international student stress due in 20 hours

Raquel Faria Chapdelaine, Louise R. Alexitch

Journal of College Student Development, Volume 45, Number 2, March/April 2004, pp. 167-184 (Article)

Published by The Johns Hopkins University Press DOI: 10.1353/csd.2004.0021

For additional information about this article

Access provided by York University (23 Sep 2014 11:15 GMT)

http://muse.jhu.edu/journals/csd/summary/v045/45.2chapdelaine.html

MARCH/APRIL 2004 ◆ VOL 45 NO 2 167

Social Skills Difficulty: Model of Culture Shock for International Graduate Students Raquel Faria Chapdelaine Louise R. Alexitch

This study expanded and tested Furnham and Bochner’s (1982) model of culture shock, employing a sample of 156 male international students in a Canadian uni- versity. Path analysis was used to assess the effects of cultural differences, size of co- national group, family status, cross-cultural experience, and social interaction with hosts on culture shock. Results suggested that social interaction with hosts plays an important role in international students’ adjustment.

Universities and colleges can benefit greatly if they become international communities that contribute to the advancement of knowledge. The outcomes of hosting inter- national students are beneficial for all parties involved as these students contribute to the enrichment of higher education, to the strengthening of relations with various countries in world trade, and to the pro- motion of global understanding (Canadian Bureau for International Education [CBIE], 1977; Cunningham, 1981). Moreover, international students provide additional financial support needed by many uni- versities whose traditional student body is changing. Specifically, in a socioeconomic analysis of the international students residing in the United States during the late 1980s, Paige (1990) estimated that American universities were receiving over 300,000 international students per academic year, and that this number was expected to increase

with the phenomenon of globalization. Paige argued that for smaller colleges and uni- versities in the United States, these inter- national students not only helped maintain high enrolment levels but also provided significant contributions for tuition revenue.

Given the increasing focus on inter- nationalization strategies by many univer- sities (Association of Universities and Colleges of Canada [AUCC], 1998), the literature on international student adjustment has expanded greatly in the past decades. Although varied approaches have been employed to examine international student adjustment, researchers and theorists agree that this population experiences an array of difficulties in their sojourn (Charles & Stewart, 1991; Das, Chow, & Rutherford, 1986; Day & Hajj, 1986; Furnham & Bochner, 1982; Klineberg & Hull, 1979; Schram & Lauver, 1988; Searle & Ward, 1990). We examined international student adjustment difficulties from a culture shock perspective. Specifically, we expanded and tested Furnham and Bochner’s (1982) Social Skills and Culture Learning Model of Culture Shock. According to this model, international students experience adjustment difficulties because they are not aware of the implicit social rules that regulate interaction in the host country. Being unaware of these social rules, international students may lack culture-specific social skills that allow for effective interaction with hosts. As a result, they experience social difficulty in situations

Raquel Faria Chapdelaine is a doctoral candidate in Psychology at the University of Saskatchewan, Canada. Louise R. Alexitch is an Associate Professor of Psychology at the University of Saskatchewan, Canada.

168 Journal of College Student Development

Chapdelaine & Alexitch

that presented no problems in their countries of origins. Thus, in this perspective, culture shock is defined as degree of social difficulty encountered in cross-cultural interaction.

The purpose of this study was twofold: First, it assessed Furnham and Bochner’s (1982) conceptualization of culture shock by determining whether international students experienced a higher degree of social difficulty in Canada than in their countries of origins. Second, it expanded the social skills and culture learning model of culture shock by incorporating into the model additional variables that have been theo- retically linked to social difficulty in cross- cultural interactions. The expanded model of culture shock was subsequently tested through path analysis.

THE MAJOR APPROACHES TO CULTURE SHOCK

The term culture shock was first introduced by anthropologist Kalervo Oberg in the late 1950s. Oberg (1960) defined it as a “disease” suffered by individuals living in a new cultural environment. According to Oberg, culture shock resulted from the loss of well- known cultural signs and symbols, causing individuals to experience anxiety, frustration, and helplessness. The term culture shock has been repeatedly redefined and renamed in the literature. For example, culture shock has been referred to as cross-cultural adjustment (Befus, 1988; Searle & Ward, 1990), culture learning (Paige, 1990), and cultural adjust- ment stress (Anderson, 1994). Despite its multiple definitions, researchers seem to agree that culture shock refers to the multiple demands for adjustment that individuals experience at the cognitive, behavioral, emotional, social, and physiological levels, when they relocate to another culture (Befus;

Brislin, Cushner, Cherrie, & Yong, 1986; Searle & Ward).

