Tuesday, February 19, 2013
Non Attitudes
Nonattitudes are problematic for pollsters if not adequately measured and discarded. When a qualifying question is added to the beginning of the survey it can help to clarify actual v. nonattitudes; however, as Asher points out in his Ohioan land use survey example it is not always an effective method for disqualifying respondents. If at the beginning of the survey which Asher presents the interviewer had asked, "Sir, tell me what comes to your mind when you hear the phrase land use AND can you give me an example?" we might have been left with a less interesting anecdote for our reading; yet, we might be given a more informed survey about actual land use issues. In the example Asher is asked for an example of a land use problem that might be salient to he and constituents at this time. If Asher had been allowed to say "I don't know" or "I am not informed enough" we might again be left with an uninteresting story; however, even this proponent of polling is sucked in to a fairly dichotomous decision making process with which he is "relatively unfamiliar." Does this example give us faith in the polling system or only inspire change for future conduction of polling? (It is completely acceptable to say, "I don't know" or "I am not informed enough about this issue" in the comments section of this blog post)
Subscribe to:
Post Comments (Atom)
I think it inspires change. As we saw later in the chapter, he started to provide instances of surveys suggesting noncommittal options, like you cleverly did in your post.
ReplyDeleteThe problem is if you provide the "I don't know" choice, the result will still be relatively ambiguous. "I don't know" can mean a multitude of different things ranging from "I am insecure about having an opinion" to "I feel strongly both ways" to "I don't care". We are still grouping together a multitude of different attitudes, non or otherwise, because people are interpreting the phrase "I don't know" differently. Operational definitions would be needed to make this more objective.
I agree that polling does not account for as many things as it should, but I think the best way to poll (aka get quick data) is to not complicate the process with a wide variety of options. One thing we failed to mention in class is the time required to conduct a survey of opinion, as in your boss wants it yesterday, and if you start giving people too many options they will not complete a survey in a quick fashion. I think the cure all to this problem is to provide something other than a direct answer which means that for the individual it is more complicated than yes or no. So I would say that the provision of such an option provides a little more faith in the interpretation of polling.
ReplyDeleteLove both of these comments! Kathleen I think you nailed Asher's description of the issues with "I don't know" answers in Chapter two. Kevin! I was thinking the exact same thing! When I am asked if I can take a survey I almost automatically gauge how much free time I have right then. Allowing the interviewee to say "This issue is more complicated for me" seems like a great idea. Thank you both for your input.
ReplyDelete