More on Proficiency

The issue of language proficiency is often of central concern in working with a bilingual population and I’ve written about it before.  Here, I want to continue that discussion and invite comments from both research and practical perspectives. So, the questions are:

What is proficiency?

How is it defined (or how should it be defined)?

How should it be measured?

All these questions are interrelated of course. So, what is proficiency? A quick google search of language proficiency yields 3,780,000 hits. Language proficiency +

grammar = 211,000 hits; vocabulary = 401,000 hits; use = 1,1550,000; exposure = 81,500; pronunciation = 63,500

I think that the differences in results among combinations has to do with different ways that proficiency is conceptualized. All of these factors go into proficiency. But, what’s the most important aspect of proficiency? From a very practical perspective I think it has to do with being able to communicate. From a research perspective however it likely depends on the question being posed. So, how proficient is person A on a measure of grammar or vocabulary (or both). What I think needs to be taken into account is that all these things are related but not perfectly. At the American Speech-Language-Hearing Association convention last year we presented data that shows similar but slightly different relationships among these in bilingual Spanish-English speaking preschoolers.

Even vocabulary comprehension vs. vocabulary production shows differences in which language and at which time children shift dominance. One nice example of this are a pair of studies (Kohnert, Bates & Hernandez, 1999; Kohnert & Bates, 2002) I think are highly informative with respect to proficiency. What happens is that Spanish-first-then-English sequential bilingual children show vocabulary gains in both languages over time. Older children score higher in both languages than younger children. Younger children show dominance in Spanish– as expected. Older children are dominance in English. What’s relevant to the issue of proficiency is that children’s cross-over to English dominance occurs at an earlier age in comprehension than compared to production. This means that at any one point in development you could demonstrate dominance in one language when measuring comprehension AND dominance in the other language when measuring production.

What about children who have language impairments? Or adults with aphasia? Or other groups of people with language-based difficulties. There I think that the task becomes even more difficult. Lack of proficiency can come from either lack of exposure or lack of language ability. What we’ve done is to look carefully at age of first exposure, amound of time each language is heard and used  by the child.  This approach isn’t perfect of course, but it’s a step in understanding the context in which the child is learning and using each language relative to how they perform on a given test (in each language).


, , , , ,

  1. #1 by Kathryn Kohnert on March 11, 2009 - 1:11 pm

    I am truly enjoying the wonderful information shared by others on this list and appreciate the opportunity to share a few of my own thoughts on this recent POST.
    I second Elizabeth’s comments on the centrality of “proficiency” to both research and practical considerations in bilingual speakers/learners. I also agree that “proficiency” varies with how it is operationally defined (thanks for the “hit counts” Liz!), what is measured and how (context-based conversation on topic X; receptive vocabulary; speech “accent”), as well as the standard of comparison used (e.g., within-speaker L1-L2 comparison; between-speaker, monolingual-bilingual comparison). As mentioned in previous POSTS, relative and absolute levels of fluency on any given task are also fluid, changing as a function of social circumstances (e.g., educational experiences, immigration, retirement) or even acquired pathology (e.g. stroke resulting in aphasia, cognitive-communication deficits following head injury). A measure of proficiency or ability in each of a bilingual participant’s languages is essential to interpreting findings from empirical studies.

    At the same time, I think an overreliance on proficiency measures can sometimes lead us down the garden path when we are making practical or policy decisions. For example, for bilingual children with suspected developmental speech or language impairments, it is often recommended that a goal of assessment be to determine the child’s relative levels of proficiency in each language—with the idea that this information can inform both the identification of a “difference vs. disorder” and also provide guidance in determining which language should be supported in treatment (e.g., treat in dominant language). Similarly for bilingual adults with acquired aphasia, an emphasis has been placed on determining relative levels of proficiency in each language, with the idea that this can again inform the “which language for treatment” decision.

    I have argued elsewhere (Language Disorders in Bilingual Children and Adults, that although it may be important to determine relative levels of proficiency in each of a bilingual’s languages as part of the process, this in and of itself does not provide answers to diagnostic or treatment planning questions. Independent measures of proficiency (testing L1 and L2 separately) may fail to capture a bilingual individual’s full ability (thus undermining diagnostic decisions in the absence of bilingual normative data). [For more information on this, refer to POST on this website on January 4 –“Testing Bilingual Children”]. In terms of intervention decisions, I would again agree with Elizabeth’s functional view of language and bilingualism and put treatment recommendations for children or adults with language disorders back into their communicative contexts. Specifically, in determining which language(s!) should be supported through treatment, I feel it is important to know which languages are needed (e.g., Spanish AND English; Hmong AND English etc) and for what purposes (literacy, oral communication; communicating with parents, spouse, siblings, colleagues, coaches, peers, health care professionals; to meet educational or vocational needs, to pray or surf the internet etc.). The long-term support of any/all languages needed, independent of their levels of proficiency as measured at a given point in time, would be consistent with this perspective. Poor levels of proficiency in either or both languages does not necessarily mean a person is not bilingual, as evidenced by the many individuals who function in bilingual environments who have aphasia, developmental language impairment, autism etc. So, I guess my long-winded point here is that I think language proficiency is an essential consideration, but that at some point what we do with information about proficiency should, at least for practical purposes, should be tempered by our general views of language and social functioning (as well as the possibilities of multiple languages within the human mind/brain . . . a topic for another POST!).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: