• AP updates guidelines on polls. No more poll results as main headline.
    7 replies, posted
AP Definitive Source | Updating our polling guidance Most important section:  “poll results that seek to preview the outcome of an election must never be the lead, headline or single subject of any story.” I don't think this will change too much, but at least AP and news organizations that use them (like local news) won't lead off with "NEW POLL SAYS CANDIDATE A HAS A 53-47 LEAD!"
Hopefully this has a positive effect on 2020. I wonder if headlines with polls created the "Hillary will win by a landslide all the polls say so" thing everyone expected in 2016.
You won't believe the poll results for the next election!
“It is more vital than ever for policymakers, journalists and citizens to become better informed consumers of surveys and data,” “We are proud to work with The Associated Press and support them in their continued efforts to promote rigorous standards for reporting on survey research and to help the public understand key characteristics of reliable data.” This is good because poll results are far too easy to misinterpret, and the fickle nature of polling is lost when the results are reported as-is devoid of context.
I'd say this is more a reaction intended to maintain the credibility of the Press. People like poll numbers. Poll numbers summarize what you want to know. Nobody who's browsing headlines wants to read, "Statistical Likelihood But Hedge Your Bets", they want to read, "IT'S A SURE THING." All of the major ones (okay, except certain ones,) lost a lot of face, a lot of faith, and a lot of pride when the 98% exploded in their hands. AP Guidelines are supposed to be what sets papers, broadcasters and the whole range of journalism apart from blogs, tabloids and word of mouth. If this had been a move with real integrity behind it, it would've happened literally decades ago, because this has been a problem in reporting for actual decades. But better late than never, so whoopee for that.
I feel as if journalists should force its readers to understand its conclusions rather than let them make assumptions just based on the numbers. For example, they should remove graphs of polls from articles altogether. Complemented with this about removing poll numbers from headlines, it'll force readers to actually read the article to find the numbers. And perhaps along the way, they'll see a realistic explanation rather than creating an assumption.
IMO, a good rule for journalists to follow would be to only utilize a minimum of a 5 day average poll number with at least 3 different pollsters, or something like that. Using single polls, from single days, from single pollsters is just bad journalism.
Are you saying graphics can be misleading? Don't know what you're talking about https://i.imgur.com/FzOMq5j.jpg (Not actually from a news article but from the Lib Dems themselves lol)
Sorry, you need to Log In to post a reply to this thread.