grr, arg
So I was having an okay day, feeling pretty good about things. Met with Elaine this morning and we worked out what will happen with my wonky results. Did some reading, made a "to do" list, ran a participant in Candice's lab. Then I made the mistake of opening up Statistica and trying to get towork on my results. The first couple regressions were fine, because they were simple, stepwise multiple regressions. But now I'm trying to get hierarchical regressions to work, and I have no idea what I'm doing. I even have the manual open and it's not helping me. There seems to be no way I can tell the program which variable I want to put in first, which is ridiculous because that's the whole point of a hierarchical regression! So I'm actually starting to seriously consider doing this by hand, which . . . is quite possibly a horrible solution. At least I have the time to do it while sitting here. Of course I don't have the formulae handy, but I could do it at home tonight. Arg. Curse you Statistica! Curse you a lot!!
[Edit: 2:44] Okay so I figured out how to get the regression thing to work, then realized I didn't want to do a hierarchical regression and had to start over again to do standard regression. Got that to work, so I have all the numbers I need. Now I just need to do a Sobel test to figure out if mediation occurred. Can anyone explain to me why I said I enjoyed this crap?
[Edit: 2:44] Okay so I figured out how to get the regression thing to work, then realized I didn't want to do a hierarchical regression and had to start over again to do standard regression. Got that to work, so I have all the numbers I need. Now I just need to do a Sobel test to figure out if mediation occurred. Can anyone explain to me why I said I enjoyed this crap?