Polling Project: Second Release

Polling Project: Methodological Statement

Polling Project: First Release

Now we are starting to get into the good stuff. The first release focused on the first five questions of the poll, mainly in order to demonstrate that the poll was a solid, accurate sample of national public opinion that was in line with the majority of other recent public polls. This release will look at the results to questions 6-9. While the really, really juicy stuff will come out in three separate releases on Monday, these questions should all tide over your appetite until that time. All of them reveal information that you won't find in any other public poll.

These four questions are the terrorism and "fear" questions.  None of these are groundbreaking in and of themselves, as several other polling outfits have asked similar questions. However, I have always been frustrated that no polling outfit who asks these questions ever releases cross tabs for these questions, specifically party ID cross tabs. Who is more worried about terrorist attacks taking place, Democrats or Republicans? Who is more concerned that we haven't captured Osama Bin Laden, Democrats or Republicans? Who feels more safe and secure since 9-11, Democrats or Republicans? I really want to know this stuff, because I think knowing whether or not there is a difference between partisans on these topics would inject an interesting new dynamic into several relevant debates. As such, I have looked for partisan ID cross tabs for a long time within the releases of many polling outfits, but I have never found them provided for these sorts of questions. I don't know why. Maybe I just haven't looked hard enough. Maybe it just never occurred to other pollsters. Maybe pollsters were worried they would be "politicizing" fear and terror by releasing such cross-tabs, even though Republicans have now run two entire elections our of politicizing fear and terror. Who knows.

Well, whatever the reason, I am happy to be able to present these numbers to you today:


Q6. How would you rate the federal government's efforts in protecting America since 9/11?



                     All         Dems      Reps      Ind/Other/No Party
Excellent     13.8%      3.2%    28.1%            10.7%
Good           40.1%     29.8%    51.3%           39.6%
Fair             29.2%     40.1%    14.7%            32.1%
Poor            11.3%     18.4%      4.6%            11.0%
Very Poor     5.5%       8.4%     1.3%               6.6%

Wow. Even though the margin of error on these sub-samples is pretty large, Democrats and Republicans obviously give the government very different ratings on protecting America since 9-11. Non-partisans seems to fall roughly in between the Democratic and Republican positions.

Q7. How would you characterize your own feelings of safety and security since 9/11?

(1-5 scale. 1 = Feel a lot safer and more secure now; 5 = A lot less safe and less secure now)



            All      Dems      Reps      Ind/Other/No Party
1         23.7      11.6        34.4                25.4
2         19.3      13.5        26.6                18.3
3         33.1      39.8        23.0                35.8
4         12.2      17.9          9.2                  9.8
5         11.5      17.2          6.9                10.7

Again, a large difference. Democrats feel a lot less safe and secure since 9-11 than Republicans. Interesting.

Q8. How worried are you that there will be another terrorist attack on the United States in the next year?

(1-5 scale. 1 = Not worried at all; 5 = Extremely worried)



            All      Dems      Reps      Ind/Other/No Party
1         18.5      14.4        22.0                 19.1
2         19.0      14.7        22.0                 20.2
3         28.8      27.9        27.3                 30.8
4         17.5      21.6        16.3                 15.0
5         16.2      21.3        12.3                 15.0

Again, Democrats are more worried, but the margin here is lower than it is on other questions.

Q9. How worried are you that the United States has not captured Osama bin Laden?

(1-5 scale. 1 = Not worried at all; 5 = Extremely worried)


            All      Dems      Reps      Ind/Other/No Party
1          23.9     17.1        31.9               23.0
2          15.6     13.0        18.9               15.1
3          24.0     21.0        24.9               25.8
4          14.0     14.9        11.6               15.1
5          22.6     34.0        12.6               21.1

Wow. A majority of Republicans in this survey are not very worried about capturing Osama Bin Laden. Only a quarter are very or extremely worried. The difference here between Democrats and Republicans is gaping.

There are lots of other cross tabs to these questions, and I can't possibly post them all right now. However, they will all be available within the next few days. I don't want to throw too much of my own analysis onto these numbers, as I think they are quite provocative in their own right. These are good, accurate numbers, which I think add a new, interesting element to the national debate on terrorism and national security. Besides, you guys paid for these numbers. I'd like to hear what you have to say first.

