I want to calculate the average standard deviation from my data set, but I don’t know which method will be correct.

I use an exponential curve fit on my raw data and and I get numbers with already calculated standard deviation. Per parallel experiment I have five replicates.

The resulting data set is, for example:

69,3 +/- 0,5

69,8 +/- 0,6

69,1 +/- 1,2

68,8 +/- 0,7

69,8 +/- 0,5

From these, I want to calculate the average. Currently, I have been using SD(avg) = 1/N*(root(sum(SD^2))) for finding average SD. Not sure it’s correct…