Poll Shows Palin Speech Helped Her Image (or Did it?)

A poll that came out late last week purports show that Palin’s speech helped her public image. However, if we look at the numbers, that claim is a weak one.

Steven L. Taylor

January 16, 2011

Comments

A comment to my post on what I (and and others consider a lost opportunity for Sarah Palin points to a poll that purports to find Americans View Sarah Palin as more Sincere and Believable after Watching Speech.

The poll comes from HCD Research, which (according to the description on their “About Us” page) seems to indicate they do focus group study that uses respondents who use dials to indicate their reactions to video and audio).

Specifically, the findings where as follows. The scale is a 1-7 scale with 7 being best and 1 being worst.

In regards to sincerity:

Democrats went from 2.69 from 2.62 (+.07)
Republicans went from 5.45 from 5.25 (+.20)
Independents went from 3.85 from 3.68 (+.17)

The likeability number are a mix (slightly down for Ds, and slightly up for Rs and Is).

The believability changes were the best:

Democrats + .10
Republicans + .32
Independents + .35

However, I have two key questions:

1) Given the scale under use and the methodology employed, are those statistically significant changes? My guess is that they are not (especially on sincerity).

2) I am unclear on how these numbers were derived and what the basis of the comparison is. If HRC does focus-group based studies , I am not sure how they can compare the reaction to this speech to a previous one, and if that is what they are doing, that by itself would explain small variations in the results.

It is also interesting to note that while Palin has the above mentioned slight improvement in terms of perceptions of sincerity and believability, there is a uniform (D, R and I) ~~decrease~~ increase (correction–I misread the table initially) of 2% on the question “Would you support Sarah Palin if she were to run in the Presidential Election in 2012?”

I have no doubt that that number is well within the margin of error. Still, it is telling, I think, that the headline focuses on the small shifts that are pro-Palin rather than the small shifts that are less so (or that are within the MOE). The most honest reading of the numbers would seem to indicate that the speech had no actual effect on public opinion of Palin.

18 responses to “Poll Shows Palin Speech Helped Her Image (or Did it?)”

Doug Mataconis

January 16, 2011 at 1:25 pm

I saw this story — I don’t really think it qualifies as a poll in the traditional sense of the word — yesterday and was unconvinced like you.

The methodology seems odd to say the least and the changes within the margin of error.

I’ll be interested to see what more traditional polls show about Palin, as I am sure we’ll see some numbers from the usual suspects over the next several days
Steven L. Taylor

January 16, 2011 at 1:40 pm

That was my initial reaction as well. I only gave it a deeper look when the comment mentioned it as an attempted refutation (or, perhaps, refudiation) of my post. I started to respond in the comments, but it grew to post-length rather quickly.
ponce

January 16, 2011 at 2:15 pm

Would you support Sarah Palin if she were to run in Presidential Election in 2012?

Independents:
Before speech: 73% No
After Speech: 71% No

Ouch!
1. Doug Mataconis
  
  January 16, 2011 at 2:18 pm
  
  ponce,
  
  Where do you get those numbers from?
  
  Just curious
floyd

January 16, 2011 at 2:29 pm

One thing is certain, nothing she could ever say or do would affect her image at OTB.
Steven L. Taylor

January 16, 2011 at 2:34 pm

@Doug: those numbers come from the survey. I made an error in my original post (now corrected). However, the point remains in force: at best these numbers show no real change.

@floyd: I will confess that she is unlikely to do any to do anything to change my opinion, although I hold out the possibility that she could. However, on balance she is quite good at reinforcing, rather than changing, my views of her.
anjin-san

January 16, 2011 at 2:36 pm

sure it would Floyd. she could’ve taken the high road. she could of said we all need to do better and I’m going to commit to just that . I certainly would of given her credit has she done this. but no, she plays the victim card. apparently it is the only card in her deck.

I for one, would be very pleased to see palin start to act like a responsible leader. if she does I may never lend support to her, but I will give her credit for becoming a serious person.
Steven L. Taylor

January 16, 2011 at 2:37 pm

For clarification purposes, the Yes numbers for inds gos from 27% to 29%.
Xrlq

January 16, 2011 at 3:21 pm

@Steven, it sounds to me like you are the one reinforcing your own stereotypes here, not Palin. Other polls may (or may not) differ but for now, the only rational read on the numbers is that Palin’s response likely helped her a little, and almost certainly did not hurt. But that answer didn’t fit the script, so here we are.
ponce

January 16, 2011 at 3:45 pm

“Where do you get those numbers from?”

http://mediacurves.com/Politics/SarahPalinonTucson/Index.cfm
PJ

January 16, 2011 at 4:25 pm

Xrlq:
“the only rational read on the numbers is that Palin’s response likely helped her a little, and almost certainly did not hurt.”

