A researcher wishes to test the differences among five population means, using the 0.05 criterion of significance. Why is it a bad idea to perform 10 t-tests for the 10 possible pairs of means?

Question 3 options:

A)

The probability of making at least one Type I error will be greater than 0.05.

B)

It is possible that the researcher will get some statistically significant results purely by chance.

C)

Both A and B.

D)

It is perfectly fine to compute multiple t tests instead of an ANOVA.