Talk:Survey sampling

From Wikipedia, the free encyclopedia

[edit] Random sampling

The article claims:

In public opinion polling by private companies or organizations unable to require response, the resulting sample is self-selected rather than random. Volunteering for the sample may be determined by characteristics such as submissiveness or availability. The samples in such surveys are therefore non-probability samples of the population, and the validity of estimates of parameters based on them is unknown. They are, however, unquestionably random samples of that sizeable subset of the population which volunteers for opinion surveys.

(Emphasis added.)

Are they really "unquestionably random samples"? What about the portion of the population that would volunteer for an opinion poll if they were included in the set of potential pollees, but for whatever reason they aren't? Someone might not be included in the set of possible polling candidates for a wide variety of reasons -- for a telephone poll, for example, indidividuals who only have a mobile phone are not polled. Neilc 12:39, 4 Nov 2004 (UTC)

The article also claims that all survey samples are random, which is not true. Most survay samples are random, however one can also take a purposive sample in which you choose a sample which will best represent the population. This method is important as taking a random sample is sometimes difficult due to cost. Statistical inference can also be model-based inference which does not rely on the randomness of the sample like design-based inference, under this inferential method a purposive sample can actually improve your estimates. Henderson, 6 April 2006 (ANU)

I had the same thoughts - there are many other factors which can prevent a sample from being random, so I removed the problematic sentence. Interesting to see that this problem was first noticed over two years ago! Brendanfox 03:36, 17 November 2006 (UTC)