2 ms·
The article is good, but the math sucks. Statistically, 1000 elements is a large enough sample for binary studies, and is consistent with the number of people t
by tomp 13y ago
The article is good, but the math sucks. Statistically, 1000 elements is a large enough sample for binary studies, and is consistent with the number of people that are usually asked about their voting preferences on most polls. Also, the size of the sample does not depend on the size of the population you want to sample, provided that the population is large enough.
Now, whether the sample was actually representative of the population in question is another matter.