Code Switching Topic: CLAN commands

[Toh An]

Dear all, 

I have some questions on the use of COMBO commands to analyse code switching topics in utterances.

Firstly, I want to do a count of specific tier combinations (the same tier; different codes), in a specific sequence across clusters using the COMBO command.

 

e.g.:

*TE1:  utterance

%xcod: 1.3

%xtsf: 4.8

*CH3: utterance

%xcod: 2.5

%xcsf: 5.4

è  I want to be able to specify the command to count the instances where %xcod:  1.3 is followed by another code on the same tier %xcod:  2.5 but across different clusters.

è To use the command to count the instances of different sequences (for example 1.1à2.5/ 2.6/ 2.7)

 

Secondly, I want to do a count of specific tier combinations (different tiers; different codes), in a specific sequence within clusters using the COMBO command

 

e.g.:

*CH3: utterance

%xcod: 1.7

%xcsf: 5.4

*TE1:  utterance

%xcod: 1.2

%xtsf: 4.2

*CLS:  utterance

%xcod: 1.6

è  I want to be able to specify the command to count the instances where %xcod:  1.2 is followed by another code on a different tier %xtsf:  4.2 within the same clusters.

è To use the command to count the instances of different sequences (for example 1.1à4.1/4.2/4.3/4.4/4.5/4.6/4.7/4.8)

 

 

Note:

As shown above some utterances have two tiers of codes. All utterances have the %xcod tier while some utterances by the speaker *TE1 have the tier %xtsf, and some utterances by the speaker *CH#/CLS have the %xcsf tier.

However, other utterances may not have two tiers, and only have the %xcod tier.

e.g.:

 

*TE1:  utterance

%xcod: 1.2

*CH4: utterance

%xcod: 2.6

*TE1:  utterance

%xcod: 3.2

*TE1:  utterance

%xcod: 3.3

%xtsf:  4.8

 

I have tried using the following COMBO commands but they do not seem to produce results in the CLAN window. Could you provide some suggestions?

Thank you so much.

[Toh An]

These are the COMBO commands which I have used which do not seem to work:

combo +t*TE1 +s"t%xcod: 1.2^*^t%xtsf: 4.5" +t%xcod +t%xtsf

combo +b2 +s"%xcod: 1.3^*^%xcod: 2.5" +t%xcod +t%xcod

[Brian MacWhinney]

Dear Toh An,

    Yes, you could probably analyze these using some form of a COMBO command.  However, the type of code chain structure you seem to be studying might be better analyzed using either the KEYMAP or CHAINS commands.  Why don't you take a look at the description of those in the manual to see if they correspond to what you are trying to do.  

--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/254d4cc7-b995-4c51-807c-b622aaedd79d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Toh An]

Thank you for the recommendations! 

For KEYMAP, the manual says that "On the coding tier, keymap will look only for symbols beginning with the $ sign. All other strings will be ignored." Do I have to code the dependent tiers starting with the $ sign? I tried using the command "keymap +b1.2* +t%xcod  @" but it does not produce any results or output.

[Brian MacWhinney]

Yes, you would need to add the $ sign.

--Brian

To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/1df5ac85-abbd-48fb-a12d-f9a2fa8596ba%40googlegroups.com.

[Toh An]

Thank you Brian!