Four major approaches have been used to explain the etiology of culture shock. These approaches offer cognitive, be- havioral, phenomenological, as well as psychological and sociocultural explanations of culture shock. Proponents of the cognitive approach to culture shock postulate that cross-cultural adjustment depends upon individuals’ ability to make correct attribu- tions about the cultural values, beliefs, behaviors, and norms of the new society. They ineffectively use their own cultures as the standard for interpreting, judging, and behaving in the new culture (Triandis, 1990). For example, whereas individuals from collectivist societies (e.g., Mexico, Philip- pines) tend to place greater value on behav- iors promoting in-group interdependence and in-group goals, individuals from indi- vidualist societies (e.g., United States, Great Britain) are likely to endorse behaviors related to independence from the in-group and to individual goals (Hofstede, 1991; Triandis). Thus, individuals from collectivist societies may interpret independence from the in-group, for example, as a sign of disrespect for the social group. In contrast, those from individualist societies may interpret the same behavior as a sign of maturity.

Proponents of the behavioral approach (Anderson, 1994) postulate that culture shock occurs because individuals do not know the systems of rewards and punishment associated with the verbal and nonverbal behaviors in the host culture. In other words, behaviors that were positively reinforced in the home country would elicit aversive stimuli in the host country. In an interview with an international student from Iran, the first author was provided with an illustrative

MARCH/APRIL 2004 ◆ VOL 45 NO 2 169

Culture Shock

example of how similar behaviors may elicit different responses across cultures. The interviewee from Iran was encountering difficulties making male friends in Canada. He explained that in his home country male friends are expected to call each other very frequently (almost daily) to demonstrate the strength of their friendship. He started experiencing difficulties in Canada when he engaged in this behavior with a Canadian male that he considered a good friend. The Canadian friend, who did not share the notion that daily phone calls are a sign of a strong male friendship, interpreted the international student’s behavior as very intrusive and almost terminated the friend- ship. Thus, whereas the same behavior elicited positive reinforcement in the inter- national student’s country of origin, it elicited an aversive response in the host country.

According to the phenomenological approach, culture shock is a transitional experience from a state of low self- and cultural awareness to a state of high self- and cultural awareness (Adler, 1975; Bennett, 1986). According to this approach, indi- viduals experience culture shock because they can no longer use their own cultural references to convey and validate central aspects of their identity in the new culture. Consider “politeness” as one of aspect of one’s self-identity. Given that social rules for politeness vary cross-culturally, one may not be able to convey and validate this aspect of self-concept in a different culture in the same way as in one’s own culture. Although self- concepts are comprised of several com- ponents, and a single invalidation incident is unlikely to undermine established identi- ties, repeated episodes may challenge some aspects of one’s self-concept.

Finally, the sociopsychological approach

(Searle & Ward, 1990; Ward & Searle, 1991) indicates that culture shock can be better understood in terms of both (a) affective or psychological adjustment and (b) social adjustment. Whereas the psychological adjustment dimension relates to individuals’ feelings of well-being in the host culture, the social adjustment component refers to individuals’ ability to interact effectively with host members (Ward & Searle). Accord- ing to this theory, the psychological dimen- sion of culture shock can be understood in terms of cultural dissimilarities and of feelings of loneliness in the host country. The social dimension of culture shock can be explained in terms of (a) individuals lacking the appropriate cultural knowledge about the host country and (b) individuals having strong cultural identities that would make them less likely to adapt to the host culture. According to this approach, cross-cultural adjustment would depend upon individuals’ ability to bridge cultural differences between their home countries and the host country, as well as on the extent to which individuals are able to develop meaningful cross- interactions with members of the host culture.

