TTR differences between EVAL and FREQ

[Leonid Spektor]

If EVAL FREQ_types are different from FREQ Types 166 to 204, but both EVAL FREQ_tokens and FREQ Tokens are the same 596, then I will need a small sample of your data that show the difference and the command lines you used for both EVAL and FREQ to figure this out. Please send those directly to me at spektor at andrew.cmu.edu.


Leonid.

> On Mar 14, 2025, at 15:44, Veronica Fletcher <veronica.n.fletcher at gmail.com> wrote:
> 
> Hi Leonid,
> Thanks for your response. I re-ran the file using FREQ on the %mor tier. This resolved the issue with # Tokens. However, I am still receiving a higher Types count with freq +t%mor (204) compared to EVAL (166). Any thoughts? Updated numbers below.
> 
> FREQ_types (from EVAL): 166
> FREQ_tokens (from EVAL): 596
> Types (from FREQ %mor): 204
> Tokens (from FREQ %mor): 596
> Types (from FREQ): 189
> Tokens (from FREQ): 593
> 
> Veronica
> On Friday, March 14, 2025 at 3:23:20 PM UTC-4 Leonid Spektor wrote:
>> Hi Veronica,
>> 
>> 	EVAL counts words on %mor tier and default FREQ command counts words on speaker tier. If your data has a lot of contractions words, then numbers will be different. For example, word (can't) will counted by FREQ as 1 word. But, because on %mor tier this word is represented as (can) and (not) EVAL will count it as 2 words.
>> 
>> You can see the difference if you run FREQ command on %mor tier. For example, command "freq +t%mor ..."
>> 
>> 
>> Leonid.
>> 
>> 
>>> On Mar 14, 2025, at 15:06, Veronica Fletcher <veronica.... at gmail.com <>> wrote:
>>> 
>> 
>>> Hi all,
>>> Our lab is noticing that we receive different Type and Token output values, depending on whether our transcripts are run through EVAL (which I believe uses FREQ counts to calculate TTR?) versus run directly through the FREQ command.
>>> 
>>> Here is some sample data from Transcript A.  The same file was used for both analyses:
>>> FREQ_types (from EVAL): 166
>>> FREQ_tokens (from EVAL): 596
>>> Types (from FREQ): 189
>>> Tokens (from FREQ): 593
>>> 
>>> What could be accounting for these differences? Apologies if this is a silly question - I could not find anything in the CLAN manual that would explain this.
>>> 
>>> Veronica
>>> The Aphasia Network Lab, Northeastern University
>>> 
>> 
>>> -- 
>>> 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... at googlegroups.com <>.
>>> To view this discussion visit https://groups.google.com/d/msgid/chibolts/5dcaff54-e165-4563-b9de-45a2f1936c8bn%40googlegroups.com <https://groups.google.com/d/msgid/chibolts/5dcaff54-e165-4563-b9de-45a2f1936c8bn%40googlegroups.com?utm_medium=email&utm_source=footer>.
>> 
> 
> 
> -- 
> 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 at googlegroups.com <mailto:chibolts+unsubscribe at googlegroups.com>.
> To view this discussion visit https://groups.google.com/d/msgid/chibolts/760455e1-fe9a-42be-a2f5-a6f5f3cbed98n%40googlegroups.com <https://groups.google.com/d/msgid/chibolts/760455e1-fe9a-42be-a2f5-a6f5f3cbed98n%40googlegroups.com?utm_medium=email&utm_source=footer>.

-- 
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 at googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/chibolts/45A980C9-C2FE-47D7-B332-AEE4F2943FEA%40andrew.cmu.edu.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/chibolts/attachments/20250314/83ccf0aa/attachment.htm>