This section describes conclusions of research findings, recommendations for cross-cultural research and limitations of this study.
Conclusions
This study examined the reliability and validity of the Chinese version of the Cross-Cultural Adaptability Inventory by using a Taiwanese sample. Each research question of the study was restated, followed a short conclusions.
Research question 1: What instruments are used to measure cross-cultural adaptation?
After a review of existing literature, five measures for cultural competence have been found and examined, including The Multicultural Awareness, Knowledge, Skills Survey-Counselor Edition-Revised (MAKSS-CE-R), Multicultural Counseling Knowledge and Awareness Scale (MCKAS), Intercultural Development Inventory (IDI), Multicultural Personality Questionnaire (MPQ), and Cross-Cultural Adaptability Inventory (CCAI).
The MAKSS-CE-R and MCKAS assess the multicultural competence for use with counselors. IDI is intended to measure the orientations toward cultural differences (Hammer, et al., 2003). MPQ and CCAI attempt to assess the cross-cultural effectiveness with the initial developmental stage of beginning with a thorough search for factors related to cross-cultural effectiveness as well as traits/skills for successfully being adapted to another culture in the literature. The MPQ was developed by Van der Zee and Van Oudenhoven in 2000 and 2001, whereas the current version of the CCAI was published in 1992 and has been used on hundreds of participants, especially those attending the cultural training programs (e.g.,Edwards, 1999; Goldstein, 1992;
Levy-Barnett, 1995; Majumdar et al., 1999; Wright, 1997). Thus, the CCAI was chosen for use in this study.
degree of adaptation, the Chinese version of the CCAI was modified the wordings of items in reference to the Chinese context, no inappropriate item dropped or new item added. This study undertook a rigorous translation process by using back translation approach with peer review, expert review and pilot test as Hambleton and Patsula (1998) suggested. They pointed out that the cross-cultural adaptation of the measure should include at least four steps: (a) forward translation; (b) back translation; (c) resolving the discrepancies between the original version and the adapted version and examining cross-cultural equivalence; and (d) Pre-testing the original and adapted versions to check for equivalence. Though this study didn’t pre-test the original version to compare the results with the Chinese version, the rigorous translation process had reduced the item biases that were likely to have occurred in the translation process.
The translation steps taken by this study were beneficial for item equivalence due to three reasons. First, back translation did help identify the discrepancy between the original and back-translated version. Second, the disadvantage of back translation could be offset by the committee approach. Hambelton (1993, cited in Geisinger, 1994) and Van de Vijver and Hambleton (1996) pointed out the disadvantage of back translation was that translators would use wordings which were correctly translated into original ones instead of in reference to cultural context once they knew their work would be back-translated. In this study, in order to enhance the translation quality, a group of graduate students who were bilinguals and in the HR field formed as a committee to not only examine the words and phrases of the translation but also reach the consensus on the appropriateness of the translation. As suggested in the literature, back translation should be combined with the committee approach (Geisinger, 1994;
Van de Vijver & Hambleton, 1996). Finally, the expert review played an essential role in the process of translation. The experts working in the CCAI publisher, who knew well what the items intended to measure, helped clarify some implicit meanings of some items because these implicit meanings could not easily identified even through back translation. Also, three professors in the HR field assisted to evaluate the translation and minimized the researcher’s and the publisher’s bias.
Research question 3: Is the Chinese version of the Cross-Cultural Adaptability Inventory reliable and valid for measuring adaptability?
The Chinese version of the CCAI didn’t have acceptable reliability and validity,
which were discussed separately below:
Reliability
In this study, internal consistency and composite reliability were examined.
