I have a bunch of data in the following format (sorry, it does not format well on here):

Key Indicator-----------First Group------------Second Group

------------------Observations---Mean-----Observations----Mean

1 Grade change

negative----------35-------------11,22%--------2--------------0,20%

no change------149------------47,76%-------316-----------32,51%

positive---------128-------------41,03%-------654-----------67,28%

I am trying to do a T-test to see whether these values are statistically significantly different for Group 1 vs Group 2. It's a bit complicated because the sample size is different for each group AND the mean is in percentages. I did read that this is possible, here: http://www.statpac.com/statistics-ca...r/percents.htm

However, I am unsure how to do it in Python...

I looked up how a regular t-test would work and found this:

from scipy.stats import ttest_ind

ttest_ind(mean1, mean2)

Or something like this?

stats.ttest_ind(a,

b,

equal_var=False)

Can I use one of these and just input the different percentages? ]]>