Freq Command on %gra Tier

[Kimberly Mueller]

Dear All,

Thanks to my previous thread with Brian, I am excited to examine grammatical relations using the MEGRASP dependency analysis.  I was unsuccessful in figuring out the correct syntax to extract the frequency of  the following grammatical relations:  CMOD, XMOD, CPRED, XPRED.
Would someone be willing to help with that?

I am hoping to achieve a subordination index in order to assess syntactic complexity in a group of healthy adults at risk for AD. 

Thank you!
Kimberly

[Brian MacWhinney]

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.

[Kimberly Mueller]

Dear Brian,

Thank you for this, it is working nicely.  I went through the manual, and decided to add a few other grammatical relations in order to capture syntactic complexity. I am hoping that you (or others) might look it over and let me know if this list appears sound?  The entire list includes the following:

CSUBJ:  the finite clausal subject of another clause
COMP: the finite clausal complement of a verb
XCOMP: the non-finite clausal complement of a verb
CPRED: a full clause that serves as the predicate nominal of verbs
CPOBJ: a full clause that serves as the object of a preposition
COBJ: a full clause that serves as object
XJCT: a non-finite clause that attaches to a verb, adjective, or adverb
NJCT: the head of a complex Noun Phrase with a prepositional phrase attached as an adjunct of a noun.
CMOD: a finite clause that is a nominal modifier or complement
XMOD: a non-finite clause that is a nominal modifier or complement.

Taking the sum of these captured and then dividing by the total number of GRs for that sample, yielding a syntactic complexity ratio. 

Thank you again for all of your help,
Kimberly

[Brian MacWhinney]

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.

[Kimberly Mueller]

Excellent.  May I send you a couple of sample transcripts with their analyses?

Many thanks,
Kimberly