Cronbach’s a for the total score of the Chinese version of the CCAI was .91. The reliability coefficients were .86 for Emotional Resilience (ER), .73 for Flexibility/Openness (FO), .80 for Perceptual Acuity (PAC), and .55 for Personal Autonomy (PA). Compared with Kelley and Meyers (1995b)’s and Kraemer and Beckstead’s (2003) results, though the reliability coefficients for the total scores were all passed .70, the reliability coefficients for Personal Autonomy in this study and in Kraemer and Beckstead’s (2003) study were below the standard, .55 and .59 respectively. Internal consistency is referred to the interrelatedness of the items (Schmitt, 1996). Personal Autonomy examines “the extent to which an individual respects others, their value systems, and how pressured a person feels to change in a cross-cultural environment” (Kelley & Meyers, 1995b, p.19). The interrelatedness of three issues included in the smallest scale in the CCAI was low and might need more examination. Cortina (1993) stated that the number of items had an effect on Cronbach’s a. Personal Autonomy which contained 7 items, the smallest scale in the CCAI, might account for the lower reliability coefficient for Personal Autonomy.
The composite reliability takes squared sum of standardized loadings and measurement error variance into consideration in a confirmatory factor analysis (CFA) fashion. The composite reliability coefficients were .87 for Emotional Resilience (ER), .74 for Flexibility/Openness (FO), .78 for Perceptual Acuity (PAC), and .39 for Personal Autonomy (PA). Personal Autonomy (.39) didn’t pass the suggested value, .60 (Bagozzi & Yi, 1988), arisen from extremely high measurement error variance.
The above-mentioned two reliability coefficients both showed that Personal Autonomy needs more examination.
Validity
below .08 which Hu and Bentler (1999) brought up, indicating adequate fit. The results were similar to Davis and Finney’s (2006). In their study, χ2(1169, N= 709)
=7182.4, p<.01; RMSEA= .08; SRMR= .07; and CFI= .70.
While investigating each item carefully, item FO6, FO14, PAC3, PA3, PA5 of which t values were smaller than 1.96 indicated no significant relations between the measured variable and its corresponding factor. Above-mentioned results suggested that the proposed four-factor model of cross-cultural adaptability requires to be modified.
In this study, there are two circumstances which have potential threat to validity.
Each of them was discussed below:
Construct inequivalence
One of the scales in the CCAI, Personal Autonomy, is intended to “deal with personal identify, values, beliefs, and empowerment in the context of unfamiliar environment and different values” (Kelley & Meyers, 1995b, p.19). In the cultural dimensions Hofstede (2001) proposed, Taiwan’s scores for Individualism were lower than America’s scores. It’s doubted that the concept of personal autonomy in the Chinese culture may not be as common as that in the United States. However, the CFA results of this study conducted in Taiwan was consistent with Davis and Finney’s (2006) conducted in the United States. Future studies should continue examine whether the Chinese and the American possess different conceptualizations of cross-cultural adaptability.
Unidentified translation errors
Though this study assumed that the thorough translation steps were conducted, there was no denying that research results was easily affected by unidentified translation errors. To some extent, efforts for reducing translation errors were made in this study. There were 66 questionnaires out of 438 considered invalid (15.1% invalid rate) due to item missing, extreme answers, and the same answer on all the items.
After an examination on the invalid questionnaires with missing values less than two items, five translated items (ER6, ER12, PAC2, PAC3, and PAC6) were regarded as problematic because more than 3 respondents failed to answer these items. Among these items, up to 9 respondents failed to respond to PAC3. The detail information can be found in Table 5.1.
Table 5.1. Problematic Translated Items
Scale Item Missing
Times ER6 If I had to hire several job candidates from a background different
from my own, I feel confident that I could make a good judgment.
如果我必須僱請幾位與我背景不同的應徵者，我有自信能做出 良好的判斷
4
ER12 I rarely get discouraged, even when I work with people who are very different from me.
我很少感到沮喪 即使當我和非常不同的人共事
3
PAC2 I have a realistic perception of how others see me.
我對別人如何看待我有實際的認知
4 PAC3 I am the kind of person who gives people who are different from me
the benefit of the doubt.
面對跟我不一樣的人，我會看他們好的一面
9
PAC6 I pay attention to how people’s cultural differences affect their perceptions of me.