Tags: Blogosphere, Democrats, polls, Republicans, terrorism (all tags)

Comments

35 Comments

The Filibuster

should be a front page diary.  Whatever the outcome this is a serious political action with potentially significant ramifications for the Dems.  You are curiously silent.....

by aiko 2006-01-28 09:59AM | 0 recs
Re: The Filibuster

Oh whatever. I've been calling for a filibuster of Alito for weeks. But you may have noticed that the community jsut spend $17K on oin a poll and I'd like to show the the results.

Serisouly, if you weren't around for when I was lvie-blggoing the hearings from the Senate Hart building, and you don't remember what I said then, well just ram it.

by Chris Bowers 2006-01-28 10:14AM | 0 recs
Re: The Filibuster

Nobody doubts your support for a filibuster. Nobody. It just needs some airtime today, tomorrow and Monday.  All is not yet lost and we don't need to act like it is.

by aiko 2006-01-28 10:38AM | 0 recs
Re: Polling Project: Second Release

OK, this is interesting. While Republicans feel safer than Democrats, and while they rate the government's efforts to protect the country since 9-11 much higher than Democrats, they are still more worried about a terrorist attack occurring in America over the next year. Someone is going to have to explain that one to me. If you feel safer, and think the government is doing a good job protecting you, should you be less worried about a terrorist attack (or vice versa?).

You may want to look at that one again (I had to look a second time myself). The Repubs are skewing lower on the worry scale than the Dems. Are you reading the results backwards?

by zAmboni 2006-01-28 10:08AM | 0 recs
Re: Polling Project: Second Release

Fixed. Thanks!

by Chris Bowers 2006-01-28 10:14AM | 0 recs
Re: Polling Project: Second Release

Thanx, for being the first to point this out. I kept reading it over and over. I think your conclusions as to question 8 should be reversed.

by murrman32 2006-01-28 10:44AM | 0 recs
People lie

Kevin Drum remarked on this a little while ago.  He noted a poll where one of the questions was whether you had voted in the last election--and 91 percent answered yes. This was regarding an election in which there was 70 percent turnout.

Taking Q9 as an example, if a democrat were in office, republicans would indicate much more concern about bin Laden's status.  

People know what polls are used for, and try to influence them in their answers.  That's what makes polling analysis an art. What a goodly fraction of these republicans are really saying is "I support the President." over and over again.

by jayackroyd 2006-01-28 10:24AM | 0 recs
One more thing

What's interesting about Q8 is it's not clear which way Bush would want you to answer. (This is a difficult line for him to walk--keeping people scared while declaring progress.   And it makes you wonder, sometimes, whether they aren't trying all that hard to get him, just as they left Zarkawi in place.)

The margins tighten because these are more honest answers.

by jayackroyd 2006-01-28 10:29AM | 0 recs
Re: People lie

a poll where one of the questions was whether you had voted in the last election--and 91 percent answered yes. This was regarding an election in which there was 70 percent turnout

Yeah, people lie.

Voter registration is a public record.

You can select a sample using 1 + dialing (that is, using a random sample of pre-existing listed phone numbers in each prefix [and area code] and adding one to the final digit - this allows you to get unlisted phone numbers). You have to know the percentage of business lines and fax numbers with this method along with a good estimate of your refusal rate or you'll add significant error to your results when you have to go back to add to the sample (if that's the case, you might as well start over). You have to screen for voter registration (people lie about things they know they should do, like vote, but don't bother to do anyway).

If you select from voter rolls and try to match the phone numbers (there are available commercial databases, or you can do it the old fashioned way - look 'em up) you won't get the unlisted phone numbers. Guess which demographic has unlisted phone numbers?

Cell phones are a problem. Specifically, households with cell phones and no land line. And those cell phone area codes no longer guarantee a geographic area. Guess which demographics fit this description?

