Two years ago, if you had told me that 70% of people interviewed in Ireland thought that drinking Guiness was good for your health, I would have actually considered it without thinking about the methodology used to gather the relevant data. These days, many people look at polls and take them as fact when they should be thinking about the data that the media and other institutions are presenting to them in various mediums. Douglas Walton in Media Argumentation: Dialect, Persuasion confirms this point:
But activists, the media, and private agencies can and often use “mutant statistics” as tactics to manipulate public opinion. They can and often convince people that even the most implausible claims are true by twisting the question wording in a poll (Best 2001, p.4).
So how are we to trust polls? How are we to discern what is true, and what is false? I think that question is best answered by each individual person. Polls are great as a public discussion tool but they should not be the basis of any voter making a judgment about a political candidate. Take the example of the current CNN Poll conducted over the telephone with 435 registered Democrats. The poll suggests that the majority of Democrats want Senator Clinton to be on the ticket, but whether that represents the larger populations is an unanswered question.
Telephone interviewing is quick and fast and allows organizations like CNN and Gallup who have the money to crank out these research reports fairly quickly in order to maximize exposure to the public and the general media. There are some problems with telephone interviews as it relates to political research:
- You cannot tell if someone is lying, telling the truth or just plain joking.
- Depending on the length of the telephone interview, the respondent might rush his or her answers if they feel that they are pressed for time.
- Telephone interviews are often not planned so the interviewer might catch the interviewee at an inappropriate moment and receive negative feedback as a result.
- The respondent cannot research the topic or even have time to formulate a response that he or she feels is appropriate therefore the interviewer might receive data that does not truly represent the interviewee’s opinion.
- Lastly the person that conducts the interview is critical to adequate data being gathered because no one likes to talk to someone who speaks in a monotone for example.
Having said all of that, I have to say I like the Gallup data because they also include Independents which are a going to be a significant voting segment come the general election. 50% of people interviewed in the Gallup Poll want Senator Obama to select Senator Clinton as his running mate. I don’t find this very surprising given how popular Senator Clinton is but despite all these numbers, I am still not sure whether the Obama campaign could pull of selecting Senator Clinton as their VP.
Ideally I would like to see some comprehensive nationwide polls indicating how strong some Obama supporters feel about that possibility? If the significant majority do not care, than I believe, even if Senator Clinton gives Senator Obama a minor boost, it might be worth it. Don’t quote me on this as this is not my final opinion, I will wait and see what the next month reveals.