MLU counts

[Yonata Levy]

Hi Nina,
I think you are right in that a composite measure will require not just MLU
but other measures. This should begin to answer Phil's query too -
we use MLU as well as percent of sentences longer than 5 when we equate
children as per developmental stage. The latter is suppose to addressed the
problems that central measures usually have.
I think that the aim is to suggest measures that are theory-free and that
can be achieved independent of grammatical analysis. Would you consider
verbs per utterance such a measure?
Yonata.

________________________

Prof. Yonata Levy
Psychology Department
The Hebrew University
Jerusalem, Israel 91905

Phone: 972-2-5883408 (o)      Fax:    972-2-5881159
            972-2-6424957 (h)      e-mail: msyonata at mscc.huji.ac.il


----- Original Message -----
From: "Hyams, Nina" <hyams at humnet.ucla.edu>
To: "'Brian MacWhinney'" <macw at cmu.edu>; "Yonata Levy"
<msyonata at mscc.huji.ac.il>; <info-childes at mail.talkbank.org>
Sent: Wednesday, February 18, 2004 7:50 PM
Subject: RE: MLU counts


> Thank you all for your input on MLU counts. I am aware of all the
> difficulties in using MLU cross-linguistically from my early work on
Italian
> and English, languages with quite different morphological systems. What I
> was interested in now was MLU as a measure within a single language with
> very rich verbal inflection, one that has not been studied to my knowledge
> -- Malagasy.
>
> With respect to the issue of devising a valid cross-linguistic measure, it
> is likely that any such measure will have to consist of several factors.
> This was done by Kamil Ud Deen (University of Hawaii) in his UCLA
> dissertation on acquisition of Nairobi Swahili. Kamil devised a composite
> measure of linguistic development that consisted of 3 factors -- MLU,
verbs
> per utterance (suggested by Virginia Valian), and frequency of
mono-syllabic
> place holders (as per Bottari et al). He ranked his children on each
> individual measure and then on all 3 measures together, which resulted in
a
> similar ranking, leading some credibility to the system. In such a system
> the degree of morphological richness of the language will not weigh as
> heavily. It might be worthwhile to try a system like Kamil's in other
> languages to see whether it is reliable cross-linguistically.
>
> Thanks again for all the advice.
> Nina
>
> -
>