How to model observed percentages (bounded from 0 to 1)

I could not find any examples for binomial. I have access to the n and the k because each row in my dataset is actually the result of a test. Each test has a total # of points and the actual points the person got. We divide those to have a percentage. So I guess I could use the total # of points as my number of trial and the actual scored points as the number of successes. But how do I specify those observed data to the Binomial distribution? The distribution needs a total # of trials n and we want to find p. The total # of points can change for each row in my dataset also.

Thanks,