Our findings indicate that cultural orientation is related to drinking behaviors among Chinese university students, similar to the findings of previous studies 
. In addition, new evidence on the influence of cultural orientation on drinking emerged when we applied a bi-dimensional model with four cultural orientation categories. Traditional-oriented individuals were more likely to be occasional drinkers and nondrinkers than marginal-oriented, bicultural-oriented or western-oriented individuals. These findings suggest that a traditional cultural orientation is associated with less drinking and less risky drinking than other orientations. Consist with Herman-Stahl’s finding 
, when controlling for the socio-demographic factors, logistic regression indicated that being either a bicultural or marginal individual increased the odds of being a regular drinker relative to someone with a traditional cultural orientation. However, no significant differences were found between western and traditionally-oriented individuals after demographics were controlled.
Although the mechanism for how cultural orientation may affect patterns of alcohol use is not fully understood, there are some possible explanations. Traditionally oriented students possible drink less because they value traditional Chinese values, which evolved over time to support low risk drinking behavior. Values that did not support low risk drinking would threaten the survival of the society.
Counter intuitively and unlike some prior research 
, students endorsing a Western orientation were not more likely to be either regular or occasional drinkers than students with a traditional orientation. Western cultural values come largely through media, advertising, and foreign visitors. This small window into Western culture tends to emphasize immediate gratification and pleasure, which would support increased alcohol. But, Western values also do not endorse some risky drinking behaviors that are part of traditional Chinese culture, like social and business toasting practices. Our findings also may be affected by using only a frequency measure for drinking. The difficulties with obtaining volume consumption measures in China have been well documented 
. For college students in the West, the predominant risky drinking pattern is high levels of consumptions at infrequent intervals. It may be that the Western oriented students do not drink any more frequently than traditional oriented students but drink more when they drink. This requires further study.
It appears that the two mixed cultural orientations create the highest risk factors for problem drinking. In both orientations, individuals do not have a clear identity. Bicultural individuals who endorse values from both the receiving and traditional cultural contexts may find themselves without a clear value system and able to vary their position depending on the circumstances 
. Marginal individuals who reject the foreign culture and do not value their traditional cultural values lack any firm value system for guiding behavior 
. It has been suggested that due to stress created by the conflicts associated with being marginal or bicultural oriented individuals might be motivated to escape such stress or conflict by consuming more alcohol 
Our findings confirmed gender differences in alcohol use, that is, males tend to be higher frequent drinkers, similar to the findings of Shell and colleagues 
. Much of the variation in gender differences in drinking can also be linked to culture 
. Our investigation found that females were more likely to be traditional-oriented than males. This suggests that gender differences may be mediated through differences in cultural orientation 
Graduate students were more likely to be regular drinkers than undergraduates. A scant amount of literature has shown the specific differences between the two populations. Our findings confirm the findings of Wang and Zhang’s study among university students in Beijing that graduate students were prone to be to drink more 
. Our findings indicate that these differences do not appear to be due to different cultural orientations in the two groups. Further study is needed to examine the origin of these differences.
Students who came from an urban area tended to be higher frequent drinkers than those from town/rural area. Zhou’s study in Hunan Province in China showed that the drinking rates were higher in the urban area (45.9%) than in the rural area (39.6%) 
. It is difficult to specifically assess the reasons for the differences in drinking behaviors between urban and rural population, however, with our findings that students from town/rural area were more likely to be traditional oriented, this is most likely due to a combination of the different cultural background and other complicated factors, such as levels of economic development, working environment, educational background and etc.
Key university attendance was associated with less drinking than general university attendance. Key universities refer to universities recognized as prestigious and which received a high level of support from the central government with different campus culture and model of teaching from general universities. Our findings suggest the type of university attended was associated with cultural orientation. Students in key university were more likely to be western-oriented. Despite this they drank less, possibly the result of the heavy time requirements for academic success that leave little time for recreational activities that might include alcohol use. It may be the case that key students, like students in Western colleges, drink fewer times but drink more each time they drink. This possibility clearly needs further study.
In summary, culture orientation had an important impact on drinking behaviors. The role of gender, hometown and university type on drinking behaviors is partially mediated through the influences of culture orientation. The significance of our findings could be considered as the basis for alcohol policy and program development. These factors that differentially contribute to alcohol use deserve careful and detailed attention in efforts to reduce alcohol drinking. Reinforcing traditional beliefs and values may be beneficial for less drinking. In the meantime, significant changes in the formal policy environment at the university level seem to be indispensable. Specific strategies of preventive intervention would meet specific population, considering the role of gender, hometown, campus environment and graduate students/undergraduates.
Several limitations of this study should be mentioned. First, it should be noted that this study only took place in a single city in China. Whether the findings would generalize to other parts of China is unknown. Second, the study is cross-sectional. Data on drinking patterns over time and cultural orientation changes over time may provide new insights into their relationship. Finally, the continued development of the CCOS is also needed to enhance reliability and validity. At the same time a more detailed analysis of the relationships of drinking and other variables could help in understanding problematic drinking pattern 
. Given these limitations, our findings need to be replicated and refined in future studies. More researches are encouraged to examine the deep relationship between specific aspects of cultural orientation and drinking patterns, determine what protective factors are provided by culture, and investigate marginal and bicultural individuals and the possible role stress plays in drinking behavior.