MLU counts

[Yonata Levy]

Dear Nina,
We have struggled with the issue of MLU for Hebrew - a morphologically
complex language-  and here are a few of the more general principles which
were adopted: -
I am not stating here language specific characteristics but there are also
some general things which I think would be relevant to most languages even
if their typology is unlike that of the Semitic languages. The problem is
how not to have inflated counts, since you want to be able to compare
children cross-linguistically and how not to give credit for rote learnt
morphologically complex forms.
Here is what we do:
1. MLU of a single utterance can never be more than 9
2. MLU of a single word cannot be more than 2
3. If the language has grammatical gender which is extensively marked on
various parts of speech, the count varies according to the sex of the child
since a girl is addressed in the feminine and a boy in the mas. and thus,
early on they encounter different forms with different frequency. a few
specific decisions are required with respect to such forms.
4. Inflected pronouns are counted as 1
5. First person forms are counted as 1

I hope this helps!
Yonata.
***********************************************
Prof. Yonata Levy
Psychology Department
The Hebrew University
Jerusalem, Israel 91905

Phone: 972-2-5883408 (w)
            972-2-6424957 (h)
Fax:      972-2-5881159
----- Original Message -----
From: "Hyams, Nina" <hyams at humnet.ucla.edu>
To: <info-childes at mail.talkbank.org>
Sent: Wednesday, February 18, 2004 1:05 AM
Subject: MLU counts


> Can anyone point me to a set of guidelines/instructions on calculating MLU
> in morphologically rich languages.
>
> Thanks,
> Nina Hyams
>