Relationship characteristics, such as attachment and loneliness, are often viewed as universal in nature, sometimes conceptualized as varying only in degree. Thus, scant attention is given to contextual considerations such as culture and gender. Scholars have argued, however, that culture needs to be the focal point of research, rather than an afterthought (Doherty, Hatfield, Thompson, & Choo, 1994; Segall, Lonner, & Berry, 1998; Smith & Bond, 1994).

There is a dearth of research supporting the cross-cultural applicability of attachment theory (van IJzendoorn, 1990). The predominant view, labeled the universality hypothesis, states that attachment bonds develop similarly in a variety of cultural or family contexts (van IJzendoorn & Sagi, 1999). In a review of over 30 samples, van IJzendoorn and Sagi found relatively similar distributions of attachment classifications (i.e., secure, avoidant, anxious) across different cultural environments. The results are supported by other researchers, who report a lack of cross-cultural variability in attachment styles among older adolescents and adults (e.g., Doherty et al., 1994; Moreira et al., 1998; Soares, Fremmer-Bombik, Grossmann, & Silva, 2001). In contrast, You and Malley-Morrison (2000) found that Korean students scored higher on a measure of insecurity than did Caucasian Americans, as also do some minority group samples (e.g., Bliwise, 1999). Cross-cultural comparisons of gender differences in attachment are also lacking, although the universality hypothesis predicts, and researchers suggest, that gender differences should be found cross-culturally (e.g., Brannen & DiTommaso, 2001; Ross & Spinner, 2001).

Another area of research that has received minimal attention from attachment researchers is the cross-cultural effects of different types of relationship factors. According to the universality hypothesis, one would expect cross-cultural similarities in the pattern of associations among such factors. The results of a cross-cultural study of the relationship between attachment security and lowered perceived loneliness tend to support the hypothesis. DiTommaso, Brannen-McNulty, Ross, and Burgess (2003) reported that male university students had higher levels of romantic loneliness, while there were no significant gender differences for either social or family loneliness.

Unlike the attachment construct, however, support for the universality hypothesis for loneliness is more equivocal. For example, in a handful of studies it has been demonstrated that higher levels of loneliness are associated with more Western cultural orientations which emphasize individualism (e.g., Bhogle, 1991; Rokach & Bacanli, 2001; Rokach & Neto, 2000). On the other hand, reports of higher levels of perceived loneliness among Chinese versus Westerners are not unknown (e.g., Anderson, 1999; Goodwin, Cook, & Yung, 2001). Such apparent contradictions may be a result of different measurement approaches, and could be clarified by specifically measuring different types of loneliness. As with the attachment literature, gender has not been a particular focus of cross-cultural loneliness investigations.

Geographical context is a second potential confound in examining both the attachment and loneliness constructs. The term “cross-cultural” is variously defined as different ethnic groups sharing the same environment, or groups from diverse cultures and geographical settings. Visiting students, who spend only brief periods in another culture, represent a convenient comparison sample.

Asian students studying in North America provide an opportunity to investigate how visiting students may differ from home students. A few researchers have examined the relationships between this group and stay-at-home students. For example, Hsu, Hailey, and Range (1986) found that Chinese foreign students experience significantly more social loneliness than do Canadian students. Leung (2001) reported that Chinese students (both migrants and sojourners) experience more loneliness than do migrants from other cultural backgrounds or Australian home students.

Researchers have demonstrated that levels of loneliness and social isolation are key predictors of sojourner adjustment (Ward & Rana-Deuba, 2000; Ward & Searle, 1990). Furthermore, researchers have indicated that the quality of relationships with family, friends and romantic partners is a significant predictor of the degree of loneliness experienced by visiting students (Sam, 2001; Stone-Feinstein, & Ward, 1990; Ward & Kennedy, 1993; Ward & Searle 1990).

While there are no comparable studies from the attachment literature, attachment theory contends that visiting students should not have less secure parental attachment, since parental bonds are viewed as largely free of the influence of one’s current environment (Bowlby, 1969). An alternative view which suggests that attachment may be context-specific has recently been advanced in reports examining romantic attachment (e.g., Baldwin, Keelan, Fehr, Enns, & Koh-Rangarojoo, 1996; Ross & Spinner, 2001). In addition, there is growing evidence that individuals hold differing attachment relationships, and do so concurrently (e.g., Ross & Spinner).

