Washington Post Poll vs Catania Internal Poll

[Anonymous]

Jeff, I will probably vote for Catania but think he has no chance because the business community has lined up behind Bowser. Not because they don't think Catania is smart---they do---but because they think he is capricious and vindictive. One business person told me that he actually thought Catania would continue to be much more effective as a councilmember---given his strengths and weaknesses.

[jsteele]

Anonymous wrote:Jeff, I will probably vote for Catania but think he has no chance because the business community has lined up behind Bowser. Not because they don't think Catania is smart---they do---but because they think he is capricious and vindictive. One business person told me that he actually thought Catania would continue to be much more effective as a councilmember---given his strengths and weaknesses.

Unfortunately, Catania's Council term ends at the end of the year. So, if he doesn't win the mayoral election, he will be out of politics. Didn't the business community support Cropp against Fenty and then Fenty against Gray?

[jsteele]

jsteele wrote:

Anonymous wrote:Jeff, did you get a chance to see the questions that were asked? The argument about the sampling is persuasive but I'd also like to know whether the questions were asked in a neutral manner.

I was actually polled, so I didn't need to see the questions. The first part of the poll was pretty standard: a few random questions about the direction of the District and so on. Then, the horse race question: if the election were held today would you vote for Democrat Muriel Bowser, Independent David Catania, etc. Then, a part where messaging was tested. This part was very biased toward Catania and described how he was raised by a single mother, etc. while Bowser was linked to a bunch of scandals. Then, the horse race question was asked again. Ben assures me that they are only counting the first horse race results and that Catania was actually ahead in the second one (believable given that Catania was presented as an American dream come true while Bowser was Boss Hogg).

If I were to critique the poll, I would ding it for being automated -- though there are various opinions on this point. My main criticism is that automated polls can't call cell phones. In the Post poll, Catania did worse among cell phone respondents than among line lines. That might support the theory that younger voters are more likely to pick Bowser because of the "D" label.

In it's favor, the polling firm is well-respected and has worked for many prominent Democrats including Cory Booker and Elizabeth Warren.

I'm replying to my own post to correct something about which I was in error. In the above post, I criticized the poll for being automated. FCC regulations prevent automated calls to cell phones and, therefore, I assumed that cellphones were not included in the poll. However, I have now learned that while landline calls were automated -- and that's what I received when I was polled -- live callers also surveyed cell phone numbers. As a result, both landlines and cell phones were included in the poll. That removes one of my concerns about the Catania poll.

[jsteele]

There is a public poll out today that might be said to support the Catania campaign argument (but with a caveat):

http://www.washingtonpost.com/blogs/mike-debonis/wp/2014/10/01/new-d-c-mayoral-poll-muriel-bowser-leads-david-catania-by-8-points/

"A new poll commissioned by a D.C.-based, pro-business advocacy group finds a tighter D.C. mayoral race than seen in previously published polling, with Democrat Muriel E. Bowser leading independent David A. Catania by eight percentage points among likely voters."
...
"The new poll differs from the NBC4/Washington Post/Marist poll in selecting respondents from a voter list rather than by dialing random numbers."

The Catania argument has been that random dialing reaches people who don't accurately say whether they are registered to vote or likely to vote and who are not well-informed about the election, answer based on party affiliation, and are unlikely to vote. The result is inflated numbers for the Democrat.

This poll would seem to support that contention. By using voter lists, it reaches people with a history of voting who are known to be registered. Both polls, as well as Catania's internal poll, found similar levels of support for Catania. The Post showed much higher numbers for Bowser, just as Catania has argued random polling would show. Therefore, this poll supports the contention that the race is much closer than the Post poll suggests.

However, here is the caveat, this poll may have under-sampled African Americans. The impact of that is something I simply don't know enough about. Obviously, sampling more AAs would probably help Bowser, but I don't know by how much.

[Anonymous]

jsteele wrote:There is a public poll out today that might be said to support the Catania campaign argument (but with a caveat):

http://www.washingtonpost.com/blogs/mike-debonis/wp/2014/10/01/new-d-c-mayoral-poll-muriel-bowser-leads-david-catania-by-8-points/

"A new poll commissioned by a D.C.-based, pro-business advocacy group finds a tighter D.C. mayoral race than seen in previously published polling, with Democrat Muriel E. Bowser leading independent David A. Catania by eight percentage points among likely voters."
...
"The new poll differs from the NBC4/Washington Post/Marist poll in selecting respondents from a voter list rather than by dialing random numbers."

The Catania argument has been that random dialing reaches people who don't accurately say whether they are registered to vote or likely to vote and who are not well-informed about the election, answer based on party affiliation, and are unlikely to vote. The result is inflated numbers for the Democrat.

This poll would seem to support that contention. By using voter lists, it reaches people with a history of voting who are known to be registered. Both polls, as well as Catania's internal poll, found similar levels of support for Catania. The Post showed much higher numbers for Bowser, just as Catania has argued random polling would show. Therefore, this poll supports the contention that the race is much closer than the Post poll suggests.

However, here is the caveat, this poll may have under-sampled African Americans. The impact of that is something I simply don't know enough about. Obviously, sampling more AAs would probably help Bowser, but I don't know by how much.

Thanks for the insight Jeff. I am with you on a lot of this, but I can't get over how far off the black/white percentages are, and how that would affect the final numbers. Do you have any insight or scuttlebutt from the Bowser camp related to what their internal polls are saying? Chuck Thies said on Twitter that her internal polls were closer to this 8 point gap.

[jsteele]

Anonymous wrote:
Thanks for the insight Jeff. I am with you on a lot of this, but I can't get over how far off the black/white percentages are, and how that would affect the final numbers. Do you have any insight or scuttlebutt from the Bowser camp related to what their internal polls are saying? Chuck Thies said on Twitter that her internal polls were closer to this 8 point gap.

I will quote what I wrote in my original article in this thread:

"Young says that he has heard that the Bowser campaign's internal polls also show a very close race."

"Young" is Ben Young, Catania's campaign manager. Other than that, I have no information about Bowser's polls. I have friends who are working in her campaign, but for reasons I don't understand, they refuse to share private information with me.

[Anonymous]

jsteele wrote:

"Young says that he has heard that the Bowser campaign's internal polls also show a very close race."

"Young" is Ben Young, Catania's campaign manager. Other than that, I have no information about Bowser's polls. I have friends who are working in her campaign, but for reasons I don't understand, they refuse to share private information with me.

Thanks, putting it all together, maybe the true gap is indeed lower than 17, but not quite as low as 8. Somewhere in the 10-12% range sounds about right based on all of this. With 27% undecided or going for fringe candidates in this poll, that seems a little more possible for Catania than when the gap was 17% and undecided/fringe was only 15%. But I would still give him 10-15% odds of winning, of course that's up from probably 5% based on the last poll. I still see a very narrow way he can put together enough votes. Nothing pops out that will swing enough people, at least right now.