return plot-object as well for sjp.(g)lm when type = "vif" about sjplot HOT 5 CLOSED

I have updated the documentation. Currently, model assumption functions should not return any plot, only the updated model with removed outliers. Perhaps this may be an option for future releases.

from sjplot.

I am not sure I understand this comment. Removed outliers?

sjp.glm produces a plot -- even if type="vif". The plot just is not returned. I can confirm that the documentation is consistent with the behaviour now. But I'd much prefer to have the ggplot object returned to be able to further tweak it to my needs.

from sjplot.

If you refer to the very last plot of model-assumptions, this is the same one as sjp.glm(fit). type = "ma" first checks the model for outliers, and if there are any, they are being removed and an updated model w/o outliers is re-fitted. This updated glm-object is returned. Set showOriginalModelOnly to FALSE to see plots for both models, original and updated.

from sjplot.

This is an example what should be possible, I'd say. And easy to achieve. Sorry, should have posted code right in the beginning:

 # prepare dichotomous dependent variable
y <- ifelse(swiss$Fertility < median(swiss$Fertility), 0, 1)

 # fit model
 fitOR <- glm(y ~ swiss$Education + swiss$Examination + swiss$Infant.Mortality + swiss$Catholic,
              family = binomial(link = "logit"))

 sjp.glm(fitOR, type="vif")$plot + theme_bw()

from sjplot.

Sorry for confusion, I was talking about type = "ma", not type = "vif" - my fault... Currently, you can get access to the plot only via last_plot() function from ggplot. I re-open this issue and may add the plot as return value as well to that function.

from sjplot.