If I have 1,000 participants ranking on a scale of 1 to 10 regarding some object how do I interpret the confidence level and margin of error of the resulting rank? I am used to of seeing 99% confidence level and 4% margin of error type notations so how do these numbers play into my sample case? And how does the resulting rank fit with the large n and Central Limit Theorem?
I am weighting each rank against the percentage of total participants to get a final result.
And could you also explain what a response distribution is related to this scenario and why it is best to assume 50%?