I have identified three archetypes in a dataset of coffee drinkers using Principal Convex Hull Analysis (PCHA): A: People who drink a little of high-quality coffee. B: People who drink a lot of high-quality coffee. C: People who drink a lot of low-quality coffee. We have hundreds of individuals distributed across this archetypal space. Each individual is represented by a mixture of the three archetypes. For example, one individual might have a composition of A = 70%, B = 20%, and C = 10%. Importantly, these three proportions always sum to 1 (100%). I would like to investigate whether these individual compositions are associated with an external variable, for example, the number of hours each person spends working per week. My initial thought was to calculate Pearson or Spearman correlations between the external variable and each of the three archetype proportions. However, because the proportions are constrained to sum to 1, they are not independent, and this approach could lead to misleading conclusions. There is a further difficulty: the association may be spread across several archetypes. Working more hours could be related to both B and C. Each archetype alone could then show a weak association, even though the combined proportion B + C could be strongly associated with the variable. What would be the appropriate statistical approach to quantify and test the association between the overall composition (A/B/C = predictors) and an external continuous variable (working hour…

Full article content could not be extracted automatically. Read the original below.