probably shouldnt be using an lm for the outcome with 4 categories anyway. is pub_date a date? i would change that to something more interpretable like time since some start date. untested: loc_rating$loc_rating_ind <- as.integer(with(loc_rating, loc_rating == 'Good' | loc_rating == 'Outstanding')); glm(loc_rating_ind ~ pub_date, data = loc_rating, family = 'binomial')

did you look thru the vignette of that package? cran.r-project.org/web/packages/margins/vignettes/…

no you multiply by 15 not divide pmin(1, p * length(p)); p.adjust(p, method = 'bonferroni')

for the categorical case, you can add another (non NA) value for the missing cases, addNA does something similar

you already know the answer: "mean difference / standard deviation" d <- (mean(M$V1) - mean(M$V2)) / sd(c(M$V1, M$V2))

@DanChaltiel hmm, good point. I'm not sure, but it seems to do with how finegray handles discrete time (or times with ties). For example, if you add a tiny amount of noise to the time variable, you get the same coefficients: run transplant$futime <- transplant$futime + runif(nrow(transplant), 0, 0.1) first then fit the models. There is a line in finegray which is newtime <- matrix(findInterval(Y[, 1:2], utime), ncol = 2) where utime are unique times, so that may be the culprit

race, ed, drace, and ded are categorical variables with more than two levels each, you can't just throw them into a model like that

stats.idre.ucla.edu/other/mult-pkg/whatstat