« The A Word... Adam L. Silverman, PhD* | Main | The limits of power »

29 April 2014

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

steve

If you look at the polling results by political identification, the vast majority of republicans disapprove, independents disapprove with a smaller majority and a sizeable majority of democrats approve.

I think it's more an issue of political branding more than anything else--just like ford guys v. chevy guys.

I believe it took the last 2 years of the Bush administration for Bush's republican approval ratings to slip below 60%, so perhaps that will be Obama's future, but it seems to take quite a bit for political realism to trump branding, imho.

Charles I

Even without looking at the research one must assume a self selection bias acting on willing respondents, just as there is for those who still give up their seats or hold the door open for little old ladies without the likes of me barking at them as I used to be so wont to do.

You can enroll yourself in various web polling panels and be emailed invites for general or self-selected category polls.

Add in that we're not rational actors, its a complete muddle no matter how apparently reflective or predictive.

William Herschel

It is striking that Obama gets his lowest grades on Ukraine. What in God's name do the people who responded to this poll want him to do? Do they want US troops to attack... somebody... in Ukraine?

Following on, I believe I have finally understood what is going on. This interview with a Russian Deputy Foreign Minister I think captures the situation succinctly:

http://rt.com/news/155572-ryabkov-ukraine-south-crimea/

"Russia is not going to repeat “The Crimean scenario” in southeastern Ukraine, said the country’s Deputy Foreign Minister Sergey Ryabkov.

“I think what happened in Crimea turned out to be a big shock for Western partners. They can’t tolerate it. And they see the same scenario in southeastern Ukraine,” he told Gazeta.ru."

It was all about Crimea. Had Crimea remained a part of Ukraine, NATO would have ultimately controlled Sevastopol. I leave it to the more strategically knowledgeable to state how big a coup that would have been and how enraged the people behind it were when, without firing a shot, Putin put a damper on their plans.

nick b

Col.,

Your comment is interesting. Are there enough accomplished people in this country to change the results of an opinion poll? I would imagine 'accomplished' to be a mixture of level of education and income. Do you have a different metric?
In polling, I find the big tells to be sample size, wording of the survey, live interviewers vs. robocallers and lastly inclusion of both cell phone and land lines. Polls are simply a snapshot of a moment in time, and are useful in that regard. I think the media hype them to be much more than they actually are. Sometimes such hype can lead to a self fulfilling prophecy.

turcopolier

nick b

I would include substantial life achievements in the mix. That would exclude a lot of ancient TV watchers like me and others who are available during the work day because they have nothing to do. pl

nick b

I get a fair amount of polling calls at the house. I am quick to answer because I find such things of interest. Usually the callers are more interested in speaking with my wife than me (she is not interested). However, the calls seem to come after business hours.
Out of curiosity, would you think accomplished people in this country make up a large enough demographic to make a difference in a public opinion poll? My guess would be no, but that's just speculation on my part.

turcopolier

nick b
"the calls seem to come after business hours." Are you there during business hours? Are there enough "accomplished people" in America? I think so. pl

nick b

Yes, I am here more often than I would like. Merely curious on the question of accomplishment. I have a good amount of education and had an interesting career and experience, but tend to consider myself much more ordinary than accomplished. Having read much you have written, I find you to be quite exceptional, and that people of your caliber are in short supply. Hence, my question.

Bandolero

Turcopolier

I think such polls have yet another more important issue: slightly different questions may bring big differences in results expressed in percentages.

It has several implications:

1. For those seeking to influence political decisions of others with polls they may ask the questions in a way that results would be most affirmative of their intentions (if answers are not satisfying to affirm the intended aim, poll results don't need to be published and a new poll can be ordered with slightly different questions)

2. For those who make political decisions they shall order their own polls from institutes they trust which ask similar questions with a slightly different spin to understand more about the opinions of the respondents

Just see the polls regarding Ukraine as an example. As I understand them, most Americans want:

- a tougher policy against Russia
- more US sanctions against Russia
- no US military aid for the Kiev government in any form

Of course, that policy, for which poll majorities seem to exist, may not make a lot of sense, because US sanctions against Russia will likely not change anything on the ground in Ukraine.

Contradictions happen also quite often in polls. So, as I understand current US polls, US people care a lot about improving US economy, while Ukraine is not important to most of Americans. So, if the US sanctions Russia for Ukraine, most US people find it good in regard to "helping Ukraine", but when they hurt the US economy, too, which sanctions always do, than most US people find the policy bad.

walrus

Col. Lang is right. The problem with these polls is self selection which skews the results.. The majority of non WaPo readers may have very different views.

Furthermore, the binary nature of the framing of the questions is indicative of the Elites views of what should concern the voters and betrays an incredibly shallow worldview.

ToivoS

WH you raise a point that impressed me. What was striking about the Crimea takeover was that it was done without firing a shot. Of course the Ukrainian forces were totally demoralized with more than half of them ready to join the Russians. Without doubt it was an operation that was one part military and one part political. This must have involved some extremely disciplined and well led troops. When it first started to break out I found it difficult to see how it could end without some significant bloodshed. After all the Ukrainian troops were still under command of their officers and they were trained to obey orders.

I have no military background but I would be interested in hearing from someone who has led troops and is interested in questions like this if that bloodless takeover was really as impressive as it struck me?

nick b

Walrus,

"The majority of non WaPo readers may have very different views."
Not sure how that would effect the survey. I don't think the poll was soley of WaPo readers. While not easy to find, the poll claimed it was a survey of 1000 people nationally. It has a margin of error of +/- 3.5%. These samples are usually created through random digit dialing, though I could not find anything published by WaPo/ABC or Abt-SRBI (the survey firm) that showed their methodology. It would've been interesting to know how many calls they had to make to get 1000 interviews.

