Complex syntax

34 views
Skip to first unread message

Cynthia Core

unread,
Aug 3, 2017, 9:29:21 PM8/3/17
to chibolts
Hi -
Are there any commands that can help me count the number of sentence types - simple/complex and count different types of complex sentences? I see commands to calculate ratios of one two verb utterances and IPSYN, but I am looking for something that is more specific to syntactic diversity and clausal diversity and that counts the number of complex versus simple sentences.

Thank you,
Cynthia Core

Brian MacWhinney

unread,
Aug 3, 2017, 9:39:08 PM8/3/17
to chib...@googlegroups.com
Yes, this is described in section 7.10.14 of the CLAN manual.

--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/0ddf9d6a-1f18-4678-b5ee-d13704204698%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Cynthia Core

unread,
Aug 23, 2017, 2:42:26 PM8/23/17
to chib...@googlegroups.com
Is there a procedure for freq of complex syntax for Spanish? I'm looking for the equivalent of the MEGRASP/freq procedure to count clausal types in Spanish using the %gra tier. 

Thanks,
Cynthia Core

On Thu, Aug 3, 2017 at 9:39 PM, Brian MacWhinney <ma...@cmu.edu> wrote:
Yes, this is described in section 7.10.14 of the CLAN manual.

--Brian MacWhinney

On 8/3/17, 6:29 PM, "chib...@googlegroups.com on behalf of Cynthia Core" <chib...@googlegroups.com on behalf of cynth...@gmail.com> wrote:

    Hi -
    Are there any commands that can help me count the number of sentence types - simple/complex and count different types of complex sentences? I see commands to calculate ratios of one two verb utterances and IPSYN, but I am looking for something that is more specific to syntactic diversity and clausal diversity  and that counts the number of complex versus simple sentences.

    Thank you,
    Cynthia Core

    --
    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+unsubscribe@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/0ddf9d6a-1f18-4678-b5ee-d13704204698%40googlegroups.com.
    For more options, visit https://groups.google.com/d/optout.
--
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+unsubscribe@googlegroups.com.

To post to this group, send email to chib...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.



--
Cynthia Core, Ph.D.
Assistant Professor
Department of Speech and Hearing Science
The George Washington University
2115 G St., NW
Washington, DC 20052
co...@gwu.edu
202 994 7364

Brian MacWhinney

unread,
Aug 23, 2017, 4:16:13 PM8/23/17
to chib...@googlegroups.com

Dear Cynthia,

There is a short, medium, and a long answer to this.  The short answer is “no”, but that answer is quite misleading.  The medium answer is that the grammatical relations of Spanish, particularly on the interclausal level, are nearly the same as those for English.  Therefore, the routine for complexity computation presented in the CLAN manual should work nearly correctly for Spanish.  The longer answer is that I could spend time working on this, but not for several weeks.  Actually, that is not a long answer.  Rather, it is a short answer that points to a long timeframe.

 

Best,

 

--Brian MacW

 

From: <cynth...@gmail.com> on behalf of "co...@gwu.edu" <co...@gwu.edu>
Reply-To: ChiBolts <chib...@googlegroups.com>
Date: Wednesday, August 23, 2017 at 2:42 PM
To: ChiBolts <chib...@googlegroups.com>
Subject: Re: Complex syntax

 

Is there a procedure for freq of complex syntax for Spanish? I'm looking for the equivalent of the MEGRASP/freq procedure to count clausal types in Spanish using the %gra tier. 

 

Thanks,

Cynthia Core

On Thu, Aug 3, 2017 at 9:39 PM, Brian MacWhinney <ma...@cmu.edu> wrote:

Yes, this is described in section 7.10.14 of the CLAN manual.

--Brian MacWhinney

On 8/3/17, 6:29 PM, "chib...@googlegroups.com on behalf of Cynthia Core" <chib...@googlegroups.com on behalf of cynth...@gmail.com> wrote:

    Hi -
    Are there any commands that can help me count the number of sentence types - simple/complex and count different types of complex sentences? I see commands to calculate ratios of one two verb utterances and IPSYN, but I am looking for something that is more specific to syntactic diversity and clausal diversity  and that counts the number of complex versus simple sentences.

    Thank you,
    Cynthia Core

    --
    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/0ddf9d6a-1f18-4678-b5ee-d13704204698%40googlegroups.com.
    For more options, visit https://groups.google.com/d/optout.



--
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/4CE8C958-6FC6-409F-ABE4-64DD11B66E9D%40andrew.cmu.edu.
For more options, visit https://groups.google.com/d/optout.



 

--

Cynthia Core, Ph.D.
Assistant Professor
Department of Speech and Hearing Science
The George Washington University
2115 G St., NW
Washington, DC 20052
co...@gwu.edu
202 994 7364

--

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.

Cynthia Core

unread,
Aug 24, 2017, 11:48:37 AM8/24/17
to chib...@googlegroups.com
Thanks, Brian.

We just ran the frequency count for the grammatical complexity analysis on our Spanish data, and we noticed that there were only two types of complexity identified: 
- CMOD and COMP

Other complex sentence types (e.g, XCOMP, NJCT, etc) are not present in the output at all, so does that mean that those types are not identified by MOR?

