VOCD manual

Amber ambalashes at gmail.com
Mon Aug 24 21:21:11 UTC 2015


Thanks for these references - really helpful discussion here. I'm going to 
use MATTR - simply because my samples are all quite small. Interestingly, I 
had CLAN compute voc-d first - and when I ran my statistical tests for any 
difference in vocd between the two datasets, there was a significant 
difference (a very small effect size). However, one dataset contained 
samples that were on average larger (i.e., more overall talk) than the 
other, and having read the references mentioned here I think the 
significant difference was likely spurious (because of the sensitivity of 
vocd to sample size - and small samples generally plus consistently 
(slightly) larger samples in the one dataset). I've had CLAN compute MATTR 
- the difference is not significant with this measure - and I feel more 
confident about its reliability for my small datasets.

Best wishes
Amber 

On Thursday, 20 August 2015 14:31:03 UTC+1, Amanda Owen Van Horne wrote:
>
> Also see 
>
> Owen, A. J., & Leonard, L. B. (2002). Lexical Diversity in the Spontaneous 
> Speech of Children With Specific Language ImpairmentApplication of D.*Journal 
> of Speech, Language, and Hearing Research*, *45*(5), 927-937.  
> Bottom line is that VOCD is better than TTR, but less transparent.  It 
> does still vary with sample size, but to a lesser degree. 
>
> McCarthy, P. M., & Jarvis, S. (2010). MTLD, vocd-D, and HD-D: A validation 
> study of sophisticated approaches to lexical diversity assessment. *Behavior 
> research methods*, *42*(2), 381-392.
> Compares various methods of computing D.  Recommends  a variety of 
> measures. 
>
> Koizumi, R. (2012). Relationships between text length and lexical 
> diversity measures: can we use short texts of less than 100 tokens. *Vocabulary 
> Learning and Instruction*, *1*(1), 60-69.
> Recommends MLTD
>
>
>
> On Wednesday, August 19, 2015 at 9:46:58 AM UTC-5, Amber wrote:
>>
>> In fact, perhaps I should just post my question on here. The CHAT manual 
>> talks about calculating D and states that it can be done on samples as 
>> small as 50 tokens. However, reliability is less good for small samples. 
>> Could anyone advise me on whether VOCD is still a more valid measure than 
>> TTR, even for small transcripts (e.g., during a shared-book reading 
>> session)? I assume that as long as the sample is large enough for KIDEVAL 
>> to be able to give me VOCD values then it is OK?
>> Thanks
>> Amber
>>
>> On Wednesday, 19 August 2015 15:36:05 UTC+1, Amber wrote:
>>>
>>> Dear all
>>>
>>> The CLAN manual contains the following link, which no longer appears to 
>>> work. 
>>>
>>> http://childes.psy.cmu.edu/manuals/vocd.doc
>>>
>>> Is there a new link to the VOCD manual/another similar article? 
>>>
>>> Many thanks
>>> Amber
>>>
>>>

-- 
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 post to this group, send email to chibolts at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/ef344fea-c3ce-41d1-8897-0c9d2bd33e6e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/chibolts/attachments/20150824/8fa957f4/attachment.htm>


More information about the Chibolts mailing list