Random sampling methods and applications

Random Sampling

Although it is always tempting to think that one's own population is the one of greatest interest to a potential politician and his or her pollsters, this would not be the case if such a future politician or even a political scientist wished his or her staff to conduct a random sampling of potential voters in the 2008 campaign. It would not be enough to simply sample the opinions of one's own class, should one desire to sample the opinions and likely voting patterns of a truly representative body of voters. First of all, many if not most of the voters in such a so-called random population would be young, by default of their student status alone. Also, in terms of future predictions of voting patterns, some might not be registered for the upcoming election, or desire to be. Even if they said they might vote a particular way, the might not become habitual voters, or they might change their location in the future, as young people tend to be more mobile, as well as less likely voters than older and more stable people (both in opinions and residency).

Also, the potential class voter's economic situation would be likely to change in the upcoming years. At the time of the survey, the typical class voter's interests would no doubt involve educational loans, rather than high property taxes, for example, or the quality of education from a student rather than a parent's point-of-view, and other specific interests particular to a younger, student age group, however large or randomly selected the class size.

Demographically, one's class might also be relatively homogeneous ethnically, geographically within the United States, and might be more likely to have liberal social views, in comparison to their parents, by virtue of their youth. A high school or college student would also even be more apt to be exposed to particular forms of media, such as the Internet, than would other voting groups.