kevinwang

Thanks, will look into this by this weekend. On Wed, 8 Dec 2021, 2:09 pm Dianne Cook, ***@***.***> wrote: > About half the math scores appear to be missing, but...

Looks like most of the missingness are from the year 2000, and this occurred across many countries. For 2018, there is only one country (Vietnam) that had missing values. ```...

Added in the GitHub version now: student_full_data/student_2022.rds We are preparing a submission to CRAN soon.

@dicook, the installed package is about 5.1MB even after reducing the number of rows. I suggest that we subset the data to just the OECD countries: https://www.oecd.org/en/about/members-partners.html. Procedures: https://github.com/kevinwang09/learningtower/blob/master/inst/sampling_student_and_school.R

The use of factor column was the main cause of a large package size. After re-curation, - We ensured that year and school_id are now integer and character columns. -...