On 16 Sep 2003 03:28:23 -0700, [EMAIL PROTECTED] (emma) wrote: > Hi there, > > Am new to posting - so let me know if haven't included enough info etc. > > I have conducted an ancova with two between subjects (group - 3 > levels) and education (2 levels) IV's and one covariate. The DV has > been transformed using sq root. When I run the analysis I have an > interaction between group and the covariate suggesting a lack of > homogeneity of regression slopes. In trying to understand the > interaction (which was unexpected) - I ran some diagnostic tests on > the covariate which I found to be skewed. I decided to transform the > covariate (using sq root in line with the DV) - and this normalised > the distribution. When re-running the ancova using both of the transformed > variables the interaction disappeared. Is this valid? Is it
If the covariate was the same thing being measured as the outcome, then you would have a hard time justifying *to me* an analysis where they were transformed differently. If taking the square root was entirely arbitrary, with no justification except for improving the looks of the test, in ways that have not been specified; then, *to me*, you would have a hard time not-doing a transformation that happens to remove an ugly interaction. > necessary to transform a covariate if: 1) it is non-normally > distributed and/or 2) to be consistent with the DV? I should probably > mention that although I have homogeneity of variance the sample sizes > are small and unequal (46/15/15). > > A further point, which may or may not have a bearing on the above, is > the issue of independance between the DV and covariate. The DV is a > percentage of overlap between words ticked on two lists - say a and b > i.e. a/b * 100 - and the covariate is b - as participants were able to > tick as many or a few words as they wanted - and so I wanted to > control for sheer number. Is this valid? Or should the DV and > covariate be completely independant? You have two variables, you then you take a ratio, and then you think of doing transformations? - I think that your initial questions were premature, by far. You don't seem to know enough to be considering subtle transformations. You need to explore your data, and explore the *logical* relationships of your measures. Figure out what *might* make sense, before you even look at a scatter chart. Then look at a scatter chart before you even think of doing any transformation or regression. You might try Judd and McClelland's book since it emphasizes models. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html "Taxes are the price we pay for civilization." . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
