can FREQ create counts for combos of codes on same tier?

[Michelle Barton]

Hello,

I have created a set of 4 $ codes for the child speaker.  I would like to be able to use FREQ to get counts of all combos of two of these 4 schemes.  I can generate the counts of each individual $ code type, and I can generate the counts for combos of codes IF I write out each type of code to count (eg, how often $Info:FACT:specific was paired with $FXN:Build), but I would like to be able to use wildcards to generate all the possible combinations.  I have tried several, but no success (I get 0s for counts).

Is there a way to use wildcards to generate the counts of combinations of codes on the dependent tier?

I have attached my tester transcript and the output from running different attempts to get these counts (the command lines are in the output files, along with my specific comments and questions as I walk through the process I tried).

Thank you,

Michelle

[Leonid Spektor]

Michelle,

If you add +c7 option to freq command with wildcards, then you will get the output you want.

Leonid.

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/956ff7a8-ee45-4ebb-9751-2a1dc3549d38n%40googlegroups.com.

[Michelle Barton]

Hi Leonid,

I have run the +c7 code you suggested, and it has worked to expand the wildcards as below:

> freq +t*CHI +t%xqif  +s"$Info:*:* $FXN*" +c7 anctestv3.cha
freq +t*CHI +t%xqif +s"$Info:*:* $FXN*" +c7 anctestv3.cha
Mon Nov 07 16:46:31 2022
freq (07-Oct-2022) is conducting analyses on:
  ONLY dependent tiers matching: %XQIF;
****************************************
From file <ANCtestv3.cha>
Speaker: *CHI:
  1 $Info:FACT:generic $FXN:Reask
  2 $Info:FACT:specific $FXN:Build
  1 $Info:FACT:specific $FXN:Cont
  1 $Info:FACT:specific $FXN:Reask
------------------------------
    4  Total number of different item types used
    5  Total number of items (tokens)
0.800  Type/Token ratio
    This TTR number was not calculated on the basis of %mor line forms.
    If you want a TTR based on lemmas, run FREQ on the %mor line
    with option: +sm;*,o%


> BUT, this output is only giving pairs of codes that are adjacent to each other,
and is missing to pairs where there is another code in between the two codes I am
using.  These codes are on the dependent tiers for speaker lines 1 and 38:
$Info:EXPLANATION:specific $FXN:Start and $Info:FACT:specific $FXN:start in the transcript I had sent in the original 

email). Both of these lines have $INI codes in between the two codes I am wanting to count.

I saw thought maybe the +c3 switch would work to catch these other codes,
but it did not work as I hoped (it read the 2 extra codes that were missing before, but did not expand the wildcards):

freq +t*CHI +t%xqif  +s"$Info:*:* $FXN*" +c7 +c3 anctestv3.cha
freq +t*CHI +t%xqif +s"$Info:*:* $FXN*" +c7 +c3 anctestv3.cha
Mon Nov 07 16:51:49 2022
freq (07-Oct-2022) is conducting analyses on:
  ONLY dependent tiers matching: %XQIF;
****************************************
From file <ANCtestv3.cha>
Speaker: *CHI:
  7 $Info:*:* $FXN*
------------------------------
    1  Total number of different item types used
    7  Total number of items (tokens)
0.143  Type/Token ratio
    This TTR number was not calculated on the basis of %mor line forms.
    If you want a TTR based on lemmas, run FREQ on the %mor line
    with option: +sm;*,o%


> Is there a way to do both - find the codes regardless of where they are on the tier
AND expand the wildcards?

Thank you,
Michelle

To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/5EBD3F5D-62AB-4A2F-A05C-147194ED88DD%40andrew.cmu.edu.

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/5EBD3F5D-62AB-4A2F-A05C-147194ED88DD%40andrew.cmu.edu.

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/5EBD3F5D-62AB-4A2F-A05C-147194ED88DD%40andrew.cmu.edu.

--

Michelle E. Barton
Associate Professor, Psychology

New College of Florida
Division of Social Sciences
5800 Bay Shore Road
Sarasota, FL 34243
(941) 487-4382
mba...@ncf.edu

[Leonid Spektor]

Michelle,

It appears that +c3 and +c7 options were never implemented to work together. Probably because no one ask for it before.

I have changed FREQ to make +c3 and +c7 options work together as you would expect. Please get new version of CLAN and the command "freq +t*CHI +t%xqif  +s"$Info:*:* $FXN*" +c7 +c3 anctestv3.cha" should work correctly now.

Leonid.

To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/CACy5FBcMVZ5wvBxUoBb1jgi4fuQY1yW%3D3W_A9mv%2B8d-9YT9Npw%40mail.gmail.com.

[Michelle Barton]

Leonid-

Yes, got it!  Thank you so much!  

Michelle

To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/BD9284CD-1E90-4A86-B8FF-4B5FC332F2D2%40andrew.cmu.edu.

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/BD9284CD-1E90-4A86-B8FF-4B5FC332F2D2%40andrew.cmu.edu.