So in principle, a statistically significant result is not attributed to chance.
I'm given 70 independent experiments on the number of actions of 500 customers over a course of a month. Each experiment is on the same set of 500 customers, but it is believed that their number of actions will increase over time.
I'm given the sample mean and alpha, and I was able to calculate the p value. Now I'm asked to find the probability that I find some significant results just due to chance.
Can someone explain to me what should I do in this case with the information I have so far?
EDIT: $H_0 = 49.75$, $H_a \ge 49.75$, $\alpha = 0.05$
I'm using $t$ test and Python scipy.stats as st
alpha = 0.05
ttest = st.ttest_1samp(data, 49.75)
pval = ttest[1]
if (pval > alpha):
significant = 1