Fred

nick,

It is the wording of the questions to get the answers desired that is also a major issue in polling.

rick

I would just add #3- people lie to pollsters. If you ask a question that has a plainly moral answer and a plainly immoral answer, it takes quite a person to be honestly immoral with a stranger. My favorite examples are health care and alternative sources of energy; when asked, huge majorities say oh yes, we will pay more so the u insured can be covered, and that they are willing to pay more I gas taxes to support AE research. 75% comes to mind for both.

Nobody wants to say out loud, screw the poor, I got mine.

Miraculosly, those majorities have not compelled action. It's almost like respondents say what they think the "correct" answer is, regardless of their actual feelings.

nick b

Fred,

I agree. I mentioned what I thought were the big tells in polling in my first comment. I read this blurb about some of the pitfalls of phone polling on the site of the firm that does the field work for Rasmussen. And it's just a slice of the problems with polling:
"different segments of the population answer the phone in different ways. For example, women answer the phone more than men, older people are home more and answer more than younger people, and rural residents typically answer the phone more frequently than urban residents."

http://www.pulseopinionresearch.com/Methodology

Paul Escobar

To all,

32% of those polled "think of themselves" as Democrats - while only 21% as Republicans.

I doubt there is actually an 11 point spread between Democrat-minded & Republican-minded folks in the United States.

IMO, this spread colours the responses.

Am I wrong?

walrus

Yes Nick, but how many when they picked up the phone said "to hell with it" and slammed it down?

nick b

Walrus,

In my reply to you above, I wondered the same thing.

nick b

Paul,

Unfortunately to answer the question, you have to rely on yet another poll. Here's one from Gallup earlier this year. They put the spread at 6 pts.

http://www.gallup.com/poll/166763/record-high-americans-identify-independents.aspx

Here's another from Pew in 2012. They have the spread at 12 pts.

http://www.people-press.org/2012/06/04/section-9-trends-in-party-affiliation/

Both polls show trends in the spread over a number of years. Perhaps this is helpful. Take it for what it's worth.

kao_hsien_chih

Records of nonresponses are kept in the raw data, if one cares to look at the real numbers. It might be theoretically possible to do serious analyses with that data since there is some real information to be had about non responders (say, based on their phone numbers and associated geographic areas--although this is harder in the era of cell phones.)

Generally, the nuts and bolts side of survey operations are much more careful and rigorous than people think and, more than that, we know our limits. The same can't be said for people who use the information we collect, though.

kao_hsien_chih

A lot of very recent polling data (beyond summaries which are the only parts most people see) are kept as proprietary information, but older data, including fairly recent ones, are released via certain academic instititions (eg Roper Center at Univ of CT). Not sure how the public can access them, though.

Generally, there is a lot of research on a lot of technical factors and their effects on surveys (they are usually much more interesting than the results themselves because they get at the question of what polling results really mean and what they don't) and that is the real social "science." Not many nonspecialists care about such things since they are not "relevant" enough or whatever....

nick b

KHC,

Thank you. It has been twenty plus years since I have been personally involved in polling/surveys. I recall we spent a lot of time and money to be sure we could be as accurate/unbiased as possible. However, our polling was not for public consumption. While no longer involved, I still find methodology very interesting, often more so than the survey results.

Charles Dekle

Col Lang,
Since I retired, I have beenI telling people that I am a gentleman of leisure. It makes me feel better and puzzles the questioner.
Regards,

kao_hsien_chih

Thanks for kind words! How people try to make sense of polling data (a little bit like intelligence of other kinds, based on the little that I know of that realm) is always something I found fascinating. The real "information" contained in polls is that certain people responded certain way to certain questions and that we can draw some inferences by using statistics, with a lot of caveats about what it is that we really know and what we don't--which gets into technical details. Nobody wants to know the technical details, it seems, but are eager to jump to grandiose conclusions that are pretty tangential to what the polls "really" say. True with a lot of other things too--unless you know the "details," things don't make much sense, but that doesn't stop a lot of "important people" from loudly grandstanding, or so it seems.

Ryan Murphy

This is what I don't like about this poll:

"Q: Do you approve or disapprove of the way Obama is handling the situation involving Russia and Ukraine?"

In the detailed answer for all adults it states that 34% approve of Obama's handing of the situation, 17% strongly, 17% somewhat.

For the breakdown of those who don't like his handling of the mess it shows 46%, 14% somewhat and 31% strongly.

For registerd voters the difference is a one percent addition to the above figures.

I find this question subject of all sorts of different interpretations. For example, I disapprove of Obama's handling because as an anti-imperialist nationalist I believe he is still meddling. If he wasn't he would tell Kerry to stop running his mouth. Kerry is making a fool out of himself and is an embarrassment. Taken at face value it would appear that I am in league with people like Krauthammer and others who are don't believe Obama is doing enough and want to try some sort of stupid military option. Nothing could be further from the truth.

On the other hand I wouldn't be surprise if a number of people who do approve of Obama's handling are under the mistaken belief that he is trying to avoid getting involved.

In short, this question is meaningless.

The comments to this entry are closed.

My Photo

August 2020

Sun Mon Tue Wed Thu Fri Sat
            1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          
Blog powered by Typepad