This study attempts to obtain clear evidence for a sensitive period for acculturation. Cheung, Chudek & Heine study Hong Kong immigrants in Vancouver, all from a broad range of age of immigration and length spent in Canada. The study aims to disentangle the duration of exposure to a host culture and age of immigration - variables which were largely confounded in previous studies and resulted in alternative hypotheses. Cheung et al. measured acculturation by accessing respondent's identification with their mainstream culture and heritage culture on the Vancouver Index of Acculturation. Results showed that the younger participants were at the time of immigration, the quicker they were to identify with the Canadian culture. Moreover, mainstream identification among younger immigrants increased with their duration in Canada. However, neither of these variable, age of immigration and length of time spent in Canada, predicted participants identification with their heritage culture. This evidenced shows that identification with heritage culture and identification with mainstream culture are independent. Cheung et al. concludes that there is evidence for a sensitive period of acculturation: Younger immigrants who are exposed to a host culture for a longer duration are better able to identify with it (Cheung et al., 2011). To strengthen the study's results, a least squares model was used to control for covariates such as gender of participant and the language of questionnaire they completed. This covariate adjustment through a least square regression adjusted for baseline differences that could have occurred by chance and affected the interpretation of results. Instead of adjusting the covariate through a regression, administering a block randomization could further improve precision. For example, equal proportions of female and male who have spent the same number of years in
Canada. Since the sample size of the study is small, precision gains from blocking would be large. Though performing covariate adjustment on variables like gender of participants could strengthen the results by eliminating variation in the outcome, Cheung et al. did not justify the reason behind choosing those specific covariates; variables that affected the results may have been chosen. Moreover, Cheung et al. should examine other potential covariates such as participant's level of education. A major weakness of this study is its low external validity. As Cheung et al., the high concentration of Chinese immigrants in Vancouver and "concomitant cultural amenities" could have explained the observed results and limit the extent to which results can be generalized to other populations and circumstances (Cheung et al., 2011). To improve external validity, researchers could sample immigrants in Vancouver who come from a heritage culture that is less prevalent in Vancouver, such as Japanese participants. The study also mentions that the negative relationship between duration Canada and mainstream identification could have been a result of a cohort effect such as the older immigrants' experiences with the Cultural Revolution (Cheung et al., 2011). The experiences in a host culture is a variable that could be extended to all participants in the study; bad experiences in a host culture could make participants