So I've read that the degrees of freedom of a Welch's Anova have decimals, so I wonder if when you report them you put the decimals or you round them up.:
a) F2, 426.67 = 58.054, p< 0.001
b)F2, 427 = 58.054, p< 0.001
I have looked for other articles for reference, but I have seen them both ways, so I am not sure if they are both admitted.
Since we do not use the DF in the interpretation of the results, there is no consequences in the rounding process. In other hand there is no error presenting the decimals. In my opinion, both options are valid what is confirmed in the literature "... I have seen them both ways ".
According to APA I would report two decimals, but I answered this within another thread where you asked the same question, didn't I?
Depends on your field. Its not wrong not to include decimals. You could report df to 1 decimal place (if any).
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA