1
$\begingroup$

I've been trying to figure out how to properly report a two way ANOVA with two between-subject variables and their interaction. Most references I found suggest to report it in the following way: F(df between-subject, df within-subject)=f-value, p=p-value, e.g. this site. But at another place I found a different way to report: F(df effect, df error) = F-value, MSE = mean-square error, p-value.

Now I am confused, because I don't have any within-subject measurements so in case the first approach is the right one I have no idea how to report my results. Here is my output from R:

                                     Df Sum Sq Mean Sq F value Pr(>F)    
 happinessLevel1                      1      5    4.73   3.449 0.0633 .  
 happinessLevel2                      1      0    0.13   0.096 0.7562    
 happinessLevel1:happinessLevel2      1    138  137.93 100.542 <2e-16 ***
 Residuals                       131810 180827    1.37

In case I report using the second version I'd have F(1,131810)=100.542, p<0.001 right?

$\endgroup$
  • $\begingroup$ Given that your references are for APA, look up what the current version of the APA manual says. The MSE is now required. So, you need to add in MSE = 137.93. $\endgroup$ – John Jun 7 '14 at 20:44
  • 1
    $\begingroup$ This question appears to be off-topic because it is about APA formatting and not statistics. $\endgroup$ – John Jun 7 '14 at 20:46
  • $\begingroup$ I found only APA references regarding how to report ANOVA. Is there another standard way to do so? And revisiting my question, you're saying that the degrees of freedom for errors are OK, right? $\endgroup$ – Pio Jun 7 '14 at 20:49
  • $\begingroup$ Especially, because the paper I want to report the results in is not in the field of psychology. $\endgroup$ – Pio Jun 7 '14 at 20:51
  • 2
    $\begingroup$ APA is widely used, not just in psychology. Essentially, you have to use whatever style the place you are sending this to requires. $\endgroup$ – Peter Flom Jun 7 '14 at 22:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.