Dear Kimberley,
To count, for example, all the cases of XCOMP in the productions by the child in the Eve corpus in Eng-NA-MOR/Brown.zip on CHILDES, I used this command:
freq +t%gra +s"%|XCOMP" *.cha +u +t*CHI
You could add more +s switches for CMOD, CPRED, and XPRED. Go through the documentation of the GRs in the manual to see if you consider any others to involve embedding. Then, I suppose you want to take account of the overall size of your corpus by running
freq +t%gra *.cha +u +t*CHI
Then you can divide by the total number of tokens you get there.
--Brian MacWhinney
--
You received this message because you are subscribed to the Google Groups "chibolts" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chibolts+u...@googlegroups.com.
To post to this group, send email to chib...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/e7c2cd6c-3f9f-4d24-a5be-ab25617cac9d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Dear Kimberly,
This seems right the right list. I guess we would need to test this out on a trial corpus or two to find out the exact precision and accuracy of this type of filter. If you have a corpus or set of transcripts that you think would be a good target for this type of checking we could focus on that.
-- Brian
To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/76071fca-0b23-4c44-a102-736cc9c0db41%40googlegroups.com.