Closed bbolker closed 6 years ago
there's a second issue here. I don't have an example, but I've made some of the dw_tidy()
logic for the by_2sd=FALSE
case parallel with the default by_2sd=TRUE
case, avoiding the "." semantics which are said (in comments) to be fragile ...
Thanks, Ben!
dwplot
does anna.omit
step after matching order of term names. I assume that's done to eliminate terms that weren't matched in the specified order, but it has bad side effects for the new version ofbroom.mixed
, which can include agroup
column in the tidied model that isNA
for the fixed effects. This modification looks for NA values only in theterm
column ...