Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
Who Knew? Data on the Efficacy of the Repeal Pledge
Oh, there is nothing like data.
Several days ago, in the wake of the Deb Fischer victory in Nebraska, I put up a post, “Mourdock, Fischer and the Repeal Pledge“. Many of the comments were by people who because they hadn’t heard* of the Repeal Pledge thought it didn’t make any difference, or didn’t believe pledges in general worked, or attributed Fischer’s victory to Palin’s endorsement , or to the two main candidates relentlessly beating up on each other (and it is true, without that groundwork she couldn’t have prevailed). I stopped responding to the comments, because I knew the pollster was in the field right then, trying to find out from Nebraska voters as best he could the answer to those questions .
[*Small digression here, for those of you who rely on news reports to know what happened: it’s probably missing stuff that no one talked about. For example, when we do messaging in a state, we almost never put out a press release about it. And where all we are doing is telling people who has signed the pledge and who hasn’t, and not encouraging anyone re. how to vote, we don’t even have anything to file. So there won’t be any press reports about the role the Repeal Pledge played; it just won’t have gotten on reporters’ radar screens.]
Just to be REALLY clear: in no lifetime would I say that the Repeal Pledge gets credit for anyone’s victory. A bad or not credible candidate could take the pledge repeatedly and it wouldn’t make any difference. Good candidates have won without it. And by definition there are many issues, personalities, and outside groups that each are helping create the outcome.
But now we have data, and the implications for candidates of what we’ve just confirmed are huge.
Back up first: in the Mourdock race in Indiana, the campaign told us explicitly after the fact that the Repeal Pledge had been way more important than we knew. (That’s the Mourdock campaign talking, not us.)
We hadn’t spent that much on promoting who had, and hadn’t, signed the pledge – certainly our outlay was drop in the bucket relative to what else was being spent – but we knew it was an important contrast point that other groups and the campaign could easily amplify, and that they had done so.
We hoped the Mourdock campaign folks were right, but there was no hard data. We noted the correlation between when the word about who had and hadn’t signed the Repeal Pledge got out, and the blowout that rapidly occurred. But we all know that doesn’t PROVE anything.
In Nebraska, we had a unique opportunity with Deb Fischer: a relatively unknown candidate (which was her biggest problem), few other elements promoting her candidacy, and even less amplification of our message. This was a chance to assess how effectively our messaging penetrated, and whether knowing that a candidate had signed the Repeal Pledge actually mattered.
So we did a survey. The results are in this Politico piece. Bottom line: This idea that Rush Limbaugh so long ago encouraged was more important than anyone would have ever dreamed.
- Obamacare/repeal was the single issue that mattered most. (Which was utterly unexpected to me, but maybe stood out because it became a point of key contrast.)
- 62% were voting for their candidate’s positions or personal qualities v. only 27% doing so against another candidate
- The act of taking the Repeal Pledge outweighed all the obvious work done on ObamaCare by the Attorney General, Jon Bruning (who didn’t take the pledge).
- Over half Republican primary voters said Fischer taking the pledge mattered to them.
- 50% of people who remembered getting our robocall also voted for Fischer! (We are talking hated ROBOCALLS people! Do you know how amazing this is? The thing that we all hang up on in the first 3 seconds and that campaigns normally have to do a ton of to get any impact? This reifies our contention that both the messenger and the message really matter. It also means this cost pennies per vote.)
- By roughly 3:1, the Repeal Pledge mattered more to voters than Sarah Palin’s endorsement (though I think her endorsement was huge in that it legitimized Fischer’s candidacy and helped move Fischer to “contender” status, which signing the Pledge alone would not have done).
- The Repeal Pledge works with self-described moderates as well as with self-described conservatives.
The Repeal Pledge isn’t just good policy, it’s good politics. We knew that by gut feel before (hey, when strangers track down the doctor who recorded your call to thank them for making a robocall, you know something unusual is going on). But now we know it by hard data. Can’t say this is true of all pledges, but it is for this one. Take that, you cynics! :)
Published in General
Fingers crossed that GOP candidates will take note of this. This pledge has more teeth because it’s about a specific piece of legislation rather than a general principle (like taxation). The robocall number is stunning. Even if Romney shies away from it, attention to Obamacare repeal by congressional candidates may create a reverse coattail effect.
Just noting that I dislike the idea of pledges, as they are manmade and thus, flawed. There are several signers of Norquist’s pledge that are regretting it and publically announcing that they are going to abandon it, now that they see the pitfalls.
Yours, too, will have pitfalls that may be unforseen. All I’m saying is it wouldn’t bother me too much if supporters later changed their minds, if they had good reasons and explained them.
Nice follow up with hard data. I still think pledge=gimmick and signing one doesn’t make a candidate any more or less conservative. The only thing that will ensure that is an electorate that is engaged and contacting candidates and elected reps proactively rather than waiting on robo calls.
Well done follow up.
Repeal may be good politics, and it is mostly good policy, but one problem: Not gonna happen. See Ben Domenech’s post from a couple of day’s ago. http://ricochet.com/main-feed/Will-Republicans-Repeal-Obamacare
The questions we should be really thinking about are, what aspects make sense, how can we make it better, and how can we move enough Democrats from places like WV move in our direction to amend it. End it is a pipe dream, and the sooner we understand that reality, the sooner independents will no longer see us as mad dogs baying at the moon.
Please allow me to propose your next pledge project:
“We in the political class recognize the unfairness of banning robocalls for the private sector, yet allowing those of us in the political sector to make them without limit. I pledge that neither I nor my supporters will use robocalling or other intrusive contact methods in any of my campaigns.”
