Hi group,
I am evaluating the correlation of two continuous variables using Spearman correlation and JMP14. I read in an article that for a similar study, since there was an additional factor that showed differences in the levels for one of those variables, they adjusted the Spearman correlation analysis by this factor. I can calculate the Spearman correlation between both variables using the Multivariate methods platform, but how can I account for this effect?
Thanks,
Dave
Thanks @Byron_JMP. This would be the reference I talk about:
https://www.ncbi.nlm.nih.gov/pubmed/30165638
Inflamm Bowel Dis. 2018 Aug 24
Vedolizumab Drug Level Correlation With Clinical Remission, Biomarker Normalization, and Mucosal Healing in Inflammatory Bowel Disease.
Al-Bawardy B, Ramos GP, Willrich MAV, Jenkins SM, Park SH, Aniwan S, Schoenoff SA, Bruining DH, Papadakis KA, Raffals L, Tremaine WJ, Loftus EV Jr.
In the Statistical Analysis section they mention:
"Spearman partial correlations were calculated to measure the association between VTLs with BMI and albumin, adjusting for dosing frequency"
Then there is an additional references discussing this matter:
https://www.ncbi.nlm.nih.gov/pubmed/29131931
Biometrics. 2018 Jun;74(2):595-605.
Covariate-adjusted Spearman's rank correlation with probability-scale residuals.
Liu Q, Li C, Wanga V, Shepherd BE.
would you mind posting the reference?
