Exporting freq (Types, Tokens, TTR), MLU, D (Optimum) to Excel

28 views
Skip to first unread message

Toh An

unread,
Nov 29, 2019, 1:08:57 AM11/29/19
to chibolts
Hi,

I have some questions about exporting freq (Types, Tokens, TTR), MLU, D(Optimum) to Excel.

Firstly, for freq, I tried using this formula to export the results to Excel but it gives me the freq counts of all languages rather than only one of the two languages in the transcripts:
 
freq +d2 +t*CHI -t%mor at ID="*target_child*"

How can I modify the above formula so that it exports the freq including Type, Token, and TTR results of a specific language only such as Mandarin instead of all languages? I have two languages - eng and zho in the CHAT transcripts and would like to have only the freq results of Mandarin words.

Secondly, how do I export MLU and D(Optimum) results of only Mandarin language in bilingual CHAT transcripts to Excel?

I tried modifying the formula for freq above to MLU and D(Optimum) but it did not output the results to an Excel sheet just like it did for freq:

mlu +d1 +t*CHI -t%mor at ID="*target_child*"

vocd +d1 +t*CHI -t%mor at ID="*target_child*"

How can I modify the above formulas so that (1) only Mandarin language values are calculated and (2) the results of MLU and vocd can be exported to an Excel sheet?

Thank you.

Leonid Spektor

unread,
Nov 29, 2019, 1:54:36 AM11/29/19
to chib...@googlegroups.com
Hi,

For just one language, assuming you have coded Mandarin utterances with "[- zho]" code, add option +s"[- zho]" to your command lines.

For Excel output you can type command by itself in the Command window and press return key, thenlook in Usage output for option(s) that "output in SPREADSHEET format". MLU command uses +d option and VOCD uses +d3 option for that result.


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/92648405-9065-4694-9479-b1758b907c7b%40googlegroups.com.

Toh An

unread,
Nov 29, 2019, 2:04:36 AM11/29/19
to chibolts
Dear Leonid,

Thank you for the prompt reply!

Let's say I have only coded English utterances with "[- eng] code, do I need to code all the Chinese utterances in the transcripts with [-zho] code? 

Also, on a word level, I have tagged English words that have occurred within Chinese utterances using @s:eng. Is there a formula that can be used to calculate only Mandarin words and exclude all English words from the calculations?

Leonid.

To unsubscribe from this group and stop receiving emails from it, send an email to chib...@googlegroups.com.

Leonid Spektor

unread,
Nov 29, 2019, 2:27:12 AM11/29/19
to chib...@googlegroups.com
If you have only two languages and you have coded only english utterances with "[- eng]", then you can use -s"[- eng]" option to get remaining Mandarin utterances/words. If you have more than one language in your transcripts, then you need to use -s"[- ...]" option for every language that you want to exclude.

To exclude words ending with @s:eng use -s*@s:eng option in addition to other options.


Leonid.

On Nov 29, 2019, at 02:04, Toh An <pops...@gmail.com> wrote:

Dear Leonid,

Thank you for the prompt reply!

Let's say I have only coded English utterances with "[- eng] code, do I need to code all the Chinese utterances in the transcripts with [-zho] code? 

Also, on a word level, I have tagged English words that have occurred within Chinese utterances using @s:eng. Is there a formula that can be used to calculate only Mandarin words and exclude all English words from the calculations?

On Friday, November 29, 2019 at 2:54:36 PM UTC+8, Leonid Spektor wrote:
Hi,

For just one language, assuming you have coded Mandarin utterances with "[- zho]" code, add option +s"[- zho]" to your command lines.

For Excel output you can type command by itself in the Command window and press return key, then look in Usage output for option(s) that "output in SPREADSHEET format". MLU command uses +d option and VOCD uses +d3 option for that result.


Leonid.

On Nov 29, 2019, at 01:08, Toh An <pops...@gmail.com> wrote:

Hi,

I have some questions about exporting freq (Types, Tokens, TTR), MLU, D(Optimum) to Excel.

Firstly, for freq, I tried using this formula to export the results to Excel but it gives me the freq counts of all languages rather than only one of the two languages in the transcripts:
 
freq +d2 +t*CHI -t%mor at ID="*target_child*"

How can I modify the above formula so that it exports the freq including Type, Token, and TTR results of a specific language only such as Mandarin instead of all languages? I have two languages - eng and zho in the CHAT transcripts and would like to have only the freq results of Mandarin words.

Secondly, how do I export MLU and D(Optimum) results of only Mandarin language in bilingual CHAT transcripts to Excel?

I tried modifying the formula for freq above to MLU and D(Optimum) but it did not output the results to an Excel sheet just like it did for freq:

mlu +d1 +t*CHI -t%mor at ID="*target_child*"

vocd +d1 +t*CHI -t%mor at ID="*target_child*"

How can I modify the above formulas so that (1) only Mandarin language values are calculated and (2) the results of MLU and vocd can be exported to an Excel sheet?

Thank you.

--
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 chib...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/92648405-9065-4694-9479-b1758b907c7b%40googlegroups.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 view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/081a9ce6-b201-4602-836f-152b55ba065c%40googlegroups.com.

Toh An

unread,
Nov 29, 2019, 3:00:17 AM11/29/19
to chibolts
Dear Leonid,

Thank you!


Leonid.

Toh An

unread,
Dec 1, 2019, 8:55:42 AM12/1/19
to chibolts
Dear Leonid,

Is there a way to automate tagging of English utterances with "[- eng]" and English words with "@s:eng" in CLAN or must the tagging be done manually?


Leonid.

Leonid Spektor

unread,
Dec 1, 2019, 10:56:32 AM12/1/19
to chib...@googlegroups.com
The tagging must be done manually.

Leonid~
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/ebc53a15-47a6-41d6-aa91-2d5f11d80640%40googlegroups.com.

Toh An

unread,
Dec 1, 2019, 1:06:59 PM12/1/19
to chibolts
Reply all
Reply to author
Forward
0 new messages