SOCIAL SKILLS AND CULTURE LEARNING MODEL OF CULTURE SHOCK

The expanded social skills and culture learning model of culture shock incorporates the key elements of the previously described approaches. The model is derived from two earlier models: the social skills model and the social skills and culture learning model of culture shock (Furnham & Bochner, 1982). The social skills model was originally developed to explain problems of intra- cultural social inadequacy. That is, the model

170 Journal of College Student Development

Chapdelaine & Alexitch

examined the difficulties of individuals who had not mastered the social rules regulating behaviors in their own cultures. The under- lying assumption is that effective interactions depend upon individuals’ knowledge of social rules and on their ability to perform the appropriate behaviors (i.e., social skills). The model posits that if individuals are not aware of the sets of behaviors considered appropriate by most members of a group in specific social situations (e.g., rules regu- lating nonintimate relationships, rules regulating conflicts), cooperative relation- ships among group members do not follow, and conflict arises (Argyle & Henderson, 1985). Social skills training would consist of identifying individuals’ unique social difficulties and providing them with the knowledge and skills required for effective interactions.

The social skills and culture learning model of culture shock was derived from the social skills model. Furnham and Bochner (1982) argued that the latter could be further developed to account for the adjustment difficulties of sojourners because of the similarities underlying intracultural and cross-cultural social difficulties. The social skills and culture learning model posits that international students lack culture-specific social skills required for effective cross- cultural interactions because they were not socialized in the host culture. As a result, they rely on their own cultural references to interpret, to evaluate, and to behave in the new culture. To the extent that these cultural references are not shared by the host culture and vice-versa, international students experi- ence adjustment difficulties.

Furnham and Bochner (1986) used the analogy of a game to explicate the dynamics of cross-cultural encounters. In a game, players typically follow the same rules. In

cross-cultural encounters, however, players follow different rules for the same game— the game of social interaction. They sus- tained that, because social rules are different, miscommunication often occurs:

From the point of view of the sender, the intended messages may not have reached the receiver, or if they did, they were incomplete, garbled or distorted. From the point of view of the receiver, the messages may have been difficult to interpret, were ambiguous and, in more extreme cases, offensive. And since receivers are also senders, once the spiral of miscommunication has commenced, it can accelerate very quickly into a vicious circle of mis- understanding. (p. 202)

Misunderstandings would, in turn, hinder the effectiveness of cross-cultural encounters. Social situations that presented no problems in international students’ countries of origins (e.g., making friends, making contact with the opposite sex) would suddenly become difficult in the host country. Because interactions with host members comprise a highly important part of the sojourn, social difficulties in cross- cultural interactions are viewed as the essence of culture shock.

Furnham and Bochner’s (1982) theory of culture shock has two major limitations. First, culture-specific social skills cannot be directly measured because there are no clear criteria specifying which social skills are considered useful in different cultures. In their research, no empirical attempt was made to examine the effect of this variable on culture shock. Second, Furnham and Bochner examined culture shock solely as the degree of social difficulty in the new country. Specifically, the test of their theory consisted of comparing the degree of social

MARCH/APRIL 2004 ◆ VOL 45 NO 2 171

Culture Shock

e

_

+

_

_

_

_

Cross-Cultural Differences

Size of Co-National

Group

Family Status in

Host Country

Previous Cross-Cultural

Experience

Social Interaction with Hosts

Culture Shock

e

difficulty experienced by international students living in England to the degree of social difficulty reported by British students. Although results supported their hypothesis, indicating that international students did experience higher degrees of social difficulty than did British students, the authors did not examine additional variables that have been previously hypothesized to influence culture shock such as (a) cross-cultural differences (Church, 1982; Ward & Searle, 1991), (b) size of co-national group in the host country (Gallois, Giles, Jones, Cargile, & Ota, 1995), (c) family status in host country (Chapdelaine & Alexitch, 1995), (d) previous cross-cultural experience (Adler, 1975; Church; Rohrlich & Martin, 1991), and (e) degree of social interaction with hosts (Klineberg & Hull, 1979; Mallinckrodt & Leong, 1992; Ward & Searle; Westwood &

Baker, 1990; Ying & Liese, 1994). The inclusion of these variables in the model of culture shock provides a greater under- standing about how social difficulties occur in cross-cultural interactions. That is, the expanded model not only tests the hypothesis that culture shock can be operationalized as degree of social difficulty in cross-cultural interaction—but it also specifies variables that are expected to influence the degree of social difficulties. Thus, compared to Furnham and Bochner’s original theory, the expanded model of culture shock provides a more detailed theoretical definition of the phenomenon.