In Missouri if the sample is selected from voter rolls you have to try and match phone numbers since the voter's phone number is not part of the statuatory database. I've never seen a commercial match rate better than 60%. If you do the match the old fashioned way you can boost the rate to 75%, though this process is tedious and labor intensive. You still miss out on 25% of the registered voters.

The great part about the Missouri voter file is that it includes voter history, so, if you're screening for likely voters you can ask them if they voted in the previous election (except for those who are newly registered and have no previous voter history) and can then tell if they're lying to you.

So, depending on your sample selection method you've got to come up with percentages of possible good and bad contacts so you can calculate your sample size accordingly. Experience and research within opinion research firms give them some of these proprietary numbers.

Refusals are another problem. The hang-ups and outright refusals in high enough numbers can screw a sample. An experienced interviewer can call back a refusal on another day and possibly convert 25% of them back to responding. It ain't easy.

There are other methods of selecting samples, some  used depending on the geographic scope of the poll.  

by Michael Bersin 2006-01-28 11:17AM | 0 recs
Re: People lie

Absolutely, I know if a pollster called me I would answers the questions primarily (whether consciously or not) to look Democrats look better.

by The Democratic Instant Message 2006-01-29 02:13AM | 0 recs
Re: People lie

Certainly in this poll I would be unsure whether to answer I feel less safe because I think this would be damning to Bush or I don't have any fear of anything because I want to convey I am not falling for Bush's fear-mongering.  In truth I do not fear an attack, but I do expect an attack in the future.

by Rob C 2006-01-31 04:32AM | 0 recs
Re: Polling Project: Second Release

Very interesting results.  

This just underlines that Rove really knows what he's doing.  The bottom line is that terrorism is fundamentally really more of a Dem issue - terrorists tend to target large cities which are disproportionately Democrat (NY twice, DC, LA, etc..).  Kansas is in fact quite safe!  Making natl security the top issue not only feeds raw meat to the base ideologically, but also speaks directly to the tangible needs of the Dem base.

Dems cannot hide from the natl security issue - it's literally life or death for their constituents!  Any Dem running for office (in '06 or '08) needs to take the fight to the Repubs & make the case that the current policies have made us LESS SAFE and that the dems do have an alternative plan and that finding Bin Laden would be the # 1 priority of the Democrats.

This is what makes the current situation so frustrating for most urban liberals (esp. NYers) - no one hates Bin Laden more than us, but we have to watch Bush botch the struggle against terrorism and then be chastised as a fifth column by TV windbags.  It's completely nuts.

by glassonion 2006-01-28 10:28AM | 0 recs
NYers

Nobody seems to have connected those two dots--we were the ones who were hit, and we wanted this guy gone by a huge margin.

by jayackroyd 2006-01-28 10:34AM | 0 recs
Re: Polling Project: Second Release

Excellent observation glassonion.

The Dems can't run away from the National Security/Iraq/Terrorism frame that Rove has repeatedly used to his success.

As you correctly point out the Dems have to go on the attack that the Repubs and Bush's incompetence and corruption have actually made us less safe and more threatened. And not getting OBL is a perfect frame.

This needs to be hammered home.

by ab initio 2006-01-28 02:00PM | 0 recs
Re: Polling Project: Second Release

Also a good indication that the Dems need to sell a good plan for the War - on terror and in Iraq..so that plus the evidence of Bushes failures will get us the independent's votes.

by David in Burbank 2006-01-29 09:54AM | 0 recs
Feeling "safer" since 9/11

I'm still trying to parse the question "do you feel safer since 9/11."  When I first read the question, I interpreted it to mean "safer than I felt before 9/11" and then I realized that it probably means "safer than I felt on the day of 9/11."  So with the latter reading in mind, I'm trying to imagine the circumstances where anyone regardless of party would actually feel less safe now than on that day, when we were all imagining bombs blowing up in shopping malls and subways and sports arenas.  Yet 34% of Democrats apparently do feel less safe and secure now, even though a larger percentage of them are city-dwellers.

Am I the only person who is having trouble with this question?  Is it because of my narrow interpretation of "safe and secure" as having to do with terror attacks?  Or is it because I live in a mid-sized southern city and not in a giant coastal one?  Someone help me out please!

