Code to support post on significance levels required for two sample z test of proportions.

twosampleprobability3.py

#Map how sample size changes as choice of p1 changes, holding all 
#else constant.

p1s = [x*.01 for x in range(96)]

data = []
for p1 in p1s:
    record = {}
    record['Probability Difference'] = p_diff
    record['Sample Size to Detect Difference'] = sample_required(p1=p1,
                                                                p_diff=.05,
                                                                alpha=.05)
    record['Confidence Level'] = '95%'
    record['Initial Probability'] = p1 * 100
    data.append(record)
    
df = pd.DataFrame(data)

Should line 9 be p1 instead of p_diff? p_diff is uninitialized in its current form. Thank you for the great example.