Sunday, March 05, 2006

"... then you turn left, no, I mean right ..." or Alice Meets the Red Queen 

Like many others occupying a more conservative position and wearing insignia on their collars, I greeted the now infamous Zogby poll of troops in Iraq with open skepticism. For me, too, it just didn't sound right. Without rehashing who paid for the poll or the "trust me" sampling methodology used, I'd like to point out two small observations I had that I haven't seen talked about elsewhere.

First, from Mystery Pollster:
According to the procedure Zogby described, respondents were intercepted randomly (e.g. they were not self selected) at multiple locations throughout Iraq (e.g. not just in the so-called "Green Zone") and interviewed using a paper questionnaire that they filled out with the assistance of an interviewer. (emphasis added)
I wonder how this "assistance" manifested itself? I'm not accusing the interviewers of altering the responses, but emphasis or clarification provided by individual interviewers could certainly affect one's answers.

Second, let's look at the response to just two questions:
Please rate the statements ... as reasons for the Iraq invasion:

8. To remove weapons of mass destruction (WMD) from Iraq

  1. Not a reason 63%
  2. Minor reason 29.3%
  3. Major reason 3.2%
  4. Main reason 1.1%
12. To retaliate for Saddam's role in the 9/11 attacks
  1. Not a reason 2.7%
  2. Minor reason 7.5%
  3. Major reason 50.2%
  4. Main reason 35.3%
From the first we can conclude that over 90% of service members in Iraq don't think the WMDs were that big a factor in us invading. This is despite all the "Bush lied about WMD" and "where are the WMD" stories and talk. Turning to the second, around 85% seem to think Saddam had something to do with 9/11, even though this was never a big news story and even the denial of such a link was often reported. This seems to either defy all logic or represent one of the most ill-informed soldiery of all time. However, if we ask "could the respondents and/or interviewers have been confused on what a '1' and a '4' mean?" these responses just might make more sense. But how would a possible misunderstanding manifest itself with the other questions?

If we conjecture a confusion on the meaning of numerical responses, I think that confusion would be restricted to questions 8 through 14 (reasons for invasion). After all, while it is somewhat logical to rank your number one reason as a '1', questions 17 through 24 are general "how much do you agree with X" and a higher number would be more natural to express greater agreement. Also, while it may be possible for confusion to be limited to a few questions, the similar format and structure would argue for an "all or nothing" situation. So, if we take a step through the looking glass with the expressed reasons we now find:Do these flipped results make more sense? In many ways yes, but it also produces a few confusing results. For example, allowing for my reversal we are left trying to understand how a majority might think we invaded to establish a democracy and yet the same majority didn't think we invaded to remove Saddam from power. In this case, the results as reported make more logical sense (i.e. we invaded to kick Saddam out but not necessarily to establish a democracy). Also the oil and bases questions more closely correlate to what I've been reading and hearing when taken in their original form.

Those problems identified with my reversi world not withstanding, though, I find myself unable to completely accept the results as reported for one basic reason. I do believe there is any way to logically look at the overwhelming number of service members in Iraq who, based upon question #8, apparently believe that WMDs had little if anything to do with invading Iraq and come to any other conclusion than there is something wrong in either how the survey was conducted or how the questions were answered. I doubt you'd find any other segment of the U.S. population, apart from the most virulent imperialist/oil-crony loving/Jew-filled conspiracy proponents, that would even approach these sorts of numbers on this one question. I'm just surprised I haven't seen anyone else seeing the same discontinuity.

This page is powered by Blogger. Isn't yours?