We ran MOR on the files using the Spanish MOR library, then the frequency for each identified syntactic type identified on the %gra tier. This procedure worked well for English, but we just noticed that there were only two types of complex sentences in the output files for Spanish. 

You mentioned the procedure used for English might work reasonably well for Spanish, so we are curious about the mismatch in the output.

Thanks,
Cynthia

On Wed, Aug 23, 2017 at 4:16 PM, Brian MacWhinney <ma...@cmu.edu> wrote:

Dear Cynthia,

There is a short, medium, and a long answer to this.  The short answer is “no”, but that answer is quite misleading.  The medium answer is that the grammatical relations of Spanish, particularly on the interclausal level, are nearly the same as those for English.  Therefore, the routine for complexity computation presented in the CLAN manual should work nearly correctly for Spanish.  The longer answer is that I could spend time working on this, but not for several weeks.  Actually, that is not a long answer.  Rather, it is a short answer that points to a long timeframe.

 

Best,

 

--Brian MacW

 

From: <cynth...@gmail.com> on behalf of "co...@gwu.edu" <co...@gwu.edu>
Reply-To: ChiBolts <chib...@googlegroups.com>
Date: Wednesday, August 23, 2017 at 2:42 PM
To: ChiBolts <chib...@googlegroups.com>
Subject: Re: Complex syntax

 

Is there a procedure for freq of complex syntax for Spanish? I'm looking for the equivalent of the MEGRASP/freq procedure to count clausal types in Spanish using the %gra tier. 

 

Thanks,

Cynthia Core

On Thu, Aug 3, 2017 at 9:39 PM, Brian MacWhinney <ma...@cmu.edu> wrote:

Yes, this is described in section 7.10.14 of the CLAN manual.

--Brian MacWhinney

On 8/3/17, 6:29 PM, "chib...@googlegroups.com on behalf of Cynthia Core" <chib...@googlegroups.com on behalf of cynth...@gmail.com> wrote:

    Hi -
    Are there any commands that can help me count the number of sentence types - simple/complex and count different types of complex sentences? I see commands to calculate ratios of one two verb utterances and IPSYN, but I am looking for something that is more specific to syntactic diversity and clausal diversity  and that counts the number of complex versus simple sentences.

    Thank you,
    Cynthia Core

    --
    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+unsubscribe@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/0ddf9d6a-1f18-4678-b5ee-d13704204698%40googlegroups.com.
    For more options, visit https://groups.google.com/d/optout.



--
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+unsubscribe@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/4CE8C958-6FC6-409F-ABE4-64DD11B66E9D%40andrew.cmu.edu.
For more options, visit https://groups.google.com/d/optout.



 

--

Cynthia Core, Ph.D.
Assistant Professor
Department of Speech and Hearing Science
The George Washington University
2115 G St., NW
Washington, DC 20052
co...@gwu.edu
202 994 7364

--
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+unsubscribe@googlegroups.com.

To post to this group, send email to chib...@googlegroups.com.

--
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+unsubscribe@googlegroups.com.

To post to this group, send email to chib...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Brian MacWhinney

unread,
Aug 29, 2017, 9:39:45 AM8/29/17
to chib...@googlegroups.com, cynth...@gmail.com

Dear Cynthia,

Apologies for the delay in responding to your second message.  I double-checked and you are right that the training set for MEGRASP in Spanish only has CMOD and COMP structures.  Because the others are not in the training set, they are not going to appear in the output.  This raises three questions. 

1. Are relevant structures miscoded in the training set?  This is certainly possible and it would be great if you could take a look to see if some obvious one are missing.

2. Is the training set just too small and not sufficiently advanced?  This is also possible.  After all, these are structures from young children.  For English, our primary training set is from the personal narratives and story descriptions provided by normal adults doing some of the AphasiaBank tasks.  We have found that use of adult language is better as a training set, even if the target samples to be analyzed are from children.

3.  Does Spanish just not include some of these relations?  This is possible.  However, this could be a result of treating things like “lo que” as CMOD when they are actually functioning as COBJ, for example.

 

-- Brian MacWhinney

    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/0ddf9d6a-1f18-4678-b5ee-d13704204698%40googlegroups.com.
    For more options, visit https://groups.google.com/d/optout.



--
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/4CE8C958-6FC6-409F-ABE4-64DD11B66E9D%40andrew.cmu.edu.
For more options, visit https://groups.google.com/d/optout.



 

--

Cynthia Core, Ph.D.
Assistant Professor
Department of Speech and Hearing Science
The George Washington University
2115 G St., NW
Washington, DC 20052
co...@gwu.edu
202 994 7364

--
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.

--
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/E7847639-1406-444F-9B82-11F36C327A3B%40cmu.edu.


For more options, visit https://groups.google.com/d/optout.



 

--

Cynthia Core, Ph.D.
Assistant Professor
Department of Speech and Hearing Science
The George Washington University
2115 G St., NW
Washington, DC 20052
co...@gwu.edu
202 994 7364

--

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.

Reply all
Reply to author
Forward
0 new messages