Expanding the Social Skills and Culture Learning Model of Culture Shock

In light of the limitations previously de- scribed, the social skills and culture learning

FIGURE 1. Expanded Social Skills and Culture Learning Model of Culture Shock

172 Journal of College Student Development

Chapdelaine & Alexitch

model of culture shock was further devel- oped in this research (see Figure 1).

Cross-Cultural Differences in Social Interaction. To address the difficulties associated with the measurement of culture- specific social skills, cross-cultural differ- ences in social interaction—the precursor of culture-specific social skills—was used in this research as a proxy for the original variable. Cross-cultural differences in social interaction were operationalized as the degree of cultural distance between inter- national students’ countries of origin and the host country. The literature (e.g., Church, 1982; Ward & Searle, 1991) has supported the hypothesis that higher degrees of cultural distance between international students’ countries of origins and their host countries are associated with higher degrees of social difficulty in cross-cultural interactions. The assumption is that the greater the cultural distance between countries, the more socially unskilled international students will be in the host country, and the more social difficulties they will encounter in their cross-cultural interactions.

The expanded model also accounts for an untested hypothesis previously specified by Furnham and Bochner (1982, 1986), stating that cross-cultural differences may have an effect on culture shock through social interaction with hosts. A high degree of cross-cultural differences may hinder the social interaction between international students and host members because of differing values, attitudes, and communi- cation styles. Lower degrees of interaction with host members may decrease the likeli- hood that international students will learn social rules and social skills pertaining to the host culture, thus increasing the degree of social difficulty in their interactions with hosts (i.e., culture shock).

Size of Co-National Group. Size of co-

national group in the host country is hypoth- esized to have an indirect effect on culture shock through social interaction with hosts. According to Gallois et al. (1995), a great amount of adjustment is required in the social interaction of cross-culturally different individuals. International students who belong to large co-national groups in the host country may rely mostly on their co- nationals for social interactions, being less likely to learn culture-specific social skills. This would lead to higher degrees of diffi- culties in cross-cultural interactions.

Family Status in the Host Country. Family status in the host country is hypoth- esized to have an indirect influence on culture shock through degree of social interaction with hosts. Previous research (Chapdelaine & Alexitch, 1995) suggests that international students who come to the host country with their partners and/or children may have less time to interact with members of the host society. As a result, these international students may be less likely to learn culture-specific social skills and, thus, experience higher degrees of social difficulty in their cross-cultural interactions.

Previous Cross-Cultural Experience. Previous cross-cultural experience is hypoth- esized to have a direct effect on culture shock. Specifically, researchers have associ- ated previous cross-cultural experience with lower degrees of culture shock (Adler, 1975; Church 1982; Rohrlich & Martin, 1991). It has been argued that previous cross-cultural experience in countries other than one’s own (including the host country) helps inter- national students realize that their own cultural references may be ineffective in the new culture. Previous cross-cultural experi- ence would facilitate the process of culture learning, ultimately decreasing the degree of culture shock.

Degree of Social Interaction with Hosts.

MARCH/APRIL 2004 ◆ VOL 45 NO 2 173

Culture Shock

Previous research has indicated that social interaction with hosts is crucial to successful cross-cultural adjustment (Klineberg & Hull, 1979; Mallinckrodt & Leong, 1992; Rohrlich & Martin, 1991; Ward & Searle, 1991; Westwood & Barker, 1990; Ying & Liese, 1994). Higher degrees of interaction with hosts are associated with lower levels of cross-cultural difficulties. The rationale is that international students who engage in high degrees of social interaction with members of host culture are more likely to learn the social rules and social skills pertaining to that culture and, therefore, experience less social difficulty in their cross-cultural interactions.

