spss complains that columns from file made using write_sav are constant about haven HOT 3 CLOSED

Please re-open if you can provide a reproducible example.

from haven.

Hi, sorry for not doing this earlier, I thought it'd be difficult and I couldn't share the original dataset. However, the error shows even using a two variable subset. It works using cars though.
I attached links to the data in rdata and sav format. Nothing odd about it as far as I can tell (and it's easy to examine the whole dataset). The data displays normally in SPSS v20 and does not look constant at all, but I cannot run a bivariate correlation between the two variables.
If I take only the first 50 rows, it works, if I take the first 200 of 209 it doesn't. I guess this explains why it works using cars, which has only 50 rows.

library(haven)
load("test.rda")
write_sav(example, path = "test.sav")

https://dl.dropboxusercontent.com/u/1078620/test.rda
https://dl.dropboxusercontent.com/u/1078620/test.sav

from haven.

I'm sorry, didn't notice this before. It's about missings. There's some missings on row 146, and that's what causes it. Changing the missings to some other value fixes it. After I changed them to a number, I can set them missing again and it still works.

from haven.