Acculturation has played a significant role in counseling psychology research (Yoon et al., 2011). Acculturation has been researched as a variable of interest for the subpopulation of Asian international students with particular attention paid to important outcomes, such as help-seeking attitudes (Li, Wong, & Toth, 2013, Yakunina & Weigold, 2011). However, there have been systematic issues in how acculturation has been measured in this subpopulation. These issues include: 1) construct underrepresentation, in which acculturation has been evaluated using English fluency as a proxy variable for the construct (e.g. Poyrazli et al., 2004); 2) not assessing important aspects of acculturation, or not assessing enculturation at all (e.g. Jung et al., 2007); and 3) measures being designed on the premise of outdated theories (e.g. Zhang & Dixon, 2003). Based on these widespread issues in the literature, the Zhang and Moradi (2013) acculturation scale was selected as a potentially useful measure that had not yet been empirically validated for use with this population. The Zhang and Moradi (2013) scale accounts for both acculturation and enculturation as distinct constructs, and contains subscales of language-related behavior, cultural pride, and cultural knowledge for both acculturation and enculturation.
The Zhang and Moradi (2013) instrument was evaluated using five structural equation models. The first was a single-factor model suggested by Kline (2011) to serve as a baseline. Further theory-driven confirmatory factor analyses (CFAs) included a bilinear, unidimensional model, a six-factor model, a hierarchical model, and a bifactor model. The Zhang and Moradi (2013) scale was also evaluated for convergent and discriminant validity, and exploratory analyses with traditional Asian values and life satisfaction were conducted.
A total of 283 participants were recruited using multiple data collection methodologies, and there was a significant attrition rate, which resulted in a final sample of 131 participants. Results of the CFAs indicated poor to marginal fit for all structural equation models conducted. The six-factor model evidenced the best fit to the data in comparison to the other models run based on the size of the chi-square statistic and a change in chi-square statistic test. Despite the superior fit, the six-factor model evidenced poor to marginal fit to the data, even after modifications allowed four error terms to covary ([¿2(3985) = 7936.31, p < .001], CFI =.582, RMSEA = .087 (90% CI = .085, .090), SRMR = .106). The reliability of these factors, based on internal consistency coefficients, was found to be adequate.
Results of convergent and discriminant validity were mixed. Convergent validity was generally found to reflect the hypothesized directions of the bivariate correlations and reached statistical significance, but only two of the correlations met the a priori r = .50 cutoff. Discriminant validity was also mixed, with most of the hypothesized relationships being supported, but small differences in language acculturation based on country of origin, and small relationships between acculturation total and subscale scores and social desirability, were also found. In terms of exploratory analyses, small correlations were found between the Zhang and Moradi (2013) total and subscale scores and adherence to Asian values, and no significant relationships were found with life satisfaction.
Implications for counseling and future research are discussed. This study was primarily limited by the small sample size and attrition, and may best be used as a pilot or exploratory study for this important area of counseling psychology research. Future research may build on the findings with a larger sample in order to provide further evidence of the scale’s factor structure, validity, and reliability.