Skip to content
Prev 13263 / 20628 Next

Results from GLMM, error bars for predictions

Dear Quentin,

Please keep the mailing list in cc.

Dropping non significant terms from an ordered factor is not ok. That would
change the interpretation of the factor. You wouldn't drop non significant
levels of an unordered factor either.

Ben's solution is about multiple comparisons with random (and fixed)
effects. You're only dealing with multiple comparisons with fixed effects.
So glht() will do the trick.

Try plotting the predicted values for all relevant combinations of the
fixed effects. I find that easier to interpret than just a bunch of
coefficients.

Best regards,


ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature and
Forest
team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
Kliniekstraat 25
1070 Anderlecht
Belgium

To call in the statistician after the experiment is done may be no more
than asking him to perform a post-mortem examination: he may be able to say
what the experiment died of. ~ Sir Ronald Aylmer Fisher
The plural of anecdote is not data. ~ Roger Brinner
The combination of some data and an aching desire for an answer does not
ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey

2015-04-27 9:30 GMT+02:00 Quentin Schorpp <quentin.schorpp at ti.bund.de>: