@MicheleNuijten did you implement automatic sample size extraction? Would be useful to find over-cited studies, i.e. studies with high
-
Show this thread
-
citation / sample size ratios (controlled for years since publication). These studies are potential weak links in the literature.
1 reply 0 retweets 0 likesShow this thread -
And hence, should be targeted by replication efforts. Fastest way to root out error in the belief network. See also https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2655966/ …
1 reply 0 retweets 0 likesShow this thread -
Replying to @KirkegaardEmil
That would be great! But too hard (impossible?) in psych lit :( My plan is too look into medical lit though! Thanks for the ref!
1 reply 0 retweets 0 likes -
Replying to @MicheleNuijten
I bet one could get a large fraction of studies using simple regex approach. One primary problem is dealing with multi-sample/study papers.
1 reply 0 retweets 0 likes -
Replying to @KirkegaardEmil @MicheleNuijten
You should get into contact with some people who routinely work with NLP. They have many advanced methods you could use/learn from.
1 reply 0 retweets 0 likes -
Replying to @KirkegaardEmil
yes! we hope to make statcheck 2.0 "smarter" so we don't have to rely on rigid regex. at least the extracted df already give info about N
1 reply 0 retweets 1 like
See email for an approach. :)
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.