PG851: GLM--ANOVA

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.

Tuesday, February 02, 2010

Results section for factorial ANOVA

Which is more correct? Or does it even matter?

A two-way ANOVA was used to examine the relationship between.....


A factorial ANOVA was used to examine the relationship between......


Thanks

posted by Lauren B @ 7:39 PM  

6 Comments:

  • At 9:21 PM, Blogger Andrew said…

    The one that best describes the analysis you ran is the correct one.

     

  • At 9:22 PM, Blogger Christie Turner said…

    what's the difference?

     

  • At 9:49 AM, Blogger Grace Liu said…

    One of them is more precise than the other.

    On your factorial lecture notes, slide 4, you can find a brief description of factorial ANOVA.

    On your factorial lecture notes, slide 5 and 6, you can find some examples of two-way and three-way ANOVA which you will probably get what two-way and three-way ANOVA mean.

    Hint: one is more precise description of an analysis than another one.

    Let me know if you cannot comprehend from the notes.

     

  • At 8:23 PM, Blogger Mari said…

    And, in case none of these comments give you big enough hints, let me say that it is best to indicate the number of levels you have in your factors. So, for instance, if you had a model that had three therapeutic conditions and three disorders (i.e., you wanted to assess the outcomes for individuals with Major Depression, Generalized Anxiety Disorder, and Borderline Personality Disorder who were randomly assigned to CBT, psychodynamic, or client centered therapy), it would be most correct to refer to this as a 3 X 3 ANOVA.

    But...factorial ANOVA is also correct...just not as precise.

     

  • At 11:25 PM, Blogger sarahmoon said…

    for question 2, the hw asked us take var_E (which originally contained 3 levels) and put them into two groups. on the writeup, do we go with 2 levels of the var_E or 3?

    hope this question makes sense!

     

  • At 7:33 PM, Blogger Mari said…

    You go with the number of levels you actually analyzed. That is, if the variable as you analyzed it has 4 levels, it would be a something X 4 ANOVA. If you collapsed those levels into 2, the 4 levels aren't there anymore, so you would use a 2 in describing the analysis.

     

Post a Comment

<< Home