Given the purpose of observation, the effects of culture should also be considered in the development of rating scale formats. A number of studies have dealt with issues related to the structure, complexity, and subjectivity of these rating formats (e.g., Hartel, 2002, p19). Hartel demonstrated that different scale formats elicit different judgment processes and affect rating acceptability and effectiveness among different observers. However, Hartel neglected to Lest for the effects of cultural differences. We assert that the observer's culture, including his or her language and way of thinking, may affect the acceptability and effectiveness of the scale formats.
Consequently, a number of approaches were proposed for this selection and development. For example, Kenny and Albright (2005, p79) identified several approaches, including self-report measures and third-person (expert) judgments. A fundamental assumption underlying these approaches is that a universal criterion can always be found to assess observational accuracy for a given observation target. Few people have questioned this assumption since the research on observational accuracy began several decades ago. However, given the purpose of the observation (i.e., collecting data for cross-cultural studies or for decisions in international management), this assumption becomes questionable. A universal criterion may not be found for all cases, especially for observation-based data involving observers' subjective judgments. An example will illustrate this problem. (Abel 1998 p94)
In the aforementioned scenario, should it be concluded that one cultural group possesses higher observational accuracy than the other? For example, should one simply reject the Asian observers' scores as poor because they lead to completely different results? For observations conducted in cross-cultural studies and for decisions being made in an international management context, it may be more interesting and useful to consider the results of both groups. The same logic is true for the external validity of the findings based on observational research. When conducted for similar purposes (collecting data for cross-cultural studies or for decisions in international management), any consideration of external validity should also include the recognition of cultural differences.
Given the purpose of the observation (i.e., collecting data for cross-cultural studies or for decisions in international management), the effects of cultural differences should be considered in the entire process of the observational study, that is, before and after the actual observation takes place. To support this argument and to provide a specific illustration of the need to consider cultural differences, we conducted a laboratory study to examine the effects of cultural differences on observers' ratings. Specifically, we examined differences between Asians and Caucasians in their observations of individual behaviour. We believed that differences in cultural orientations would ultimately generate differences in judgment making. A brief review of the literature regarding key cultural differences between Caucasians and Asians is presented prior to the presentation of this study. (Allport 1933 p77)
Discussion
The findings support the argument that the effects of culture should be considered at two stages: both before and after observation is conducted. First, a consideration of culture is important for the appropriate selection of observers, whether this selection is ultimately for the purpose of research data collection or for such applied uses as performance appraisal. Selecting observers from only a single ethnic group may not always be appropriate. In the international management context, for example, if Caucasians are selected to assess the behaviour of those employees who will be sent to the Asian countries, the observers may ignore some behaviour that Asian people find undesirable. As a result, employees selected by the Caucasians may not be successful, based on behaviour and style relevant to that organizational context.
On the other hand, an individual whose cultural orientation reflects a lower sensitivity to the observed behaviour may provide a lower score for the presence of this behaviour observed in a target individual. For example, compared with people of Western cultures, those of Asian cultures are more likely to be characterized as being obedient and working submissively, as noted in our earlier discussion of cultural differences. As we expected, the scores generated by Asian observers were significantly lower than those of the Western observers on this type of behavioural descriptor. (Ambady 2004 p31)
Conclusion
The results of this study also suggest that differences in culture can impede effective interaction among managers from different cultures. This implication is important for practitioners working in an international management context, as well as within culturally diverse organizations. For example, when employees are evaluated, the cultural or behavioural preferences of observers from different countries or cultures may lead to different assessment results, even though these observers may have actually observed the same behaviour. Biased data may be collected from the observations, and this bias can affect the quality of management decisions. (Erez 2003 p85)
Finally, this study has some limitations because we did not measure each of the cultural values directly. Without these measurements, the relationships between the cultural values and the scores from the observers were difficult to confirm. In other words, the next stage is to collect a more comprehensive set of empirical data that more directly ties cultural differences to observational differences. Also, it should be pointed out that, at present, our knowledge regarding the effects of cultural differences on observation and judgment making is very limited. This important issue merits further research attention. Continuing attempts must be made to capture the complexity of the relationship between cultural differences among observers and their observation or judgment making.
References
Abel, T. M., & Hsu, F. L. K. (1998). Some aspects of personality of Chinese as revealed by the Rorschach Test. Rorschach Research Exchange and Journal of Projective Techniques, 13, 285–301.
Allport, G. W., & Vernon, P. E. (1933). Studies in expressive movement. New York: Haffner.
Ambady, N., & Rosenthal, R. (2004). Thin slices of expressive behaviour as predictors of interpersonal consequences: A meta-analysis. Psychological Bulletin, 111, 256–274.
Betancourt, H., & Lopez, S. R. (2002). The study of culture, ethnicity, and race in American psychology. American Psychologist, 48, 629–637.
Erez, M. (2003). Toward a model of cross-cultural industrial and organizational psychology. In H. C. Triandis, M. D. Dunnette, & L. M. Hough (Eds.), Handbook of industrial and organizational psychology (Vol. 4, pp. 559–607). Palo Alto, CA: Consulting Psychologists Press.
Hartel, C. E. (2002). Rating format research revisited: Format effectiveness and acceptability depend on rater characteristics. Journal of Applied Psychology, 78, 212–217.
Hui, C., & Triandis, H. (1996). Individualism-collectivism: A study of cross-cultural research. Journal of Cross-Cultural Psychology, 17, 225–248.
Kenny, D., & Albright, L. (2005). Accuracy in interpersonal perception: A social relations analysis. Psychological Bulletin, 102, 390–402.
Pepitone, A., & Triandis, H. (2005). On the universality of social psychological theories. Journal of Cross-Cultural Psychology, 18, 471–498.
Triandis, H. (1999). The self and social behaviour in differing cultural contents. Psychology Review 96, 506–520.
Triandis, H. C., McCusker, C., & Hui, C. H. (1995). Multimethod probes of individualism and collectivism. Journal of Personality and Social Psychology, 59, 1006–1020.