Attachment relationships are believed to fall roughly into three domains: parents (i.e., family, friends, and romantic partners; see e.g., Brannen & DiTommaso, 2001). One might expect that visiting students would score lower on the peer and romantic attachment dimensions since they are arguably further removed from peer networks than are their home student peers. On the other hand, parental attachment security may not be significantly different for these two groups given their heightened focus on peer relationships during the early college years. DiTommaso and Spinner (1993, 1997) proposed a model of loneliness that corresponds to the aforementioned three proposed attachment domains. Thus, it could be argued that visiting students would score higher on all three types of loneliness, due to feelings of isolation from both their family support system and peer social network.

The present study is an attempt to investigate the influence of culture on attachment and loneliness, by examining visiting Chinese students and home students at a small Canadian university. We hypothesized that the Chinese students would have lower reported peer and romantic attachment security, while there would be no differences between the groups on parental attachment. Regarding loneliness, we proposed that Chinese students would have higher levels of all three types of loneliness. In addition, we hypothesized that there would be similar gender differences within both samples; that is higher levels of peer and romantic attachment security and lower levels of romantic loneliness were expected among females of both cultures. In accordance with the universality hypothesis, we proposed that attachment and loneliness would be significantly associated across cultures and genders. In particular, we believed that the complementary types of attachment and loneliness (e.g., parental attachment and family loneliness) would be most strongly associated.

Inventory of Parent and Peer Attachment (IPPA) Armsden & Greenberg, 1987). The IPPA consists of two, 25-item measures of attachment in adolescence, one for each of parent and peer relationships. Each of the scales consists of trust, communication, and alienation subscales. Each item is rated on a 5-point Likert scale. Levels of attachment security are determined by subtracting the alienation scores from the combined trust and communication scores. Reliabilities for the IPPA were high, ranging from .91 (parents) to .93 (peer) across all students.

In the present study, the IPPA was adapted to measure romantic attachment (Brannen & DiTommaso, 2001). The overall alpha for the romantic scale was .93, with a range of .92 (parents) to .94 (romantic) among Canadian students. However, reliabilities were somewhat lower for the Chinese sample: .85 (parents), .88 (romantic), and .89 (peers). Correlations between types of attachment for the Canadian sample were slightly higher than were those reported in our previous research (Brannen & DiTommaso). The parent domain correlated with peer (r = .62, p < .001) and romantic (r = .69, p < .001). The peer and romantic types were also strongly associated with each other (r = .68, p < .001). Although these correlations are very high, other results suggest that they do represent different constructs (Brannen & DiTommaso). The correlations for the Chinese sample were all significant, but more similar to what we found in our previous research; r = .33, p <. 05; r = .41, p <. 01, and r = .38, p <. 01, respectively. Clearly, more research should be conducted to determine the viability of the tripartite model of attachment that we have proposed.

Social and Emotional Loneliness Scale for Adults -Short Form (SELSA-S; DiTommaso, Brannen, & Best, 2004). The SELSA-S is a 15-item measure that uses a 7-point Likert scale to assess perceived loneliness along three dimensions: family, romantic, and social. The SELSA-S possesses good convergent, concurrent, and discriminant validity versus a number of criteria (DiTommaso et al., 2004). For the current study, alphas ranged from .80 (social) to .91 (romantic) for the overall sample. The Canadian sample had alphas of .89 (romantic), .84 (family), and .85 (social). The coefficients had a greater range for the Chinese sample, with a moderate alpha of .68 for social loneliness to a high of .95 for romantic loneliness. Intercorrelations among subscales were found to be significant, but can be considered moderate at best. The correlations between the three types of loneliness for the Canadian sample were r = .05 (ns) for romantic and family loneliness, r = .43 (p < .001) for family and social loneliness, and r = .21 (ns) for social and romantic loneliness. The same correlation between romantic and family loneliness was found for the Chinese sample, while the other two correlations were somewhat stronger. Social and family loneliness were strongly associated, with r = .33 (p < .05), as were romantic and social loneliness (r = .58, p < .01). Given that these different domains assess different forms of the same construct – namely loneliness – these associations were expected.

Our hypotheses for cultural and gender differences in attachment and loneliness were supported. That is, there were no significant differences between the means of the Chinese and Canadian groups for parental attachment security.

Parental attachment was strongly predictive of family loneliness and accounted for 53% of the variance, F(1, 207) = 235.04, p < .001. Next, country of origin was entered, with no significant change in the amount of variance explained over and above parental attachment, F(1, 206) = .053, p = .82. Finally, the interaction of attachment and country of origin did not uniquely predict family loneliness, F(1, 205) = 1.64, p = .20.

Romantic attachment was highly associated with romantic loneliness, F(1, 196) = 51.98, p < .001, with the former explaining 21% of the variance in the latter. Culture also made a significant unique contribution to the prediction of romantic loneliness, F(1, 195) = 7.94, p < .005, accounting for 3% of the variance in romantic loneliness scores. The interaction between culture and attachment did not add to the amount of variance explained over and above the unique contributions of each variable, F(1, 194) = .01, p = .76.

