On 12 Mar 2004 09:00:42 -0800, [EMAIL PROTECTED] wrote: > > I really need a hand with a regression analysis for my thesis. > I'll try as best I can to summarize/explain. I have a DV, time on > task. I have a number of IVs, demographics of the participant, > cognitive tests. > > Generated a series of bivariate correlations for time on task. > Took all significant correlates of Time On Task (11 variables) and > entered them in a hierarchical regression, the ordering based on the > strength of the correlation.
Well, that's a bad idea right there. You can see my stats-FAQ for comments about stepwise regression, and further references. How many variables did you *start* with? - Since there were a lot, your procedure puts you into an exploratory mode where testing has been fatally undermined; and the "stepping" is mainly useful for selecting a *concise* model with few variables, assuming that you are confident that they really matter, and cover what else matters. > > My problem/question is this: models 1, 2, 3, and 7 are > significant. 4, 5, 6 are not. What the heck do I report? > Do I do a new regression based on this? If I report model 7 as > significant, do I report that the variables added in 4 - 6 were not > significant? > >And while I am at it, in the coefficients table, one of the variables > in model seven is listed as significant, and in the final model (all > 11 variables) only two are significant? What the heck do I report > here? I think that the term here is "nominally significant". When your program prints a p-level based on the number of variables, there is an implicit assumption that *this* model is the only one that was ever under consideration. Since you had a lot more, none of those tests are good. You would need a table that showed, "What is the p-value for the F for the best 11 out of 25 variables..." and so on. You might try Judd and McClellan, "Data Analysis, a model comparison approach." -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html - I need a new job, after March 31. Openings? - . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