HYPOTHESES

To reiterate, there are two major hypotheses in this study: First, concerning the nature of culture shock, it is hypothesized that, if culture shock is indeed manifested as social difficulty in cross-cultural interaction, then international students should report higher degrees of social difficulty in Canada than in their countries of origin. Second, con- cerning the expanded model of culture shock, the following propositions were derived:

1. Higher levels of social interaction with hosts would be associated with lower levels of culture shock.

TABLE 1.

Sample Demographic Information by Country of Origin

Country of Origin

China India Iran Other Total (n = 31) (n = 29) (n = 22) (n = 74) (N = 156)

Age

< 24 — 05 (17) — 01 (01) 06 (04)

25 – 34 23 (74) 22 (76) 09 (41) 54 (73) 108 (69)

35 – 44 08 (26) 02 (07) 12 (55) 17 (23) 39 (25)

> 45 — — 01 (04) 02 (03) 03 (02)

Marital Status

Single 05 (16) 22 (76) 03 (14) 38 (52) 68 (44)

Married 25 (81) 07 (24) 19 (86) 35 (47) 86 (55)

Separated or Divorced 01 (03) — — 01 (01) 02 (01)

Program of Study

MA 16 (52) 18 (62) 01 (05) 36 (49) 71 (46)

PhD 15 (48) 11 (38) 21 (95) 36 (49) 83 (53)

Postdoctorate — — — 02 (02) 02 (01)

Note. Numbers in parentheses represent the percentage of respondents in the particular group.

174 Journal of College Student Development

Chapdelaine & Alexitch

2. Higher levels of cross-cultural differ- ences in social interaction would be associated with higher levels of culture shock.

3. Higher levels of cross-cultural differ- ences in social interaction would be associated with lower levels of social interaction with hosts.

4. Size of co-national group would be nega- tively correlated with social interaction with hosts. Specifically, international students belonging to larger co-national groups would experience lower levels of social interaction with host members than would international students from smaller co-national groups.

5. Family status in host country would be negatively correlated with social inter- action with hosts. That is, international students accompanied by their partners and/or children would experience lower levels of social interaction with host members than would unaccompanied international students.

6. Previous cross-cultural experience (including previous experience in the host culture) would be associated with lower levels of culture shock.

METHOD

Participants

A total of 195 male international students enrolled in graduate programs at a mid-sized Western Canadian University were available to take part in the study. The largest groups of international students were from China, Iran, and India. Of these 195 international students, 156 (80%) participated in the research. Female international graduate students were not selected for participation because they were underrepresented in this

population at the time of the study. The sample was comprised of students from China (n = 31, 20%), India (n = 29, 19%), Iran (n = 22, 14%), and 35 additional countries (n = 74, 47%). The latter group included students from countries in South and Central America, Europe, Africa, and Asia.

Demographic information for the sample (age, marital status, and program of study) is presented in Table 1. The majority of participants (n = 108, 69%) were between 25 and 34 years. Forty-four percent of partici- pants were single (n = 68) whereas 55% were married (n = 86); only 1% (n = 68) was separated or divorced. Forty-six percent (n = 71) were enrolled in master’s programs and 53% (n = 83) were registered in doctor- ate programs; 1% (n = 2) were enrolled in postdoctorate programs.

Previous cross-cultural experience was defined as the length of time spent in Canada and in countries other than the home country. The large majority of the sample (n = 113, 72%) indicated that, besides their current stay in Canada, they had not had any previous cross-cultural experiences. Only 28% (n = 43) reported having cross-cultural experiences in both Canada and other foreign countries. The median for length of time spent in Canada was 31.50 months (SD = 24.83 months, Mode = 10 months, Range = 8 to 166 months). The median and the mode for length of time spent in other foreign countries was zero (SD = 21.49 months, Range = 0 to 132 months).

Procedures

The International Student Advisor Office and several international student associations helped us advertise the study. The total population of international male graduate students at the university was contacted by

MARCH/APRIL 2004 ◆ VOL 45 NO 2 175

Culture Shock

phone or e-mail about their willingness to participate. Those who volunteered their participation were then asked to take part in group testing sessions, each consisting of 4 or 5 students. The testing sessions lasted approximately 45 minutes and participants were financially compensated for their participation.

