In APA 6.0 manual there is information (pp 113) about reporting degrees of freedom for Pearson correlation coefficient - and for parametric correlation it is sugested to report df's.
But what about nonparametric correlations? Should we report df's also or not? Is there any paper about it?
...Of course, I skip discussion on whether or not we should report df for nonparametric tests at all; I'm interested in APA standard guidelines/papers.
No, you do not need df for nonparametric correlation. Df are important if you run into t or F distribution. You can check what is the (approximate) distribution of Spearman correlation. That will tell you why you do not need df.
Dear Anil
Thank You for information, but that's not what i need. There were a lot of discussion about whether to report df for nonparametric tests and how to treat them (some information for example here: http://geographyfieldwork.com/SpearmansRank.htm - can be used for beginners).
But i'm asking about APA standard guidelines/papers. Are there any guidelines in the APA standard: report it or not report; somewhere accurately described what is expected, or some literature that I can make a clear referral?
Yup, we can use common sense or other ideas, but what APA standard says? :)
Regards
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS