I have a data with 2 groups (fit/healthy) and dietary intake data as ordered categories (less frequent to more frequent as 1,2,3,4,5). I have to assess the relationship between the groups and the dietary data. Hypothesis being "frequency of certain food groups has a direct/indirect relationship a particular food group" as such. What would be the appropriate test to measure this? I have some potential confounders such as BMI, gender, smoking etc. which I have to correct for, so Chi-square is out of my options.
$\begingroup$
$\endgroup$
1
-
$\begingroup$ This problem statement is not clear: "frequency of certain food groups has a direct/indirect relationship a particular food group". Can you clarify which of the two variables mentioned in the title is the predictor and which is the response? $\endgroup$– dipetkovCommented Apr 21, 2022 at 22:12
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
This is a job for ordinal regression, which seeks to find the relationship between an ordinal outcome and a set of predictors of any scale. The most common example is ordinal logistic regression, which models the log odds of being in the next higher category as a linear function of the predictors. See Bürkner and Vuorre (2019) for a tutorial.
