Hi together,
I'd like to conduct a mixed-design ANCOVA in R with two covariates.
I used the afex package to run the ANCOVA:
BU_PK_trust_ancova = aov_ez("id", "trust_rating", data2_long, within= c("consultant_type"), between= c("Condition"), covariate= c("BU.Relevanz", "product_know"), factorize = FALSE)
The problem I have is that the two covariates "BU.Relevanz" and "product_know" have a correlation of 0.459.
So far, I couldn't find a threshold value for that correlation, on one page it only says, the correlation should not be too high.
Does anyone of you know what value I can use here?
Thank you!
I don't think that there is a threshold set for that correlation. You can use VIFs to quantify the degree of multicollinearity.
Are you interested in their relative impact? If not, then their collinearity shouldn't be an issue. It seems that you only want to control for their joint influence, so their collinearity should be irrelevant.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Software
- R Statistical Package