The Edge: Competitive Edge Blog

Marty Tries to Use a Poll to Pull a Fast One October 27, 2015

Print Friendly, PDF & Email

Marty Tries to Use a Poll to Pull a Fast One
October 27, 2015

AtkinsBlockMy previous blog post identified an appalling lack of transparency in the reports of the polling conducted for Assemblywoman Toni Atkins. Bad stuff there.

Not to be outdone, Senator Marty Block released his own internal campaign polling. He used the GaronHartYang Research Group for his polling. This is the same outfit that conducted an infamous Chargers poll a few months back. Block’s October 9 press release made no mention of it, trumpeting that his poll was conducted by the “NBC News/Wall Street Journal Pollster” insinuating it was a media poll rather than his campaign poll.

Block’s release blares, “Poll Shows Block with Double Digit Lead in 39th Senate Race.” Having learned earlier of the contrary results of Atkins’s “mystery poll,” any lead for Block, let alone a double-digit one, is cause for raised eyebrows.

Someone’s gotta be lying, fudging, or just plain wrong.

Block’s release claims the “pollster puts the race at Block 46 – Atkins 35.” Later it authoritatively intones, “the results of this fully representative sampling of likely district voters shows 46% for Block and 35% for Atkins.”

What’s going on here?

The big, fat, glaring problem is that Block’s release heralds the results of a question that was asked after arguments related to the candidates were made. His claim that he’s leading Atkins relies on information given to respondents during the survey interview. The results Block’s touting do not reflect current public opinion – Block’s “lead” is not real.

Block’s pollster did ask a head-to-head trial heat question earlier in the poll before respondents were exposed to the messages. Those results show, as Atkins’ mystery poll did, that the Senator is losing to the Assemblyman.

Block’s spurious claim to a lead is based on “advocacy polling.” Advocacy polls are designed to elicit a specific result from people to suit the client’s needs. They tee up the respondent with “what-if” scenarios before asking the question that gets the toutable results. Knowing the news media rarely digs into the methodology of a poll and simply reports “the numbers” as presented, the campaign hypes the results of their advocacy poll as if it really means something.

Don’t get me wrong, I don’t blame the media (much), and I certainly don’t blame the voters for misunderstanding these polls. I don’t even solely blame the pollsters, though I often regard them as co-conspirators. No, the real bad guys are the campaigners and, ultimately, the candidate.

In the case of the Block poll, we need to see all the message-testing questions that led up to the “money” question. It turns out that question is about Assemblywoman Atkins’ supposed willingness to trash California’s environmental laws. We have no clear picture as to what other messages teed up the results Block wants us to focus on.

Although we’re told the poll’s audience is “fully representative” across political affiliation, age and gender, how could we take this on face value? But the most serious problem is that Block is so eager to portray himself as winning that he lies in his press release. I don’t know how anyone can vote for someone who does that.

In this test of campaign ethics I give Atkins three demerits, but I give Block five demerits because his transgression is more egregious. Block’s dishonesty is worse than Atkins’ lack of transparency, in my opinion.

It doesn’t matter to me who wins. I don’t live in the district and I’m not invested in any candidate in the race. What does matter is seeing pollsters and the public lose when political campaigns fail to play by the book.

This entry was posted in Polls, San Diego. Bookmark the permalink.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *


Featured Research Download

Are you targeting the "persuadables" of your campaign? Download our seven step process to winning an election.