Hello everyone,
I was wondering whether anyone would be so kind as to give me advice about how to present Spearman rho correlation results in a table?
From my understanding, and from looking at various different research papers, you have to report correlation coefficients, sig and number of participants. Is this correct?
It would be great if someone could ping me an example!
Many thanks in advance,
Evelina
What you describe will certainly do.
The trouble with Spearman's rho is that it has no actual interpretation. You cannot explain what a rho of 0·6 actually is. The alternative, Kendall's tau-b is interpretable. It is based on the proportion of pairs of observations in which the two variables are positively correlated. In fact, it is
(proportion of positively correlated obs x 2) - 1
So you can reverse engineer the proportion of positively correlated obs by reversing the formula. A tau-b of 0·6 is
(0·6 + 1) ÷ 2
= 1·6 ÷ 2
= 0·8
So 80% of the pairs of observations show a positive correlation. I find that more intuitive.
Maybe this video could be helpful if using SPSS
https://youtu.be/tXD4Ny0DPOg
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication