The main consideration when we have a large number of tests is that probability of getting a significant test due to chance alone increases. This will increase the type 1 error (rejecting the null hypothesis when it's actually true).
Therefore we need to consider the Bonferroni Effect which happens when we make many tests. Ex. If our significance level is 0.05 but we made a 100 test it means that the probability of getting a value inside the rejection rejoin is 0.0005, not 0.05 so here we need to use another significance level which's called alpha star = significance level /K Where K is the number of the tests.
