![]() ![]() As a general rule, it’s not a good idea to go out of your way to try to interpret or explain the difference between a p-value of. The fact that one test is significant and the other isn’t doesn’t itself mean very much, especially since I kind of rigged the data so that this would happen. What does this mean? Should we panic? Is the sky burning? Probably not. ![]() When we ran the Student test (top row), we did get a significant effect but the Welch test on the same data set (second row) is not (t(23.03)=2.03, p=.054). The big difference here is that our result isn’t significant anymore. Not too difficult, right? Not surprisingly, the output has exactly the same format as it did last time too. You may recall the test output had two rows, one for equal variances assumed and one for equal variances not assumed. When SPSS computes a t-test, it will output both the Student t and Welch's t. To run a Welch test in SPSS is pretty easy. ![]() Like the Student test we assume that both samples are drawn from a normal population but the alternative hypothesis no longer requires the two populations to have equal variance. Figure 10.10: Graphical illustration of the null and alternative hypotheses assumed by the Welch t-test. What matters is that you’ll see that the “df” value that pops out of a Welch test tends to be a little bit smaller than the one used for the Student test, and it doesn’t have to be a whole number. It doesn’t really matter for our purposes. … which is all pretty straightforward and obvious, right? Well, perhaps not. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |