Discrepancy between KIDEVAL and VOCD statistics

[Nick Riches]

Hi.

I'm posting a query on behalf of a student who has noted a high VOCD, calculated using KIDEVAL. The child is aged 4;2. The VOCD score in KIDEVAL is around 85. However, when the VOCD program is used, the score falls to 59. I've reproduced the commands below

kideval @ +LinkAge +leng +t*CHI: +dfp~"4;-4;5"
vocd  +t*CHI @

The KIDEVAL VOCD is close to the population mean, suggesting that it's being calculated in a consistent manner. However, I'm struggling to understand the discrepancy with the score calculated by the VOCD program.

NB I would like to upload the manuscript itself, but the student project has quite strict confidentiality rules.

Thanks

Nick

[Leonid Spektor]

Nick,

    I am afraid I can't replicate your observation using my sample file. My VOCD command result for  "D_optimum average:  97.70" and KIDEVAL result for "VOCD D_optimum_average" is 97.74. I am going to send to you directly the sample file and command lines I used, so that you can confirm my findings. If you get different results, then see if your version of CLAN is older then the latest version on talkbank.org/clan web site. If you get similar results, then I am afraid I will not be able to replicate and identify the reason for discrepancy between KIDEVAL and VOCD values.

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 post to this group, send email to chib...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/92cdc6cf-1564-4450-8fa3-561d471a78d1%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.