User blog comment:Longguilol/Fairy Tail: Character's Battle Ratings/@comment-5274001-20120711225910/@comment-5274001-20120716230120

Alright, I really didn't wanna do this, but you leave me with no option.

In statistics and survey methodology, sampling is concerned with the selection of a subset of individuals from within a population to estimate characteristics of the whole population.

Researchers rarely survey the entire population because the cost of a census is too high. The three main advantages of sampling are that the cost is lower, data collection is faster, and since the data set is smaller it is possible to ensure homogeneity and to improve the accuracy and quality of the data.

Here is an online survey size calculator: http://www.raosoft.com/samplesize.html

Assuming that the FT fan population is one million (a very generous number), a survey population of 139 is SUFFICIENT to guarantee a confidence level of 95, i.e. the results will be within a 5% margin of error -- for characters like Erza, Mirajane, Laxus, etc, who are expected to get a 90% response distribution.

For the other normal characters (50% response distribution, i.e. the result will have an average of 5/10), around ~384 votes are enough to guarantee a result with a confidence interval of 95%.

I believe that I made my point clear here. I am not sure what you're trying to get at here.