profandyfield / discovr

discovr package for R to accompany Discovering Statistics Using R and RStudio
133 stars 25 forks source link

Issues in tutorial "Multilevel model" #7

Closed ellen-ying closed 3 years ago

ellen-ying commented 3 years ago

In section "Fitting multilevel models": There are several places where the font effect doesn't show: image

The overall intervention effect should be 6.79 instead of -4.3: image

profandyfield commented 3 years ago

Thanks these comments are super helpful (and your others). I'm going to work on correcting these errors over the next few weeks and will post an update here soon. For now, I wanted to thank you for the helpful feedback.

ellen-ying commented 3 years ago

Thanks these comments are super helpful (and your others). I'm going to work on correcting these errors over the next few weeks and will post an update here soon. For now, I wanted to thank you for the helpful feedback.

Hi Andy! While you fix the errors, can you also help to check if there is anything wrong with the module on the growth model? I have trouble opening the tutorial and here is the warning I got:

Quitting from lines 699-704 (discovr_15_growth.Rmd) Error in ref_grid(object = list(modelStruct = list(reStruct = list(id = c(0.388982154049936, : We are unable to reconstruct the data. The variables needed are: time_num intervention Are any of these actually constants? (specify via 'params = ') The dataset name is: rehab_growth_tib Does the data still exist? Or you can specify a dataset via 'data = ' Calls: sourceWithProgress ... %>% -> -> -> do.call -> ref_grid Execution halted

(Not opening a new issue since there are already a lot from me...)

Thank you very much!

profandyfield commented 3 years ago

These have been fixed. The growth model tutorial, not sure what the issue is but I found a typo in the arguments for emtrends so I'm hoping that correcting that sorts it out. Would you mind doing a force reinstall and reporting back?

remotes::install_github("profandyfield/discovr", force=TRUE, dependencies=TRUE)
ellen-ying commented 3 years ago

These have been fixed. The growth model tutorial, not sure what the issue is but I found a typo in the arguments for emtrends so I'm hoping that correcting that sorts it out. Would you mind doing a force reinstall and reporting back?

remotes::install_github("profandyfield/discovr", force=TRUE, dependencies=TRUE)

I force reinstalled and I saw the errors in the multilevel model tutorial have been fixed. However, opening the growth model module still gives the same error message...

profandyfield commented 3 years ago

I've tried something else, would you mind trying a forced install again and reporting back (sorry to ask, but I can't reproduce the error on my machine so it always knits for me without an error).

ellen-ying commented 3 years ago

I've tried something else, would you mind trying a forced install again and reporting back (sorry to ask, but I can't reproduce the error on my machine so it always knits for me without an error).

Amazing! It works right now. Thank you very much!