So I have a bunch of scores 0-100 and a two sets of weights, the frequency which is the number of people in the sample with an average score among those people. Then I have another value that is the fraction of people in that group for the given score. An example may look like this:
"Average Score of sample" "Number of people in sample" "Subset of people in sample" 65 8 .75 87 4 1.0 34 1 1.0 50 3 .33333 72 9 .66666
Here is a link to a Google spreadsheet that provides three examples of the calculations in questions on a set N=100: Spreadsheet Example
So my question is, on a larger sample size, can I use the fraction as the weight(Column M) and still expect similar results to the Frequency * Fraction (Column N) weighting approach? How could I go about analyzing data to find a bias between the two approaches?
Thanks for your insight! If any additional information is needed please let me know!