In general, when should you use "less than or equal to" or "greater than or equal to" symbols to write one-tailed DIRECTIONAL null hypotheses? For example, I have seen a null hypothesis for one-tailed directional tests look identical to non-directional null hypotheses, where the H0 is just written as u = XYZ (whatever the population mean is), even when there are key words in the question such as "increase" or "decrease." And I've seen null hypotheses for one-tailed DIRECTIONAL tests written as u >/= XYZ or u =XYZ. Sorry if this is hard to understand without an actual example. Curious if there's a general rule of thumb.