can bilinguals "replace" monolinguals in experimental data collection?

[Ambridge, Ben]

I think it depends exactly what the question is. Obviously anything where you are trying to collect norms or compare across different languages is out, but if you're looking at a very specific self-contained question within a particular language, I don't think it would be that big a problem.

For example, suppose you wanted to investigate whether there's a correlation between the frequency of particular irregular English past-tense forms in some corpus and children's correct production of these forms (versus over-regularization errors). Suppose you found this pattern in English-dominant bilinguals. Of course, the other languages may well affect overall rates of correct performance/error and add some noise, but it would be very difficult to argue that this confound was responsible for the pattern of results and hence that the same pattern wouldn't be observed for monolinguals. Indeed, as long as you have access to some monolinguals, you could show that the same pattern holds for both (if it does).

Having said that, I'm sure Caroline is right that the study would be very difficult to publish - if only because reviewers don't always think through whether or not any apparent confound is actually relevant to whatever claim is being made.

Ben

From: info-childes at googlegroups.com [mailto:info-childes at googlegroups.com] On Behalf Of Rowland, Caroline
Sent: 14 September 2011 09:26
To: info-childes at googlegroups.com
Subject: RE: can bilinguals "replace" monolinguals in experimental data collection?

Dear Susannah,

I agree with Marilyn that the solution is not to raise the age-level - the differences between bilingual and monolingual verb and grammar acquisition are too complex for this to be a workable solution. As far as I'm aware, there is no consensus about which (if any) aspects of syntax and morphology  bilinguals develop more slowly. In theory, you could probably overcome some of the problems by finding a reliable way to assess how much of each language the children hear (Caroline Floccia at Plymouth here in the UK has a good scale for this), controlling for what the "other language" is and then doing fancy statistics to take account of these in regression analyses.  However, I think you will have a more important problem in practice which is that you may find it virtually impossible to publish your findings.  I can see reviewers raising a whole raft of objections to the idea of treating bilingual acquisition as equivalent to monolingual acquisition in this way.  Sorry.

What about embracing your population and trying to  replicate some of the current influential monolingual studies with bilinguals?  Because many current theories of acquisition see some role for input/exposure - though they differ in the exact role they attribute to exposure - they have interesting implications for bilingual acquisition. There you would be at a distinct advantage because you could look at things like amount of exposure to each language as part of your predictions, rather than as a confound.

Caro

From: info-childes at googlegroups.com [mailto:info-childes at googlegroups.com] On Behalf Of marilyn vihman
Sent: 14 September 2011 08:06
To: info-childes at googlegroups.com
Subject: Re: can bilinguals "replace" monolinguals in experimental data collection?

Dear Susannah,

I am curious as to what your thinking is with regards to the age of your bilingual participants: You want to look at verb learning, and you assume that you need to raise your age-level by one or even two years to observe the same phenomenon in bilingual children? In other words, you expect a 1-2 year delay in level of grammatical knowledge or...? I don't work in this area (3-4-yr old grammar), so others will have to respond, but I don't see any reason to raise the age level at all. Bilingual children have somewhat smaller lexicons in each language but their progress is not notably smaller, as far as I know or have observed.

Another question is: Will you be looking only at English, since I imagine the bilinguals in Vancouver speak a wide range of different languages, which you presumably don't have the resources to investigate? Then I think the main thing will be to be sure that English is the dominant language, since a language that is NOT the primary one for a child, i.e., not the one used with the most regularity, is much more unpredictable in terms of how the grammar develops (some areas can be expected to show lags or heavy influence from the stronger lg.). In contrast, I suspect that learning advances in the dominant language are not significantly different from those of monolingual children in that language, even though there is always some influence from the other language - but in a group of children whose 2nd language is not the same, the differences would cancel out in a sense - i.e., whatever patterns you see, should be comparable to English-only learners. Maybe not all reviewers of your study would see it this way - but after all, bilingualism is really the norm, so it seems odd to have to avoid studying bilinguals to gain knowledge about lg. dev. generally!

I'm sure others will have something more substantial to say, but the question is so interesting, I had to respond!

-marilyn

On 14 Sep 2011, at 03:41, Susannah Kirby wrote:

Dear Info-Childes community,

I have a research conundrum, and I'm hoping you can assess one possible solution to it that I've come up with.

I have been investing verb-learning in monolingual children, but in my current location (Vancouver, BC), monolingual children are nearly impossible to find!  On the other hand, bilingual kids are extremely easy to recruit.

I'm wondering how methodologically unsound it would be to allow bilingual children to participate (not mixed in with monolinguals, but as their own participant group), and then to recruit slightly older children. So for instance, my target age range for monolinguals is 3-4 years old; for bilinguals, I might use 4-5 (or even 5-6) year olds. I would also ask for parents to estimate what percentage of the day the kids hear English input, and shoot for, say, a 50%+ range.

Is this solution too problematic to even try? I can see reasons why it might or might not work, but I'm almost at the end of my rope, in terms of my recruitment problems.

Thanks in advance for any insight and suggestions you can offer!

Best,
Susannah Kirby
SFU Linguistics


--
You received this message because you are subscribed to the Google Groups "Info-CHILDES" group.
To post to this group, send email to info-childes at googlegroups.com<mailto:info-childes at googlegroups.com>.
To unsubscribe from this group, send email to info-childes+unsubscribe at googlegroups.com<mailto:info-childes+unsubscribe at googlegroups.com>.
For more options, visit this group at http://groups.google.com/group/info-childes?hl=en.

Marilyn M. Vihman
Professor, Language and Linguistic Science
V/C/210, 2nd Floor, Block C, Vanbrugh College
University of York
Heslington
York YO10 5DD
tel 01904 433612
fax 01904 432673




--
You received this message because you are subscribed to the Google Groups "Info-CHILDES" group.
To post to this group, send email to info-childes at googlegroups.com.
To unsubscribe from this group, send email to info-childes+unsubscribe at googlegroups.com.
For more options, visit this group at http://groups.google.com/group/info-childes?hl=en.
--
You received this message because you are subscribed to the Google Groups "Info-CHILDES" group.
To post to this group, send email to info-childes at googlegroups.com.
To unsubscribe from this group, send email to info-childes+unsubscribe at googlegroups.com.
For more options, visit this group at http://groups.google.com/group/info-childes?hl=en.

-- 
You received this message because you are subscribed to the Google Groups "Info-CHILDES" group.
To post to this group, send email to info-childes at googlegroups.com.
To unsubscribe from this group, send email to info-childes+unsubscribe at googlegroups.com.
For more options, visit this group at http://groups.google.com/group/info-childes?hl=en.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/info-childes/attachments/20110914/190f348a/attachment.htm>