Reputation: 1
How can I combine two significantly associated variables (A and B) into one component and test for the effect of variable C on the two combined?
Variable C is an external factor that causes variable A and B to change (body related factors). I know that variable A and variable B measure something very similar and are correlated. I want to combine them to increase precision.
How should I got about doing this? Is using PCA to create a composite score a good idea? I'm a little bit lost and can't wrap my head around that. I've read literature doing similar things and different methodologies but I wasn't sure how to approach the issue.
I'm not looking for something flawless, just something robust enough to strenghen my anaylsis.
Apologies if my terminolgy is unclear, or inaccurate this is a little bit new to me. thanks!
I have shown the effect of variable C on each of them separately, have shown that both correlate with high significance (p<0.00001).
Upvotes: 0
Views: 12
Answers (0)
Related Questions
- SEM with lavaan in R, problems specifying model with correlated subscales. WARNING: Could not compute standard errors
- Is it possible to combine separate boxplot summaries into one and create the combined graph?
- Can I combine two strings into one, and use the combined name to assign a data frame to?
- Automatic procedure to select combination of vectors with best predictive power
- Combine aggregated variables into one variable in R
- How to combine different variables into one variable in R
- How can I combine values from three variables into one variable?
- How can we combine predictors from two different linear models into one?