I hate the quant section. Can someone explain where/why I should use the one-tailed test and where/why I should use the two-tailed test? With a significance of 95 or whatever, I look at .025 for the two-tailed test and .05 for the one-tailed right?
It really just depends on the question you’re trying to answer.
To keep it simple for the purposes of the exam, consider the following. For example, if you’re told that someone suspects the mean (whatever, return, # soccer balls in a coaches car, etc) exceeds 5, you would use a one tailed test with the null and alternative as Ho: mu =<5 and Ha: mu>5, respectively. If they only said so and so believes the mean is 5 (or isn’t 5), you would have Ho: mu = 5 and Ha: mu not equal to 5.
You’re correct with your understanding of where to put alpha for a one or two tailed test when using the critical value tables.
Hope this helps.