I'm running a cross tab of one categorical variable against a dummy variable. This dummy variable indicates the presence or absence of a company characteristic(if the company is industrial it should be 1 and 0 otherwise). I this case, when running the exact test, shoul I consider the Exact Sig. 1 sided or the 2 sided? Thanks
2 Answers
Whether to use 1 sided or 2 sided tests does not depend on whether the independent variable is categorical or not. It depends on your hypotheses. One sided hypotheses (e.g. $H_0$ industrial companies will be lower or equal to nonindustrial companies on the dependent variable) get 1 sided tests. Two sided hypotheses (e.g. $H_0$ industrial companies will not be the same as nonindustrial companies on the dependent variable) get 2 sided tests.
That said, you may have trouble getting results published with 1 sided tests published - it may depend on the field - unless theory is very strong in your direction.
Also, doing 1 sided tests takes away some of your ability to be surprised by the data; as my favorite professor in grad school says: "If you're not surprised, you haven't learned anything".
Test of independence is always one-tailed (right-tailed). So you should consider Exact Sig. 1-sided. Because you are rejecting the null hypothesis for large values of chi square.