2* Did the United States torture terrorism suspects?The 58% No to question 3 is isn't so impressive if you care about informed opinions, because there are 37% above who still don't think the U.S. tortured, making their perspective on the justice of prosecution next-to-useless. If we can mostly subtract the 37% crazies from that 58%, that could leave us with as low as 21% who realize there was torture but think it shouldn't be investigated. That gives us roughly as many for as against prosecution. (need to see poll's crosstabs for a more accurate figure).
42% Yes
37% No [what universe do these people live in?]
21% Not sure
3*Should the Obama administration do more investigating to find out how the Bush administration treated terrorism suspects?
28% Yes
58% No
13% Not sure
4* Does the release of CIA memos on interrogation techniques help the image of the United States abroad or does it endanger the national security of the United States?
28% Helps the image of the United States abroad
58% Endangers the national security of the United States
14% Not sure
The 58% Nos to question 4 are also sadly misinformed, as Rahm explained. Plenty of information was already available about the techniques used. New information provided in the memos is:
a) Quantified extent of use (waterboard counts and time limits, for instance)
b) The embarrassingly bad legal reasoning
Update: In the comments, Metavirus points points out some of the problems with Rasmussen's polling, especially the false binary choice of question #4.
Thinking about this, I realize question #3 is also poorly worded. It's pitting the Bush administration with the Obama administration, when it should be pitting the Bush administration with justice.
The Attorney General's decision to investigate should be independent, which also makes polling irrelevant to it. 99% of Americans could be for or against investigating the Bush administration, but either way wouldn't change the legal analysis.
But it's still interesting to get a picture of how informed or biased Americans are. Here's Gallup's last poll from several months ago:
Better-worded questions and methology there, for sure, with a new update coming out next week.
Bottom line, for me: Only pay attention to Rasmussen election-tracking polls, which are much more valuable since question-wording isn't a problem.
for the life of me, i don't understand how anyone takes rasmussen seriously. their performance during the 2008 election was ABYSMAL and their results were consistently far-flung outliers. it's a real shame that conservative-run polling outfits like Rasmussen are painted with the veneer of respectability. here's a post i remember from ambinder during the campaign: http://marcambinder.theatlantic.com/archives/2007/10/scott_rasmussen_the_elites_ver.php . And a more recent analysis from Yglesias: http://yglesias.thinkprogress.org/archives/2009/02/scott_rasmussens_conservative_friendly_question_wording.php
ReplyDeleteIt will be interesting to see how wrong Rasumussen was once the new Gallup poll comes out next week: http://www.huffingtonpost.com/2009/04/24/gallup-to-release-poll-on_n_190817.html
ReplyDeleteIt would be pretty shocking if the public sentiment represented in Gallup last poll on the topic back in February (which showed 2/3s of americans favoring investigations) would somehow have shifted AGAINST investigations in a month's time. Here was the earlier results: "While no more than 41% of Americans favor a criminal investigation into any of the matters, at least 6 in 10 say there should be either a criminal investigation or an independent probe into all three. This includes 62% who favor some type of investigation into the possible use of torture when interrogating terrorism suspects, 63% who do so with respect to the possible use of telephone wiretaps without obtaining a warrant, and 71% who support investigating possible attempts to use the Justice Department for political purposes." http://www.gallup.com/poll/114580/No-Mandate-Criminal-Probes-Bush-Administration.aspx
Oh, and by the way, how about the sneaky phrasing of this question:
ReplyDelete"Does the release of CIA memos on interrogation techniques help the image of the United States abroad or does it endanger the national security of the United States?"
Pretty stark binary choice, isn't it? Think about that for a second -- who is going to say that the release of the torture memos HELPED the image of the US? It's a stain on all of us. So the only other choice is that the release endangered our security? I call bullshit and shenanigans.
"there are 37% above who still don't think the U.S. tortured, making their perspective on the justice of prosecution next-to-useless."
ReplyDeleteOpinions on what constitutes torture vary wildly. Their overly narrow definition of torture is no less useful than the overly broad definition of others.
As for Rasmussen...Rasmussen is ranked as one of the most accurate pollsters at 538. And it was exactly on target for the popular vote breakdown in the presidential election.
Everyone always attacks poll questions if they don't like the results. I do it myself.
Coming up with better poll questions is a bloggy cottage industry, but in this case Gallup's questions were unquestionably better for getting nonpoliticized answers.
ReplyDeleteI've always liked Rasmussen's election tracking polls, where question-wording isn't a problem. I should have remembered to ignore Scott's non-election polling.