Peer attachment significantly predicted social loneliness, F(1, 215) = 116.67, p < .001, with 35% of the variance in social loneliness explained by peer attachment. Culture made no significant change to the prediction equation, F(1, 214) = 1.20, p = .275. Finally, the interaction between attachment and culture made no significant unique contribution to the prediction of social loneliness, F(1, 213) = 1.74, p = .188. Peer attachment, however, remained a unique predictor, t = -9.84, p < .001.

At a glance, our results may seem to support the position that there are cultural and gender differences in attachment and loneliness, but that attachment overrides these differences when predicting loneliness. It could well be that the visiting Chinese students were experiencing increased loneliness and greater insecurity because they were removed from their peer networks. Further, it could also be that women are generally more secure and less lonely as compared to men.

There is an alternative and equally viable explanation for the pattern of results obtained. The measures of attachment and loneliness used in the current study may better reflect the egocentricity typical of Western individualism (Rokach & Neto, 2000; Rothbaum, Weisz, Pott, Miyake, & Morelli, 2000). Increased emphasis on the individual may limit the measures’ sensitivity among individuals exhibiting a more sociocentric interpersonal orientation, one characteristic of the visiting Chinese participants in the present study. On the other hand, it is generally accepted, however, that women have a more sociocentric orientation than men do, suggesting that the measures employed might not be guilty of an egocentric bias. Clearly, more study into interpersonal orientation of individuals with regard to relationship factors is required before any conclusions can be drawn. Kashima and Hardie (2000) have developed a measure which evaluates levels of individual and collective self-representations which could provide insight into the debate over whether culture or interpersonal orientation accounts for the differences found in the present study.

There are other considerations that further impede any generalization from the results of the present study. First, one cannot infer that the Canadian and Chinese students in the present study necessarily espouse the theorized values of their respective cultures (i.e., egocentric versus sociocentric). As previously suggested, measuring collective and individual orientation at the personal level could provide valuable insight into the applicability of such labels. Second, the gender differences reported here suggest that there may be larger intracultural than cross-cultural differences in attachment and loneliness (van IJzendoorn & Sagi, 1999). Third, researchers using diverse groups must be aware of the implications of using a narrow definition of culture, measured in nonrepre- sentative samples, and then generalizing the results to an entire population.

We offer the results of the present study as evidence of the cross-cultural validity of attachment and loneliness, and the diversity of these constructs among cultures and genders. Further, we suggest that the numerous ways in which our findings can be interpreted as a result of geographical distance, inadequate measures, or cultural orientation may serve as a cautionary tale for other researchers. Chao (2001) has argued that the development of a culturally sensitive model of attachment requires the integration of universal plus culturally specific components. This approach could also be applied to the study of loneliness. Until such an integrated model is developed, we would argue that the results of any cross-cultural studies of attachment and/or loneliness, including the present one, should be interpreted with caution.

Of interest, the present study does highlight the importance of attachment and loneliness as measures of psychological adjustment for all students, regardless of culture. Also, the multiplicity of attachment relationships was supported by these results, and we urge researchers to consider this in future studies. In conclusion, the present results lend credence to the notion that gender and culture are crucial considerations of any research project, and should not be overlooked or considered secondarily.

Anderson, C. A. (1999). Attributional style, depression, and loneliness: A cross-cultural comparison of American and Chinese students. Personality and Social Psychology Bulletin, 25, 482-499.

Armsden, G. C., & Greenberg, M. T. (1987). The inventory of parent and peer attachment: Individual differences and their relationship to psychological well-being in adolescence. Journal of Youth and Adolescence, 16, 427-454.

Baldwin, M. W., Keelan, J. P. R., Fehr, B., Enns, V., & Koh-Rangarojoo, E. (1996). Social-cognitive conceptualization of attachment working models: Availability and accessibility effects. Journal of Personality and Social Psychology, 71, 94-109.

Bhogle, S. (1991). Perception of loneliness as an index of culture and age. Psychological Studies, 36, 174-179.

Bliwise, N. G. (1999). Securing attachment theory’s potential. Feminism and Psychology, 9, 43-52. Bowlby, J. (1969). Attachment and loss: Vol. 1: Attachment. London: Hogarth.

Brannen, C., & DiTommaso, E. (2001). The role of attachment and interpersonal competence on individuation: A mediational approach. Presented at the Annual Meeting of the Canadian Psychological Association, Quebec City, Canada.

Chao, R. (2001). Integrating culture and attachment. American Psychologist, 56, 822-824.