And by the way, on 9/11 I was living about 30 miles outside a giant east-coast city and wasn't particularly afraid for my own immediate safety...

by Christopher 2006-01-28 11:49AM | 0 recs
Re: Feeling "safer" since 9/11

I have to agree. I have no idea how I would respond to this question. I don't feel more safe or less safe relative to pre-9/11 times. It's a non-issue for me. A terrorist attack in the middle of suburbia, on me or my family, is of such a low probability as to be laughable.

Terrorists aren't some unstoppable horde of zombies. They're pseudo-militarist idealogues with limited resources and reach. As such, they are going to focus their efforts in places where they have the maximum effect for the minimum amount of effort.

by Geogriffith 2006-01-28 01:42PM | 0 recs
Re: Feeling "safer" since 9/11

I see it as: after the revelation of how unsafe we actually were all along, do feel the govenemnt has made it harder or easier for terorsits to strike again.  Clearly we were vunerable to the 9/11 attack before 9/11 only we didn't realize it.  However, we had a government who was suppose to protect us and they failed -- so the question is, are they doing a better job now then before 9/11?  Have the legislative and administrative changes made by our goverenment made us safer or have they made no real or diffenece or have made things worse (i.e. Iraq now breeding ground for terrorists.)

But I see your point and it's hard say how people will interpret the question -- though I suspect they won't take it as litterally as you did.

by David in Burbank 2006-01-29 10:03AM | 0 recs
Re: Polling Project: Second Release
There's a context to the question. Right before it, we asked respondents to rate the federal government's efforts "in protecting America since (9/11)". There's the context: the time since the attack and not the day of.

So in this question, immediately following the one noted above, respondents rated their own feelings of safety and security since that time, "where 1 means you feel a lot safer and more secure now and 5 means you feel a lot less safe and secure now". Thus, measuring personal safety in the period since the attack is what the question does.

Hope that clears it up for you. Thanks for asking.

by Sun Tzu 2006-01-28 12:09PM | 0 recs
Re: Polling Project: Second Release

Thanks! So the question is not really measuring a direct comparison of feelings of safety, but rather measuring our impression of the direction our security is generally heading?  I.e. pessimism vs. optimism?  I apologize for being dense on this but I am very literal on these things!

by Christopher 2006-01-28 12:32PM | 0 recs
Re: Polling Project: Second Release

Exactly, public perception of the direction  since 9/11: more secure, less secure? Do we feel safer or no? The results are very interesting and I'll post more on this tomorrow. No apology necessary, amigo. It's not your average poll question. Exactly! :)

by Sun Tzu 2006-01-28 01:02PM | 0 recs
Re: Polling Project: Second Release

In the response to Q6, the numbers in the Dem column sum up to less than 90%.  What's up with that?  Perhaps a typo in the 'Fair' category?  (It seems unlikely that the weighted average of 20.1, 14.7 and 32.1 would be as high as 29.2 - you'd have to have considerably more than 90% of the population in the third category for that to have a hope of being true.)

by RickD 2006-01-28 02:40PM | 0 recs
Re: Polling Project: Second Release

I am pretty sure with some random calculations show that the Dem number for "Fair" on question 6 should actually be around 40%.  Add the 20 percent or so that is missing to that and you get about the correct percentage for the "All" category for "Fair".

They may want to look at that.

by zAmboni 2006-01-28 07:44PM | 0 recs
Re: Polling Project: Second Release

It's not your average poll question. Exactly! :)

Yes, this is what I wanted. Usually the big pollsters ask the same darn questions over and over. Though this is good in tracking the shifts in opinion it leaves much to be desired when it comes to the overall picture.

Thanks guy for all your work. I am so enjoying this.

by mpower1952 2006-01-28 02:49PM | 0 recs
Re: Polling Project: Second Release

I still say the Dems best commercial is the one featuring Bush saying about bin Laden "Dead or Alive", followed by his "i'm not really concerned about bin Laden", followed by a montage of terrorist attacks following his "I'm not concerned" remarks....