Measures

Cross-Cultural Differences in Social Inter- action. Cross-cultural differences in social interaction were operationalized as the degree of cultural distance between Canada and international students’ countries of origin. The close-ended version of the Cultural Distance Index (CDI) (Babiker, Cox, & Miller, 1980) was used to measure the differences in social life between Canada and international students’ countries of origin. Using a scale ranging from 1 (no differences at all) to 5 (very different), participants indicated how different Canada was from their home countries in regards to aspects such as friendship, marriage, educa- tional systems, language, and so forth. Higher scores are associated with higher degrees of cultural distance between the host country and the country of origin. Previous research with international students in New Zealand (Searle & Ward, 1990) indicated that the instrument was reliable, yielding a Cronbach’s alpha of .85.

Size of Co-National Group. Size of co- national group was determined by the number of compatriots studying at the same university. The list specifying the number of students in each nationality category was provided by the International Student Advisor’s Office. The largest co-national groups were comprised of international students from China, Iran, and India.

Family Status. Family status in the host

country was defined as the number of family members accompanying international stu- dents in Canada. International students received one of four possible scores: 0 = they came to Canada alone, 1 = they came to Canada with their spouses (or partners), 2 = they came to Canada with their spouses and their children, and 3 = they came to Canada with their spouses, their children, and their relatives.

Previous Cross-Cultural Experience. Previous cross-cultural experience was defined as the length of time spent in (a) Canada and (b) other foreign countries in months. The score for previous cross- cultural experience was formed by adding the total amount of months spent in Canada to the total amount of months spent in other countries.

The Social Interaction Scale. We devel- oped the Social Interaction Scale (18 items) for the purposes of this study. The items were developed according to a pilot study with international students (Chapdelaine & Alexitch, 1995) in which students identified common types of social interactions with hosts, co-nationals, and other international students in the host country. The scale measures degree of social interaction in Canada, which was defined as the frequency with which international students engage in common social interactions (e.g., “sharing meals,” “going to parties,” “going to mov- ies”). The instrument is comprised of three subscales measuring degree of social inter- action in Canada with (a) Canadians, (b) co- nationals, and (c) international students who do not belong to one’s co-national group. Higher scores are associated with higher degrees of social interaction with the aforementioned groups.

Culture Shock. Culture shock was opera- tionalized as degree of social difficulty in

176 Journal of College Student Development

Chapdelaine & Alexitch

cross-cultural interaction and was assessed through the Revised Social Situations Questionnaire (RSSQ, 35 items). We devised the RSSQ based primarily on Furnham and Bochner’s (1982) Social Situations Ques- tionnaire (SSQ). Specifically, the SSQ was revised because the instrument had been originally employed to assess culture shock in international students living in England in the early 1980s. In a pilot study with international students living in Canada, Chapdelaine and Alexitch (1995) found that students considered some of the original SSQ items to be unclear and that changes in

the wording were needed to make the questionnaire items more relevant for this population. For example, item 4 on the original SSQ, “Going to discotheques or dances” was changed to “Going to clubs or bars” in the RSSQ, because the term “disco- theque” is no longer used and because “dances” presented ambiguity for inter- national students who attended exclusive dances with members of their own co- national groups. In the pilot study, inter- national students indicated that the modified item was clearer and more effective in describing a situation involving a greater

TABLE 2.

Means, Standard Deviations, Ranges, and Reliability Coefficients

Cronbach’s Variables M SD Range N Alpha

Culture Shock Scale (RSSQ)

Social Difficulty in Canada Subscale 90.36 29.52 24 – 168 144 0.91

Social Difficulty in the Country of Origin Subscale 63.02 20.63 20 – 131 142 0.89

Social Interaction in the Host Country Scale

Social Interaction With Canadians Subscale 43.79 11.89 22 – 83 145 0.91

Social Interaction With Co-nationals Subscale 51.13 12.80 18 – 76 143 0.91

Social Interaction With Other International Students Subscale 38.69 11.54 18 – 70 143 0.92

Cross-cultural Differences in Social Interaction Scale (CDI) 38.70 07.74 15 – 53 155 0.78

Size of Co-National Group 33.84 27.87 01 – 71 151 —

Family Status 0.78 0.89 0 – 02 156 —

Previous Cross-Cultural Experience 45.21 32.77 08 – 174 149 —

Note. For the variables “size of co-national group,” “family status,” and “previous cross-cultural experience,” the calculation of Cronbach’s alpha was not applicable.