DiTommaso, E., Brannen, C., & Best, L. (2004). Validation of the social and emotional loneliness scale for adults. Educational and Psychological Measurement, 64, 99-119.

DiTommaso, E., Brannen-McNulty, C., Ross, L., & Burgess, M. (2003). Attachment styles, social skills and loneliness in university students. Personality and Individual Differences, 35, 303-312.

DiTommaso, E., & Spinner, B. (1993). The development and initial validation of the Social and Emotional Loneliness Scale for Adults (SELSA). Personality and Individual Differences, 14, 127-134.

DiTommaso, E., & Spinner, B. (1997). Social and emotional loneliness: A reexamination of Weiss’ typology of loneliness. Personality and Individual Differences, 22, 417-427.

Doherty, R. W., Hatfield, E., Thompson, K., & Choo, P. (1994). Cultural and ethnic influences on love and attachment. Personal Relationships, 1, 391-398.

Goodwin, R., Cook, O., & Yung, Y. (2001). Loneliness and life satisfaction in three cultural groups. Personal Relationships, 8, 225-230.

Hsu, L. R., Hailey, B. J., & Range, L. M. (1986). Cultural and emotional components of loneliness and depression. The Journal of Psychology, 121, 61-70.

Kashima, E. S., & Hardie, E. A. (2000). The development and validation of the Relational, Individual, and Collective self-aspects (RIC) Scale. Asian Journal of Social Psychology, 3, 19-48.

Leung, C. (2001). The psychological adaptation of overseas and migrant students in Australia. International Journal of Psychology, 36, 251-259.

Moreira, J. M., Bernardes, S., Andrez, M., Aguiar, P., Moleiro, C., & de Fatima Silva, M. (1998). Social competence, personality and adult attachment styles in a Portuguese sample. Personality and Individual Differences, 24, 565-570.

Rokach, A., & Bacanli, H. (2001). The experience of loneliness in Canada, Argentina, and Turkey. Psychological Studies, 46, 77-87.

Rokach, A., & Neto, F. (2000). Causes of loneliness in adolescence: A cross-cultural study. International Journal of Adolescence and Youth, 8, 65-80.

Ross, L. R., & Spinner, B. (2001). General and specific attachment representations in adulthood: Is there a relationship? Journal of Social and Personal Relationships, 18, 747-766.

Rothbaum, F., Weisz, J., Pott, M., Miyake, K., & Morelli, G. (2000). Attachment and culture. American Psychologist, 55, 1093-1104.

Sam, D. L. (2001). Satisfaction with life among international students: An exploratory study. Social Indicators Research, 53, 315-337.

Segall, M., Lonner, W., & Berry, J. (1998). Cross-cultural psychology as a scholarly discipline: On the flowering of culture in behavioral research. American Psychologist, 53, 1101-1110.

Smith, P. B., & Bond, M. H. (1994). Social psychology across cultures: Analysis and perspectives. Boston: Allyn & Bacon.

Soares, I., Fremmer-Bombik, E., Grossmann, K. E., & Silva, M. C. (2001). Attachment representation in adolescence and adulthood: Exploring some intergenerational and intercultural issues. In P. M. Crittenden & A. H. Claussen (Eds.), The organization of attachment relationships: Maturation, culture, and context. New York: Cambridge University Press.

Stone-Feinstein, B. E., & Ward, C. (1990). Loneliness and psychological adjustment of sojourners: New perspectives on culture shock. In D. M. Deats, D. Munro, & L. Mann (Eds.), Heterogeneity in cross-cultural psychology (pp. 537-457). Lisse: Swets & Zeitlinger.

van IJzendoorn, M. H. (1990). Developments in cross-cultural research on attachment: Some methodological notes. Human Development, 33, 3-9.

van IJzendoorn, M. H., & Sagi, A. (1999). Cross-cultural patterns of attachment: Universal and cultural dimensions. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (pp. 713-734). New York: Guilford.

Ward, C., & Kennedy, A. (1993). Where’s the ‘culture’ in cross-cultural transition? Comparative studies of sojourner adjustment. Journal of Cross-Cultural Psychology, 24, 221-249.

Ward, C., & Rana-Deuba, A. (2000). Home and host culture influences on sojourner adjustment. International Journal of Intercultural Relations, 24, 291-306.

Ward, C., & Searle, W. (1990). The impact of value discrepancies and cultural identity on psychological and socio-cultural adjustment of sojourners. International Journal of Cross-Cultural Relations, 14, 251-263.

You, H. S., & Malley-Morrison K. (2000). Young adult attachment styles and intimate relationships with close friends. Journal of Cross-Cultural Psychology, 31, 528-534.