Rove had made it clear that he wants "National Security" to be the issue, and that he plans on casting dems as "weak" on National Security...

these numbers show that Dems need to, and can, solidify the idea that the White House is not showing the necessary leadership on National Security....

by plukasiak 2006-01-28 03:11PM | 0 recs
Re: Polling Project: Second Release

The obvious problem with featuring the Bush flip-flop on the importance of killing/capturing Bin Laden is that come October it may actually happen. A Rove dream come true.

by grumpo1 2006-01-29 04:41AM | 0 recs
Re: Polling Project: Second Release

Another good reason why Dems ought to be calling on Bush and the Republican majority everyday for a withdraw plan - no matter how bad an idea it might actually be.  If there is constant pressure on Bush and Republicans for a withdraw plan then when the inevitable withdraw begins they will look weak. And if they refuse to layout a plan by election time they will continue to suffer the ire of the families of dead American children sent to Iraq.

It might be good for a Dem (Murtha's) plan to be offered as the Dem strategy as well, so folks know what they're for for as well as what they're voting against -- timed right of course.

by David in Burbank 2006-01-29 10:14AM | 0 recs
Re: Polling Project: Second Release

This should have been done in 2004 - with the clip of the Bush interview where he says "I don't think the 'war on terrorism' can even be won."  

Since no ad aired to frame this message for the Dems leading up to the '04 election, the reaction to bin Laden's pre-election video appearance was, 'time to rally around the President' instead of, 'why hasn't the government caught the terrorist responsible for 9/11 yet?'

by Mr DC 2006-01-29 11:25AM | 0 recs
Re: Polling Project: Second Release

What do these answers tell us?  Dems & Reps both trust their team and don't trust the other, independents are in between?  Dems think the Reps ought to get the rap for not capturing Osama as Bushie said they would, while Reps would like this issue to go away?

Where is the surprise in any of this?

by Good 4 A Merica 2006-01-28 03:34PM | 0 recs
Q6 totals

Why is it that the sum of the Democrats' percentages in Q6 is only 80%? Did a fifth of Democrats refuse to answer the question?

All the other columns sum to 100% (or close enough). Is this odd?

by abw 2006-01-28 04:15PM | 0 recs
Re: Polling Project: Second Release

Pretty much confirms for me something most of us have known for some time now. A significant number of Republicans are inherent cowards. :)

Thanks for clearing that up.

(For those who think I am wrong, fine. :) Just think back to your school days and the school bully. The bully was always the loudest and the most angry, and even violent. But when seriously confronted, they ran. Just like Republicans). How many Republicans in the Administration actually served in an armed conflict? Compared to Democrats? There is a reason the Republicans are called Chickenhawks, and it goes all the way down the line.

by Kryten42 2006-01-28 09:51PM | 0 recs
Re: Polling Project: Second Release

Hey, not seeing any coverage of the poll in more traditional media settings (or whatever we call it).  When that comes in, hope you post the links.

by The lurking ecologist 2006-01-28 10:47PM | 0 recs
Re: Polling Project: Second Release

A more constructive way of putting it...

A lot of Republicans are suffering from cognitive dissonance.

by eehret 2006-01-29 10:25AM | 0 recs
Central page for Polling Project

Is there a central page where all the posts on the Polling Project are gathered?

If not, I would like to suggest one.

If so, I would suggest a prominent link on the main page.

This is a big deal and should be easy to find all in one place, and easy to direct people to.

by scottmaui 2006-01-29 08:12PM | 0 recs
Methodology and presentation of results

You mention "the margin of error on these sub-samples is pretty large" but do not provide the margins of error. That would be useful knowledge and highly relevant for interpreting the data.

You could also clarify the value of these results by providing the n, the subset of N respondents (N = total sample size) for each cell. This is related to the issue above and to another comment in your post, that other polls often don't present such crosstabs. One reason they don't is that margins of error quickly become huge in the cells, and thus few statisticians feel quite comfortable attributing meaning to the numbers. The smaller the number of respondents in any given cell, the more likely that margins of error are swamping any apparent variation across sample subsets. (and vice versa)

by beachrat 2006-01-30 12:47PM | 0 recs

Diaries

Advertise Blogads