Does interviewer 'traits' have an impact on polls?

The most clear cut answer to this question is yes. As talked about in class today, race, gender, age and socioeconomic status of an interviewer can have an impact on poll responses. In addition, body language and political affiliation (if known) can also have an impact. For example, although not a necessarily a 'trait' the way that CNN and Fox news word their results for their polls are quite different. CNN headlines a poll saying "Generational Split Over Healthcare Reform," whereas, Fox news disreagards the age difference and headlines a poll saying, "Americans Prefer the Current System to Obama's Healthcare System." Both media outlets who are polling the public are dictating the poll to slide it in a persuasive direction that would appeal or is likely to agree with a majority of its viewers.

However, in addittion, these polls could be skewed by the way that they were conducted. The Fox news poll was conducted from a population of registered voters via telephone. The CNN poll was also conducted by telephone. Therefore, is this poll representative? As discussed in class, this may be a disproportionate population representation because some individuals are not registered voters (Fox news poll) and some may not have telephones (both polls). Although one can appreciate the efficiency of polling via telephone. It may be quite easy to argue that these polls hardly represent and equal distribution of the population.