我會留意人們的文化差異是如何影響他們對我的看法
3
Recommendations
This study provided an initial attempt to evaluate the psychometric property of the Chinese version of the CCAI based on a Taiwanese sample. However, the contributions towards the development of this instrument for another language group aren’t complete yet. According to the result of research question 2, the recommendation for cross-cultural research was below:
Utilizing the statistical techniques to enhance item equivalence
In order to reach item equivalence, this study adopted judgmental evaluation on the translation. There were still five translated items regarded as problematic which might result from the translation problem. Van de Vijver and Hambleton (1996) suggested the appropriateness of the translation should provide not only judgmental evidence but also statistical evidence. Three of the better known and useful approaches are logistic regression, the Mantel-Haenszel procedure, and item response theory (Hambleton & Patsula, 1998).
Synthesizing findings of research question 3, the results pointed to some possible problems not only in the construct level but also the item level in the original CCAI.
This study could be a cross-cultural comparable study and the results were consistent with the validation study conducted by Davis and Finney (2006) in the United States.
The recommendations for the examination of the original CCAI were discussed below:
Construct level
The fit indices of GFI (.70); CFI (.67); and NNFI (.66) were smaller than the cutoff and the exploratory factor analysis (EFA) failed to produce an interpretable solution. Compared the scales in the CCAI with the relevant literature, Emotional Resilience, has been seldom discussed in the literature. However, stress management to some degree overlaps with the concept of emotional resilience, focusing on a person’s ability to cope with stress (Black, 1990; Hammer et al., 1978; Stening &
Hammer, 1992; Yamazaki & Keyes, 2004). Flexibility is identified as an important ability for successful adaptation (Arthur & Bennett, 1995; Black, 1990; Hannigan, 1990; Yamazaki & Keyes, 2004). In addition, CCAI have the dimensions of cultural sensitivity (Perceptual Acuity) (Leong, 2007), also referred to as empathy, which is the critical dimension of multicultural competence (Cui & Van Den Berg, 1991;
Ruben, 1976; Hannigan, 1990). In terms of distinct dimensions of cultural competence, CCAI has the dimension of Personal Autonomy, which is less frequently mentioned in the literature.
Item level
Among 49 measured variables, specifically, there is one measured variable which loads on Flexibility/Openness (FO14) and Perceptual Acuity (PAC3), of which t values were smaller than 1.96, showing no significant relations between the measured variable and its corresponding factors. This study additionally examined the model of the unidimentional measurement with a deletion of FO14. The t value of PAC3 (9.07) was larger than 1.96.
PAC3/FO14 I am the kind of person who gives people who are different from me the benefit of the doubt.
面對跟我不一樣的人，我會看他們好的一面
Aside from PAC3 and FO14, t values of FO6, PA3, and PA5 was smaller than 1.96 as well. Besides, these items showed no significant mean difference between low-score group and high-score group at the .05 level. Despite the fact that t values and critical ratio (CR) suggested problematic items in the original CCAI, one should caution that the problems might magnify when translated into Mandarin Chinese.
FO6 Impressing people different from me is more important than being myself with them.
讓他人對我留下印象，比做我自己更重要。
PA3 If my ideas conflicted with those of others who are different from me, I would follow my ideas rather than theirs.
如果我的意見與其他不同於我的人有衝突，我會依照自己的意見，而非 他們的。
PA5 I prefer to decide from my own values, even when those around me have different values.
即使當週遭的人的價值觀不同於我，我仍偏好以自己的價值觀來做決定。
Validating an instrument for another cultural group costs time and efforts. Future studies should continue to revise the Chinese version of Cross-Cultural Adaptability
however, based on their cognition of cross-cultural adaptability. It is valuable to examine whether Taiwanese and U.S. people possess different conceptualizations of cross-cultural adaptability and perform different kinds of behavior associated with these conceptualizations. Inviting cross-cultural specialists in Taiwan to revise the Chinese version of the CCAI is one of the solutions to alleviating construct inequivalence.
Limitations
This study was delimited to the evaluation of the Chinese version Cross-Cultural Adaptability Inventory, exclusive of other measures for cross-cultural competence.