The only rational read, is that the changes are within the margin of error.
PJ

January 16, 2011 at 4:34 pm

“A new national study among 1,437 self-reported Democrats, Republicans and Independents revealed that Americans indicated that Sarah Palin was more sincere and believable after viewing her speech in response to the shootings in Tucson.”

“1,437” and “viewing” are the two most interesting words here.

It’s obviously not a focus group, since the size is too big. Neither was this regular phone polling since the polled was viewing her in the middle of it.

So, it’s some sort of internet study. How was it done? How was the 1,437 picked? Did they register themselves? Obviously, they would need to have access to a computer, and be enough computer savy to do it. They also needed to be motivated.

I’d argue that the study is flawed, and that you can’t draw any kind of conclusions from it.
Steven L. Taylor

January 16, 2011 at 4:40 pm

@Xlrq:

As PJ notes, and as I said above, the numbers are clearly within the MOE at worst or represent very, very small changes at best. As such, the study is not evidence of much of anything.

@PJ: exactly. Part of my skepticism about the poll/study is that it is wholly unclear how the data was collected or how the sample was selected.
Smooth Jazz

January 16, 2011 at 5:21 pm

“@Steven, it sounds to me like you are the one reinforcing your own stereotypes here, not Palin. Other polls may (or may not) differ but for now, the only rational read on the numbers is that Palin’s response likely helped her a little, and almost certainly did not hurt. But that answer didn’t fit the script, so here we are.”

Bravo, Bravo, Bravo; You took the words right out of my mouth. This is a Anti-Palin site, on par with DailyKOS, Tbogg, Huff Post, as you can tell from the all the Lib commenters who flock to here to bash her day after day. Your point about fitting the “script” is so appropo WRT to this blog and the New York/Wash DC echo chamber where the same Liberal opinion makers sing kumbaya to each other all day. I suspect the posters of this blog are having trouble digesting a survey like this which runs counter to the canard they have been pushing all week, that Gov Palin is finished
.

Perhaps most important, notice it took a response to a Steven Taylor get a separate post on this survey since it was favorable to Gov Palin. I’m sure if the survey was unfavorable to her, OTB would more than likely have a post up right away without being prompted – as Doug M often does when he gets a DailyKOS/PPP or other Lib poll that is unfavorable to Gov Palin.

I honestly don’t believe this blog is capable of processing any survey that is favorable, or at least neutral, to Gov Palin. Their first instinct I’m sure is to pooh pooh and try to negatively spin any survey that is in her favor, in order to reinforce their own negative biases. All week they’ve been hiving up “the video hurt her” meme and declaring her relevancy over, and along comes a survey, of over 1400 respondents from all parties, which refutes that conventional wisdom. And they are tying themselves in knots trying to explain it away, especially that main Palin Hater Doug M.

The survey does not convey anything about her in a specific election matchup, but is DOES indicate that the video DID NOT HURT HER amongst the general public – a concept the posters and members of this blog have a hard time dealing with.
Steven L. Taylor

January 16, 2011 at 5:28 pm

@Smooth Jazz:

You assert:

I honestly don’t believe this blog is capable of processing any survey that is favorable, or at least neutral, to Gov Palin

Perhaps you missed the last sentence of my post:

The most honest reading of the numbers would seem to indicate that the speech had no actual effect on public opinion of Palin.

That would be the very definition of “neutral.”

This post was not about Palin, actually, but about the poll itself and the most reasonable interpretation of the numbers.
Ernieyeball

January 16, 2011 at 8:37 pm

SJ sez (with disdain I assume):
…the New York/Wash DC echo chamber where the same Liberal opinion makers sing kumbaya to each other all day.

I ask would it be okay if they sang “Jesus Loves Me”?
jwest

January 17, 2011 at 6:36 am

It’s encouraging that someone at OTB is looking beyond the headlines of polls, but as always, only on those polls that are favorable to Palin.

Had the same scrutiny been applied to the latest NBC/WSJ poll comparing Obama and Palin’s favorability, one would have questioned that 23% of the people contacted were in union households. This is a curious, considering only 12.3% of the population is unionized and a portion of that figure must be two or more union members living at the same address.

Still, it’s nice to know you’re paying attention.
Xrlq

January 24, 2011 at 9:09 am

You seem to have an odd definition of neutral. Positive movement is not neutral even though it’s within the MOE. All the MOE means is that we can’t be 95% certain that Palin’s speech helped her. It doesn’t change the fact that there’s more evidence it did than didn’t. Hence my choice of the word “probably,” which both of you conveniently ignored.

Put differently, if a small increase within the margin of error is too little evidence to conclude that the speech helped her, it’s no evidence at all to support your claim that it didn’t.