Sampling a dataset with 95% confidence level?

Question

I have a large dataset and want to analyze a statistically significant sample (95% confidence level, 5% confidence interval).

The data samples contain commit information, so I cannot calculate mean or standard deviation. How to calculate the number of samples I need to check?

Answer 1

You're asking the wrong question. Sample size required has nothing to do with the percent confidence interval. The larger the sample size, the smaller the absolute range will be on the confidence interval.

That said, if you're performing some kind of regression in this hypothesis test, your data needs to meet certain standards.

I have no idea what "commit information" means. Is that related to github or something? What exactly are you trying to do?

Answer 2

I have no idea what you're asking

Answer 3

Your whole question shows you have no clue what you are talking about.