The Chinese version of the CCAI, specifically, was adapted to common used language in Taiwan. There are two limitations in this study which should be pointed out:
1. Nonrandom sampling
Appropriate populations for the instrument (see section of Validity of the Instrument) weren’t easy to access, and not large enough to conduct a random sampling. However, through the purposive sampling technique, the sample reached a level of similarity to two groups of appropriate populations, “people who work in a multicultural or culturally diverse environment” and “people who frequently interact with people with people form other cultures.”
2. Criterion validity
Criterion validity has been difficult to establish in this area of research, as there are no generally accepted criteria for measures for cross-cultural competence.
However, future studies can examine the psychometric quality of Cross-Cultural Adaptability Inventory (CCAI) against an existing instrument for cross-cultural competence.
In conclusion, this study provided additional reliability and validity information on the Cross-Cultural Adaptability Inventory, and also provided non-English speaking samples to improve its generalizability. Based on the results for this study, further studies and endeavor are recommended in order to continue to re-examine and improve the measure for cross-cultural adaptability.
REFERENCES
Anderson, L. E. (1994). A new look at an old construct: Cross-cultural adaptation.
International Journal of Intercultural Relations, 18, 293-328.
Arthur, W. J., & Bennett, W. J. (1995). The international assignee: The relative importance of factors perceived to contribute to success. Personnel Psychology, 48, 99-114.
Avallone, A. J. (1997). Cognitive styles of successful expatriates. Unpublished
doctoral dissertation, Nova Southeastern University, Fort Lauderdale-Davie, FL.
Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models.
Journal of the Academy of Marketing Science, 16, 74-94.
Bentler, P. M., & Bonett, D. G. (1980). Significance test and goodness-of-fit in the analysis of covariance structures. Psychological Bulletin, 88, 588-606.
Black, H. T., & Duhon D. L. (2006). Assessing the impact of business study abroad programs on cultural awareness and personal development. Journal of Education for Business, 81, 140-144.
Black, J. S. (1988). Workrole transitions: A study of American expatriate managers in Japan. Journal of International Business Studies, 19, 277-294.
Black, J. S. (1990). The relationship of personal characteristics with the adjustment of Japanese expatriate managers. Management International Review, 30, 119-134.
Brislin, R. W. (1980). Translation and content analysis if oral and written materials. In H. C. Triandis and J. W. Berry (Eds.), Handbook of Cross-Cultural Psychology:
Methodology (Vol. 2, pp. 389-444). Boston, MA: Allyn and Bacon.
Caster, S. L. (1995). Career development and cross-cultural effectiveness of Ghanaian immigrants in Atlanta. Unpublished doctoral dissertation, Georgia State University, Atlanta, GA.
Choi, H. K. (2000). Preparing Korean missionaries for cross-cultural effectiveness.
Unpublished doctoral dissertation, Asbury Theological Seminary, Wilmore, KY.
Chuprina, L. A. (2001). The relationship between self-directed learning readiness and cross-cultural adaptability in United States expatriate managers. Unpublished doctoral dissertation, The University of Tennessee, Knoxville, TE
Clemens, C. R. (2002). A descriptive study of demographic characteristics and
234-253.
Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78, 98-104.
Cui, G., & Van Den Berg, S. (1991). Testing the construct validity of intercultural effectiveness. International Journal of Intercultural Relations, 15, 227-241.
D’Andrea, M., Daniels, J., & Heck, R. (1991). Evaluating the impact of multicultural training. Journal of Counseling and Development, 70, 143-150.
Davis, S. L., & Finney, S. J. (2006). A factor analytic study of the Cross-Cultural Adaptability Inventory. Educational and Psychological
Measurement, 66, 318-330.
Edwards, B. J. (1999). The effect of multicultural training on the cross-cultural adaptability of college students. Unpublished doctoral dissertation, Nova Southeastern University, Fort Lauderdale-Davie, FL.
