RELY

[Brian MacWhinney]

Sophie,
    You can work on the two participants separately by creating a file for each.  Copy the master file to a second file and use the TRIM command to remove one participant from each.  For your second question, we would need your file along with the command you are using. You could send that to me or Leonid (spektor at andrew.cmu.edu <mailto:spektor at andrew.cmu.edu>).  One file that illustrates the problem should be enough.

--Brian MacWhinney

> On Jul 28, 2020, at 3:08 PM, Sophie Ambrose <sophie at sophieambrose.com> wrote:
> 
> I'm helping one of our labs new to using CLAN conduct reliability analyses on their coding tier using RELY. (They have converted word documents with their transcripts to CLAN files and are adding coding tiers.) They'd like to have Cohen's Kappa and ideally separately for the two participants. 
> Questions:
> 1) Is the only way to get reliability data separately for each speaker to convert the transcript into two separate transcripts where each only contains one speaker? 
> 2) I'm using rely +t%xcod +dm1 +c -t* +d20 @ to try to get Cohen's kappa. There are 38 errors and 237 matches, but the output says Cohen's kappa coefficient: -1.#IND and I'm not sure what to make of that.  
> I'd appreciate any help! I have tried attaching sample files and am having trouble, so I may try posting those as a reply.
> Thank you,
> Sophie
> 
> -- 
> 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 on the web visit https://groups.google.com/d/msgid/chibolts/cb079174-197d-4734-9b34-526c181a6ddcn%40googlegroups.com <https://groups.google.com/d/msgid/chibolts/cb079174-197d-4734-9b34-526c181a6ddcn%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 on the web visit https://groups.google.com/d/msgid/chibolts/CD2150C9-39AC-412C-8FD4-75D1A7686DD2%40andrew.cmu.edu.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/chibolts/attachments/20200728/1302931d/attachment.htm>