Wikipedia entry on sample (October 2, 2006). Category: Wiki pages
In earlier weblog entries, I described my efforts to improve a page on Wikipedia, binomial proportion confidence interval, that was flagged as being too technical. One of the Wikipedia pages that I linked to, Sample (statistics), was identified as a stub, so I took some efforts to improve this page as well.
I linked to a variety of other pages related to the concepts of a statistical sample:
- Statistical population
- Sampling (statistics)
- Bias (statistics)
- Random sample
- Simple random sample
- Systematic sampling
- Stratified sampling
- Cluster sampling
- Nonprobability sampling
- Snowball sampling
The quality of the writing in these uneven, of course, but generally quite good. The pages on random sample and simple random sample are a bit muddled, and I might insert an example that points out the distinction. A random sample is a sample where every individual has the same probability of being selected whereas a simple random sample is a sample where possible sample of the same size has an equal probability of being selected. So an example might consider a sample of size 4 from the population of the integers 1, 2, 3, 4, 5, 6, 7, and 8. There are 70 such samples. A sampling approach that would produce a random sample but not a simple random sample is to flip a coin and to select all the even integers if the coin comes up heads and to select all the odd integers if the coin comes up tails. Any individual member of the population has a 50% chance of being selected, but certain samples, such as 1, 2, 3, and 4 would have zero probability of being selected. The example cited here is actually an example of systematic random sample because you randomly start at either 1 or 2 and then select every second member of the rest of the population.
Previous weblog entries on this topic are:
This work is licensed under a Creative Commons Attribution 3.0 United States License. It was written by Steve Simon and was last modified on 04/01/2010.