Fan, X., Thompson, B., & Wang, L. (1999). Effects of sample size, estimation methods, and model specification on structural equation modeling fit indexes.
Structural Equation Modeling, 6, 56-83
Farh, J. L., Cannella, A. A., & Lee, C. (2006). Approaches to scale development in Chinese management research. Management and Organization Review, 2, 301-318.
Flamini, B. (2005). The influence of multicultural training on the cross-cultural adaptability of university business students.Unpublished doctoral dissertation, Nova Southeastern University, Fort Lauderdale-Davie, FL.
Gelles, R. S. (1996). Expatriate adjustment and performance, and spouse adjustment.
Unpublished doctoral dissertation, United States International University, San Diego, CA.
Geisinger, K. F. (1994). Cross-cultural normative assessment: translation ad adaptation issues influencing the normative interpretation of assessment instrument. Psychological Assessment, 6, 304-312.
George, M. E. (1991). Cultural diversity in public elementary schools: An examination of principals' cultural adaptability and student achievement.
Unpublished doctoral dissertation, University of La Verne, La Verne, CA.
Goldstein, D. L. (1992). A comparison of the effects of experimental training on sojourners’ cross-cultural adaptability. Unpublished doctoral dissertation, Florida International University, Miami, FL.
Goldstein, D. L., & Smith, D. H. (1999). The analysis of the effects of experiential training on sojourners cross-cultural adaptability. International Journal of Intercultural Relations, 23, 157-173.
Hambleton, R. K., & Patsula, L. (1998). Adapting tests for use in multiple languages and cultures. Social Indicators Research, 45, 153-171.
Hammer, M. R, & Bennett, M. J. (1998, 2002). The Intercultural Development Inventory. Portland, OR: The IDI Corporation.
Hammer, M. R., Bennett, M. J., & Wiseman, R. (2003). Measuring intercultural sensitivity: The intercultural development inventory. International Journal of Intercultural Relations, 27, 421-443.
Hammer, M. R., Gudykunst, W. B., & Wiseman, R. L. (1978). Dimensions of intercultural effectiveness: an exploratory study, International Journal of Intercultural Relations, 2, 382-393.
Hannigan, T. P. (1990). Traits, attitudes, and skills that are related to intercultural effectiveness and their implications for cross-cultural training: A review of the literature. International Journal of Intercultural Relations, 14, 89-111.
Haslberger, A. (2005). Facets and dimensions of cross-cultural adaptation: Refining the tools. Personnel Review, 34, 85-109.
Hoelter, J. W. (1983). The analysis of covariance structures: Goodness-of-fit indices.
Sociological Methods and Research, 11, 325-344.
Hofstede, G. H. (2001).Culture's consequences : Comparing values, behaviors, institutions, and organizations across nations(2nd ed.). Thousand Oaks, CA:
Sage Publications
Holder-Ballard, C. B. (2006).Cultural competence in dental hygiene students.
Unpublished doctoral dissertation, The University of Memphis, Memphis, TN.
Hwang, F. M. (2004, January). The investigation on the issues of evaluation of overall fit indexes in structural equation modeling. Paper presented at the Meeting of Development of Structural Equation Modeling 2004, Fu Jen Catholic University, TW.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55
Hughes, K. E. (2003).The effect of the International Business Institute study abroad program on cross-cultural adaptibility among international business students.
Unpublished doctoral dissertation, Capella University, Minneapolis, MN.
Minneapolis, MN: National Computer System.
Kelley, C., & Meyers, J. (1995b). The Cross-Cultural Adaptability Inventory manual.
Minneapolis, MN: National Computer System.
Kim, B. S. K., Cartwright, B. Y., Asay, P. A., & D’Andrea, M. J. (2003). A revision of the Multicultural Awareness, Knowledge, and Skills Survey-Counselor Edition.
Kim, B. S. K., Cartwright, B. Y., Asay, P. A., & D’Andrea, M. J. (2003). A revision of the Multicultural Awareness, Knowledge, and Skills Survey-Counselor Edition.