New England Journal of Medicine on Missing data

[Carlo Lazzaro]

Dear All,
for those, like, me, who are interested in missing data analysis, I've found this video (https://evidence.nejm.org/doi/full/10.1056/EVIDstat2200) funny but with a sound educational aim.

[Clyde Schechter]

Carlo, that link gives a 404 Error (page not found).

[Carlo Lazzaro]

Clyde:
I've found that the link (https://evidence.nejm.org/doi/full/1...VIDstat2200065) was limping this afternoon (your morning there) but worked 100% for me by copying and pasting it in my Google browser (I've just tried now successfully).
Sorry for that and thanks for pointing this out.

The full reference is:
The Case of the Missing Data
Kate Takvorian, M.D.,
Chana A. Sacks, M.D., M.P.H., and
Charles C. Hardin, M.D., Ph.D.

Published March 22, 2022
NEJM Evid 2022; 1 (4)
DOI:https://doi.org/10.1056/EVIDstat2200065

[Clyde Schechter]

Thanks. That works.

I think their definition of MAR is stated in an oversimplified way that is, taken literally, incorrect. But at least it has the virtue of communicating that MAR does not mean what most people who haven't studied the subject think it means. And it does emphasize that even small amounts of missing data can be a problem. So, on balance, I think it's a really good introduction to the subject. Thanks for posting that link.

[Carlo Lazzaro]

Clyde:
happy that it woks.
Yes, I do share your point about the risk of oversemplification (in fact there are tons of textbooks on this topic, that is not easy at all) about the MAR mechanism.
Personally I've found the NMAR mechanism example really funny!

[Clyde Schechter]

Actually I was most amused by the MAR example because the old person who has difficulty using smartphone apps is something I could really identify with!

[Carlo Lazzaro]

Yes, I see the issue Clyde!
From my 50 onwards, I couldn't ask too much out of my lenses!!!