Research Article  |   June 2011
How “Proficient” Is Proficient? Subjective Proficiency as a Predictor of Bilingual Listeners’ Recognition of English Words
Author Notes
  • Correspondence to Lu-Feng Shi: lu.shi@liu.edu
  • Editor: Sheila Pratt
    Editor: Sheila Pratt×
  • Associate Editor: Jean-Pierre Gagne
    Associate Editor: Jean-Pierre Gagne×
Article Information
Special Populations / Cultural & Linguistic Diversity
Research Article   |   June 2011
How “Proficient” Is Proficient? Subjective Proficiency as a Predictor of Bilingual Listeners’ Recognition of English Words
American Journal of Audiology, June 2011, Vol. 20, 19-32. doi:10.1044/1059-0889(2011/10-0013)
History: Received March 19, 2010 , Revised July 1, 2010 , Accepted February 2, 2011
American Journal of Audiology, June 2011, Vol. 20, 19-32. doi:10.1044/1059-0889(2011/10-0013)
History: Received March 19, 2010; Revised July 1, 2010; Accepted February 2, 2011
Web of Science® Times Cited: 3

Purpose: English proficiency must be considered when a bilingual individual is to be evaluated clinically with English speech material. This study describes the minimum level of self-reported English proficiency that identifies bilingual individuals who may perform on par with monolingual listeners on an English word recognition test.

Method: A total of 125 normal hearing bilingual listeners rated their English proficiency in listening, speaking, and reading on an 11-point scale. Other related linguistic variables were also obtained. A randomly selected Northwestern University Auditory Test No. 6 (NU-6) list (50 English monosyllabic words) was presented to all participants at 45 dB HL in quiet.

Results: Over 90% of the listeners self-rated to have at least “good” proficiency in English listening, speaking, or reading. Of these participants, more than 30% did not achieve a monolingual normative level in English as delimited by binomial distribution. Composite proficiency ratings across language domains better predicted word recognition performance than self-ratings for listening proficiency only. Combining language dominance and age of English acquisition with proficiency ratings further improved prediction specificity.

Conclusions: Self-rated English proficiency can predict bilingual listeners’ performance on the NU-6 test. For desirable sensitivity and specificity in predicting monolingual-like performance, a minimum rating of 8 out of 10 across all language domains is recommended.

Acknowledgments
I wish to thank the volunteers for participating in this study. I also thank Keren Zahavi for her helpful comments on an early version of this article.
Order a Subscription
Pay Per View
Entire American Journal of Audiology content & archive
24-hour access
This Article
24-hour access