Using census data to test the critical-period hypothesis for second-language acquisition.

In a previous article (Hakuta, Bialystok, & Wiley, 2003), we tested the critical-period hypothesis for second-language learning using 1990 U.S. census data. Our findings failed to support a critical period for second-language acquisition, demonstrating greater evidence for a steady decline throughout the life span. Stevens (2004) questioned these findings, suggesting they might have resulted in part from the sampling and estimation procedures used. In this Commentary, we report additional analyses addressing each issue raised by Stevens, demonstrating the failure in each case to generate more compelling evidence for a critical period for second-language acquisition.