How can I statistically analyze the correlation between basic physico-chemical properties (which have a single value per depth) and aggregate-related parameters (which have multiple values per depth due to different aggregate sizes)? Should I average the aggregate parameters per depth, or is there a better statistical approach to ensure meaningful correlation analysis?
Thank you in advance
The most statistically robust approach is to use mixed-effects models. These models account for the hierarchical structure of your data, where aggregate-related parameters vary within each depth, and physico-chemical properties remain constant. Mixed-effects models treat depth as a random effect, allowing you to analyze correlations while preserving the variability in aggregate data without oversimplification through averaging.
Afiaa Raheem Khudhair Thank you so much for your suggestion. Highly appreciated
Show the physico-chemical parameters and aggregate-related parameters as percentages (in the form of an undefined number) and perform a correlation analysis. On this way I made a correlation between the results of the physico-chemical analysis of water and the results of the saprobiological analysis based on malacological and ichthyological material. I send you my paper about it.
Milan B. Arambasic Thank you so much for your suggestion and reference paper. Highly appreciated
- Badges
- Science topic