I get that a lot, but I insist that I am a realist.
If signing the pledge was so important, why didn’t it do anything for Stenberg? Fischer received 41.1% of the vote, but Stenberg only got 18.8%. I don’t doubt that Obamacare will be the biggest issue in the general election, but it was Bruning’s and Stenberg’s unfavorable ratings that decided the outcome of the primary.
Re. Stenberg signing: he did not sign until about 4pm on Monday afternoon. At that point all the calls had gone out, ditto web advertising which had been going on since Friday. We immediately amended the website to reflect that there were now two signers, and ask people to thank him, and sent out an alert. But it was way too late. I do not have the cross tabs with me, but as I recall none of his voters listed ObamaCare as a main concern. I hypothesize that if you were a voter who didn’t want Bruning, and you cared about ObamaCare, then Fischer signing the pledge was part of the margin that led her to be the non-establishment choice.
One is tempted, given the continuing, perhaps perpetual, cynical and insulting implication of your inquiry, to question why anyone would possibly want to waste $15,000 on polling and even bother with 300 actual voters when we could pocket the money ourselves and over a few beers with some dice have a few laughs just making up the data?
But as I largely gave up sarcasm when I turned 14, and since I know that had you not hit “post” so quickly, with a little reflection and so as not to seem boorish you would have instead phrased it along the lines of “I don’t know much about polling but am curious: why is it ok to use only 300 people and how do you make sure you get a fair sample?”, I shall confirm with the pollster what I believe to be the case and post it when I hear back.
It will be interesting to see how Bob Kerrey will try to spin his past support for national health care and universal mandates into something that will get him more votes than Ben Nelson could have received.
Of course you can get a decent result with 300 people, more is always better, though. The question isn’t so much the number, it is how were they selected. It would be great to know whatever weighting was done as well. Of course it’s more convenient for you to rephrase my concern in a manner that is more easy for you to deal with.
double post
Heather, let me respond to your post in the most direct fashion possible.
When informed that one candidate in a GOP primary has taken the Repeal Pledge and another hasn’t, I will vote for the former every time.
You see, it really is a “big effing deal!”
Elect No Democrat Anywhere, Ever.
So Heather, how were the 300 voters, a rather small sample, selected? Would Gallup or Rasmussen find your sampling and polling methods sound?
Ok, Cylon. How would you propose proving that you are correct? Are you going to try and tell me that The Lady of the Lake extended Excalibur from her bosom and made you King? ( kudos to Monty Python ). Just being contrary is not an argument.
Per pollster: “Statistically speaking, when dealing with any population above 50,000 the margin of error remains the same based on the number of interviews. Whether you dealing with a population of 50,000 or 5 million, 300 interviews will always yield the same margin of error (+/- 5.65%). Of course, if you increase the sample size, say to 500, you reduce the margin of error. But only by about 1%. Most professional pollsters feel very comfortable using a universe of 300 when dealing with a universe of primary voters.
“In Nebraska, we took a proportionate representation of voter turnout in the actual primary election from the day before by county, grouped the counties into Media markets and set a gender quota of 50/50 by media market and large counties, then dialed off a list of those voters who voted in at least one Republican primary election over the past few years. In a survey where we are required to speak to those who actually voted, we needed closer to 50 numbers (v. usual 25-30) per complete.
“We only weight if a number (e.g. gender) is way off. In these numbers no weighting was needed.”
Thanks for the follow up from your pollster. I’m satisfied your sample was well selected. I have a few more questions if you’d indulge my cynical and boorish curiosity, though.
Your press release says that “of those that recall receiving” a robo-call Fischer got 50 percent of the vote. The important question left unanswered, though, is how many of the 300 remembered getting a robo-call?
You also claim that Fischer won a plurality of those who remember seeing internet ads about the pledge. How many of the 300 remembered seeing the online ads?
You cite the pledge being more important than the Palin endorsement among respondents by 35% – 14%. But there were obviously other options to that question, seeing as those percentages account for less than half the respondents. What exactly was the question asked and what were the other choices?
(cont.)
Furthermore, just because more respondents said the pledge was more important than the Palin endorsement, that doesn’t necessarily mean that the pledge was a particularly meaningful factor in the voters decision. The pledge and the Palin endorsment could have been 19th and 20th in order of priority among the list of things voters considered. Saying the pledge was more important than the Palin endorsement isn’t necessarily any more meaningful than someone saying that the entree special at the restaurant where their spouse proposed was more important in their response to the proposal than what the weather was like that night. Chances are, neither factor was necessarily very important in their response, but if asked which mattered more, one might say the entree had a bigger impact.
(cont.)
So, again, a sincere thank you for following up with the information about your polling sample. But I’m still not buying that the self-authored press release Politico posted really tells us anything worthwhile about how much impact your pledge made in electing Mourdock and Fischer.
You can call me cynical and a boor or whatever other name you want, but you still haven’t established that you’ve done anything but spun some cherry-picked data. You were “tempted” to dismiss my skepticism (which is the proper term for my resistance to your horn-tooting, not cynicism) by asking me why your group would waste the money on such a poll. Well, I may not be a polling expert, but I do know something about marketing and PR, and I know having some pseudo-scientific “data” to sell the efficacy of your efforts is great material throw into a case study and fund raising pitch. Besides, I’m sure your all still had a little cash left over for your beers even after spending the 15k, didn’t you.
All right, I admit I was a cynic… Mr Obama promised to end cynicism, and he failed – what can I say?
Exactly per some of the responses above. Don’t fix it, get rid of it. It is an albatross. Lets hope the professional Republicans can figure that out.