On 5 Oct 2002, Christopher Larsen wrote: > I am looking at doing an analysis of some data on memory performance > on a standardized test of recall memory. The norms on each scale of > the test are 100 with a standard deviation of 15. I am wanting to > compare my data set to the normative data. Which would be a superior > test to use, a Z test or a One Sample T-Test???
Presumably you want to compare the mean of your data set (not the data set itself) to the mean of the normative data. Depends partly on how strong your reasons are for supposing that the Ss in your sample are largely like the Ss in the normative sample. (If you're dealing with a specialized subpopulation, there may be no reason to believe that either the mean OR the standard deviation of the normative data are applicable.) Basically: which variance do you propose to use? If the normative 15^2, a Z test is appropriate; if you intend to use the variance in your sample, you're into a t-test, and quite possibly a two-sample t-test (supposing you know, or can find out, the size of the normative sample). [That's possibly oversimplifying, of course, but it's a start.] OTOH, you may also be wanting (though you didn't say so) to compare your sample variance with the normative variance: especially if you suspect your sample to be likely to show restriction in range compared to the normative group, hence smaller variance. (This could be due either to greater homogeneity in your sample than in the normative group; or to apparent greater homogeneity due to restriction of range because of floor or ceiling effects.) In such cases, using the sample variance will make your t-test more sensitive than it would be with the normative variance, a point that might require attention and acknowledgement in your discussion of results. ----------------------------------------------------------------------- Donald F. Burrill [EMAIL PROTECTED] 56 Sebbins Pond Drive, Bedford, NH 03110 (603) 626-0816 [Old address: 184 Nashua Road, Bedford, NH 03110 (603) 471-7128] . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
