I am analysing a dataset using a log-level regression in R.
My two key variables of interest are called MTenure and CTenure. Both are significant and the respective unstandardised coefficients are 2.460e-03 and 3.391e-03. They are measured on the same scale and I interpret this as CTenure being able to have a greater impact on my DV than MTenure.
I noticed when calculating the standardised coefficients with lm.beta() that my standardised coefficients suggest a different picture as MTenure has a standardised coefficient of 0.44, which is greater than CTenure's coefficient of 0.19. So the standardised coefficients would suggest that MTenure has a greater impact on my DV, whereas my unstandardised coefficients suggest the opposite. Can this make sense?
Some things I noticed during the analysis: MTenure has some 0-values, CTenure is only just significant (p = 0.04) relative to MTenure (p < 0.01) and when running lm.beta() I get the following error message, which is a bit cryptic to me:
In var(if (is.vector(x) || is.factor(x)) x else as.double(x), na.rm = na.rm): Calling var(x) on a factor x is deprecated and will become an error. Use something like 'all(duplicated(x)[-1L])' to test for a constant vector.
The reason that I am using standardised coefficients in the first place is to compare the explanatory power of MTenure and CTenure to other variables in the model that are measured on different scales.
I'd be very grateful for any help!