Post followed up with Hart partner Guy Molyneux, who confirmed — and defended — the polling method. The gift card enticement is one that Hart has been criticized for employing before.

Media outlets from Mother Jones to BuzzFeed to the New York Daily News ran with the poll unquestioningly anyway.

Last year, Washington Post media reporter Erik Wemple picked apart coverage of a previous poll by Hart, which claimed 90 percent of fast-food workers had their wages stolen. In that poll, people were also encouraged to take the survey through Facebook ads.

“FAST FOOD EMPLOYEE? 2 MINUTES, AND YOU COULD SCORE $100 BUCKS,” that Facebook ad read.

Senior Hart employees did not respond to Watchdog.org’s request for an interview or comment.

The poll was commissioned by the National Council for Occupational Safety and Health.

Hart’s final sample of 1,426 adults was weighted to make sure demographics were representative of the nation’s workforce, but doesn’t mention when the survey was conducted or provide other details, as any poll from Gallup, Pew Research, or any other widely trusted research organization does.

In fact, the Associated Press’ Stylebook sets the standard for using polls this way: “Only a poll based on a scientific, random sample of a population — in which every member of the population has a known probability of inclusion — can be considered a valid and reliable measure of that population’s opinions.”

Hart claims to be one of the “leading survey research firms,” on the “cutting edge of change in the field of public opinion.”

Experts generally agree that best polling methodologies entail random samplings either over the phone or in person, although online polls can sometimes be relatively accurate.

William Galston, an adviser to former President Bill Clinton who runs polls for the Brookings Institution, said online polls are more and more common these days.

“Online polling is by no means unusual these days,” Galston told Watchdog.org. “Assuming that the combination of the sample plus weighting ends up mirroring the national pool of fast food workers, there’s no reason to be especially dubious of the results. Sure, in-person surveys continue to be the gold standard, but you can get reasonably reliable results on-line if you’re careful, which the Hart folks are.”

— Kathryn Watson is an investigative reporter for Watchdog.org’s Virginia Bureau, and can be found on Twitter @kathrynw5.

[This article courtesy of Watchdog.]