Repetitions in child production

[Bruno Estigarribia]

Well, the issue is whether you want to err on the side of overestimating
MLU (by excluding repetitions) or underestimating it (by including them).
Part of the problem is that not all repetitions have the same status, or
the same origin. Treating immediate repetitions differently from
non-immediate repetitions is somewhat arbitrary. Non-immediate and
immediate repetitions can both be a sign of perseveration, for example.
Then, how do you want to treat perseverative utterances? If perseveration
is a hallmark of the child's language, then what does it mean to calculate
MLU excluding repetitions? Are we calculating an upper bound on the MLU
instead of the actual average? Is this MLU reflective of child's competence
or just artificial?
You either code your transcript in a way that differentiates repetitions
that should not be counted (e.g. clarification repeats) from those that
should be (e.g. perhaps perseveration) or... I don't know :-) Maybe one can
calculate several MLUs with random exclusions of sentences and you can get
a confidence interval for it. But this would be a poor substitute for
making individual decisions about repeated utterances.
Note finally that repetitions are treated completely differently in the
calculation of type-token ratios. In that measure, keeping repeats is
crucial. Perseverative or repetitous language brings TTRs down... are we
equally concerned?


On Fri, Jun 6, 2014 at 3:32 AM, Brian MacWhinney <macw at cmu.edu> wrote:

> Dear Diane,
>     As noted by Natasha Rakhlin, repetitions are typically excluded from
> MLU.  This is done automatically and by default in the CLAN MLU program,
> using the marking of repetitions with the <text> [/] and <text> [//]
> symbols.  There is also a program called REPEAT that can add these marks if
> they are not present, but it is safer to do this carefully when
> transcribing.
>
> — Brian MacWhinney
>
>
> On Jun 5, 2014, at 6:04 PM, Diane Lillo-Martin <
> diane.lillomartin at gmail.com> wrote:
>
> Dear all,
> I would be very interested in knowing how you treat self-repetitions for
> various types of analyses, especially MLU.
>
> In particular, I'm thinking of both single word and phrase repetitions,
> such as a sequence of repeating the word 'ball' while playing with the
> ball; or repetitions of phrases such as 'want cookie', either immediate or
> with some intervening talk.
>
> I know that immediate repetitions are generally not counted according to
> the criteria established by Brown. I believe that they would be included in
> the CLAN calculation of MLU unless they are specifically tagged and the
> command is set to exclude them.
>
> If you work with MLU or other measures for which it is relevant, how do
> you treat such repetitions?
>
> Many thanks in advance,
> Diane
>
> ----------
> Diane Lillo-Martin
> University of Connecticut
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "Info-CHILDES" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to info-childes+unsubscribe at googlegroups.com.
> To post to this group, send email to info-childes at googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/info-childes/CAG5pRn%3DRhNa9w8o6hSjMm89NKGqOHe969x-5C9OfsDuR1YAevw%40mail.gmail.com
> <https://groups.google.com/d/msgid/info-childes/CAG5pRn%3DRhNa9w8o6hSjMm89NKGqOHe969x-5C9OfsDuR1YAevw%40mail.gmail.com?utm_medium=email&utm_source=footer>
> .
> For more options, visit https://groups.google.com/d/optout.
>
>
>  --
> You received this message because you are subscribed to the Google Groups
> "Info-CHILDES" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to info-childes+unsubscribe at googlegroups.com.
> To post to this group, send email to info-childes at googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/info-childes/FB66A613-71C6-4351-B668-D5E06DAE9DF0%40cmu.edu
> <https://groups.google.com/d/msgid/info-childes/FB66A613-71C6-4351-B668-D5E06DAE9DF0%40cmu.edu?utm_medium=email&utm_source=footer>
> .
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups "Info-CHILDES" group.
To unsubscribe from this group and stop receiving emails from it, send an email to info-childes+unsubscribe at googlegroups.com.
To post to this group, send email to info-childes at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/info-childes/CAMWGojnux3b-FD15dNYCjcm%3DpBPvCSbb-FTF4m6GONOvqmBbaQ%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/info-childes/attachments/20140606/cdc8c944/attachment.htm>