Saturday, October 04, 2014

Polling: The Battle Against Rationality Continues

Responding to a poll, whether it is a political questionnaire or assessing a consumption good or service, is a costly enterprise for the rational consumer. Polls traditionally involve long phone calls, generally placed around dinner time when the family is sitting down to eat, and include very many questions regarding levels of satisfaction and premature opinions. Responding to these questions in a comprehensive way that accurately conveys how the voter feels can be a costly and time consuming matter. Furthermore, not everyone has a home phone and prefer to only carry cell phones, which may mean the results will not be representative of the entire population. In effect, the costs of polling have increased. However, the face of polling is changing. In 2012, Microsoft, through Xbox, conducted a presidential poll. Polls via text messages are going to be a new strategy to gather information, as well as polls via television, which allow the poller to skip ads to give information instead.

The question remains, however, if these new methods will be enough to make responding to polls worth while. Even though the new strategies are more convenient because they allow people to respond to questions on their own time, they still require the voter to take time out of their day to answer questions that may not even have an effect on political parties' platforms. The rational voter will still abstain unless the benefits are high enough to make it worth while, e.g. if they really hate ads.

1 comment:

Patrick Lorenz said...

According to economic analysis, the populace will not participate in the poll unless the benefits of doing so outweigh the costs. Even if the costs were minimally above zero, most people would probably still would not participate because the benefits actually are zero. However, for some groups with extreme preferences participating in polls might seem to be rational; if the sample of those who choose to participate in a poll is unrepresentative (which it almost undoubtedly is) then citizens can cause the distribution of voter preferences to appear to lie farther to their side than it actually does, which results in candidates running closer to their preferences than those of the actual median voter. However, this makes the party on their side less close to the median voter and thus more likely to lose, so it is probably irrational to attempt to influence the results of a poll in this manner.