the audit

A Web survey isn’t a poll, CNBC

The network's tweet creates a misleading media narrative on the veep debate
October 12, 2012

Sign up for The Media Today, CJR’s daily newsletter.

Whoever was running the CNBC Twitter feed last night didn’t know the difference between a scientific poll and a Web poll:

As I’m writing this, that misinformation has been retweeted 4,838 times, favorited 405.

Eh, it’s just the Twitters, you say. Problem is, that tweet was seen by countless more readers, including ones with megaphones at Politico, The Guardian, and The Daily Caller, and more, who used the bogus numbers to report, erroneously, that more polls called the debate for Ryan than for Biden. That spread the misinformation much further.

Sign up for CJR’s daily email

Politico puts this headline “above the fold” on its home page: “Snap polls: Ryan 2, Biden 1” and writes that “A CNBC snap poll shows a similar result, with 53 percent of respondents identifying Ryan as the winner. By contrast, 41 percent thought that Biden won, while 6 percent believed that neither won.”

The Daily Caller reported this in a story that reported, falsely, that “Most polls and media talking heads gave the advantage in the vice presidential debate to GOP vice presidential candidate Paul Ryan”:

A CNBC poll said that 56 percent thought Ryan was the winner, while 36 declared Biden to be the winner.

Less surprising is that Michael Barone of The Washington Examiner made the boo-boo too, as did National Review and Breitbart.com.

Still all of it is just plain sloppy.

First, why would CNBC of all networks, do a snap poll of the vice presidential debate? If anyone at NBC was going to do such a poll, it would be NBC or MSNBC. CNBC’s usually showing get-rich-quick infomercials at that time of night.

Second, it’s clear that Politico didn’t even bother to actually visit the CNBC website to look at this “snap poll.” If it had, it would have seen that it was a SurveyMonkey-style Web poll, which is worse than useless: Unrepresentative samples that are easy to manipulate.

Third, neither actually links to their bogus source.

A little over an hour after its first “poll” tweet, CNBC tweeted again with dramatically different numbers (and making the same error):

Not to excuse Politico and The Daily Caller for their poor reporting, but CNBC’s misleading tweets instigated a false media narrative that spread far beyond its Twitter feed.

If you want to have Web polls on your site, fine. But don’t report on them for crying out loud. It’s just not smart.

Ryan Chittum is a former Wall Street Journal reporter, and deputy editor of The Audit, CJR’s business section. If you see notable business journalism, give him a heads-up at rc2538@columbia.edu. Follow him on Twitter at @ryanchittum.