search

langcog / web-cdi

7 stars 5 forks source link

clinical reports for English short forms and CDI III #489

Closed vmarchman closed 7 months ago

vmarchman commented 7 months ago

@HenryMehta Here are the clinical report templates for:

English L1 English L2A English L2B English CDI III

Each of these would have two versions, one for the Download Selected Data --> Download Clinical pdfs and one with the "adjusted benchmarks" for the Download Clinical pdfs with adjusted benchmark option. EnglishL2BShortFormClinicalReport.docx EnglishL2AShortFormClinicalReport.docx EnglishL1ShortFormClinicalReport.docx EnglishCDI3ClinicalReport.docx

HenryMehta commented 7 months ago

@vmarchman I have done L1, loaded to dev and tested. I have a concern. It looks to me like L1 scoring is including Understands in Understands and Says. I think is should be the other way around. But this might just be my test data. Please could you check the clinical form is correct and the that it is scoring correctly. Thanks

vmarchman commented 7 months ago

Hi @Henry @.***> You are right. The score for "understands and produces" should be only the ones marked with "produces". The score for "Understands" should be the SUM of the words marked as "understands" and "produces". The score for "understands and produces" is usually bigger than the one for "produces".

Also, can we delete columns "children_comforted" to "kindergarten_dpw" from the outputs?

Thanks!

-vm

On Wed, Nov 22, 2023 at 3:38 AM Henry Mehta @.***> wrote:

@vmarchman https://github.com/vmarchman I have done L1, loaded to dev and tested. I have a concern. It looks to me like L1 scoring is including Understands in Understands and Says. I think is should be the other way around. But this might just be my test data. Please could you check the clinical form is correct and the that it is scoring correctly. Thanks

— Reply to this email directly, view it on GitHub https://github.com/langcog/web-cdi/issues/489#issuecomment-1822608018, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB2TUTC3J4DL3NG2IPNKQ4LYFXP2PAVCNFSM6AAAAAA7RHW6A6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMRSGYYDQMBRHA . You are receiving this because you were mentioned.Message ID: @.***>

HenryMehta commented 7 months ago

@vmarchman I understand you've got Thanksgiving so I'm not expecting a response, but all of these are now ready to test in dev. The L1 scoring will update overnight tonight.

vmarchman commented 7 months ago

Hi @HenryMehta These look good.

But, please delete columns children_comforted to child_ask_words_say from outputs.

Thanks!

HenryMehta commented 7 months ago

@vmarchman I cannot remove these fields from the download. They have been added for other researchers and the download doesn't currently work on the fields used for a specific study but all fields available. I could rewrite the download to be study specific but would take some time and I would suggest done as a separate issue