How to compute the standard error with only sample size, mean of percentage and interquartile range?

I have a percentage which is the rate of reduction. For example, each patient was measured on a scale several times, and the reduction rate was computed by (time 2-time 1)/time 1, I think. So, the reduction rate measures how much the performance is reduced in the second time.

I have two conditions for this dataset: 1. I have a total sample size, overall mean of reduction rate for all patients (one number), and interquartile range; 2. I have a total sample size and individual reduction rate.

My question is how to compute the standard error for these two conditions? I am suddenly confused about the equation. For the first condition, I thought I could compute the standard error simply by using SQRT(p*(1-p)/n), but I am not sure if this is suitable for my scenario. Could anyone help me with it? Thank you so much!


Active Member
im not totally sure what your asking butt im prutty sure SQRT(p*(1-p)/n) is not teh formula you seek. what you got sounds like a 'percent change from baseline' and is usally treated as a continous variable, i stdev/sqrt(n) or something, or whatever.