Homework
Do you calculate an Fmax for a t-test? If so, how would you do this?
posted by ashleynevin @ 7:54 PM
This blog is for the use of students enrolled in either section of PG851. It is intended to be a forum for asking, discussing, clarifying, and helping.
posted by ashleynevin @ 7:54 PM
6 Comments:
At 8:18 PM,
Unknown said…
and also what kind of post-hoc would we use for an ANOVA prior to analyzing the whole ANOVA
At 7:31 AM,
Mari Clements said…
You compute an Fmax the same way, no matter what the analysis is.
You would NOT do a post hoc prior to computing an ANOVA.
At 4:16 PM,
Unknown said…
In regards to doing the ANOVA, do we then do the F-Max to see which Kind of Post Hoc to use, one in which equal variance is assumed or not?
At 9:49 PM,
T.J. said…
Hello, Unknown. :)
Yes, whether you violated the assumption of homogeneity of variance will affect what kind of post-hoc tests you do.
With similar cell sizes, you use F-max, but remember you can also use Levine's Test if the cell sizes are very unequal. Remember that outliers and two-tailed tests are also ways to violate homogeneity of variance, so look at more than just F-max/Levine's. (Slide 16 Week 1)
With a violated assumption of homogeneity, you can use the post-hoc tests of Adjusted Scheffe's, Dunnett T3 or Dunnett C depending on (n), or Games-Howell. (Slide 22 Week 2)
At 10:06 PM,
Unknown said…
In response to T.j. Is there an SPSS thing that we can do that will tell us if there are outliers. I don't know what would count as outlier and what wouldn't, how much of a difference would an outlier have from the normal scores?
At 7:22 AM,
Mari Clements said…
Unknown, see http://www.blogger.com/comment.g?blogID=37905226&postID=6120432978165046873
for the answer to your question about outliers.
Post a Comment
<< Home