Originally Posted by
meester69
uh, no. You can take a small sample or a large one. For example, if you know there are one million birds in an area, and you catch 10 of them, that would be a 'small' sample. If you catch 10,000, that would be a 'large' sample.
The standard error refers to the variance of possible samples from a given population. It decreases in proportion to the square root of the sample size. While 'large' is a subjective word in any case, it is clearly possible to apply fairly objective standards to it - just as we would all agree Bill Gates is a 'rich' man, it is also perfectly possible to take a sample that would be agreed upon as 'large'.
I could not agree more with this post by meester69 ^. It is ridiculous to suggest that all samples are small. The sample size must be considered in conjunction with the population size. Statistical methods allow us to state conclusions with a certain degree of confidence (for example 95%). Sample/population size play big roles in this calculation.