MARCH/APRIL 2004 ◆ VOL 45 NO 2 177

Culture Shock

degree of interaction between themselves and host members.

The RSSQ is comprised of two sub- scales: Social Difficulty in Canada and Social Difficulty in the Country of Origin. These subscales measure whether inter- national students experience a higher degree of social difficulty in Canada than in their countries of origin (i.e., culture shock). The questionnaire items describe social situations that are potentially difficult for international students (e.g., “getting to know people well”; “making friends”; “understanding jokes, humor, and sarcasm”). Culture shock scores are calculated by subtracting the degree of social difficulty in the country of origin from the degree of social difficulty in Canada. A previous pilot study (Chapdelaine & Alexitch, 1995) employing the RSSQ indi- cated that the instrument subscales were reliable. Specifically, the Cronbach’s alphas obtained for the Social Difficulty in Canada and the Social Difficulty in the Country of Origin subscales were .89 and .84, respectively.

RESULTS

Table 2 shows the means, standard devi- ations, ranges, and Cronbach’s alpha reli- ability coefficients obtained for the scales employed in the study. Reliability analysis indicated that all scales and subscales employed in the study were reliable. To provide a validity check for the Culture Shock Scale, we included in the question- naire booklet the single item, “How difficult is it for you to socially interact with Cana- dians?” Participants were asked to rate their degree of difficulty on a scale ranging from 1 (not difficult at all) to 5 (very difficult). Rating scores from this item were then correlated with culture shock scores, which

were calculated by subtracting degree of social difficulty in the country of origin from degree of social difficulty in Canada. We predicted that culture shock scores would be positively correlated with score ratings from the single item. As expected, the two variables were significantly positively correlated (r = .46; p < .01), indicating that higher degrees of culture shock were associ- ated with higher degrees of overall difficulty in the interaction with Canadians.

One of the major hypotheses of this study posited that culture shock is manifested as social difficulty in cross-cultural inter- action. To test this proposition, we predicted that international students would report higher degrees of social difficulty in Canada than in their countries of origin. Prior to assessing differences, some item ratings had to be recoded for the subscales measuring social difficulty in Canada and social difficulty in the country of origin. Speci- fically, because the rating scales for both subscales included a “never experienced” category, it would not be possible to assess differences accurately if participants indi- cated that they had experienced social difficulty in one country but not in the other country. When comparing differences, participants’ social difficulty scores for one specific country would be unduly inflated. To avoid this problem, ratings were recoded in the following manner: If participants indicated experiencing social difficulty in specific situations in one country, but having never experienced the same situations in the other country, their rating scores were changed to “never experienced” for both countries.

After recoding, a paired t-test analysis was performed between social difficulty in the host country and social difficulty in the country of origin. Results showed that, as

178 Journal of College Student Development

Chapdelaine & Alexitch

–.44***

–.34**

–.19*

Cross-Cultural Differences

Size of Co-National

Group

Family Status in

Host Country

Previous Cross- Cultural

Experience

Social Interaction with Hosts

Culture Shock

R2 = .29 E = .84

–.28**

R2 = .20 E = .89

*p < .05. **p < .01. ***p < .001.

.28*

–.39*

expected, participants experienced a higher degree of social difficulty in Canada (M = 90.09, SD = 29.62) than in their countries of origin (M = 62.68, SD = 20.57), supporting the hypothesis that international students experienced culture shock, t(139) = 11.99, p < .001. The difference between degree of social difficulty in the host country and degree of social difficulty in the country of origin was used in subsequent analyses as participants’ culture shock scores. The overall social difficulty scale ratings were 2.59 for Canada, and 1.79 for the country of origin indicating that, in general, partici- pants experienced almost no social difficulty in their countries of origin and low levels of social difficulty in Canada.

We performed independent t tests to assess whether program of study, marital status, and age were associated with culture

shock. Regarding program of study, only participants in master’s and doctoral pro- grams were included in the analysis because they comprised 99% of the total sample (n = 154). Results indicated no significant differences in degree of culture shock between participants enrolled in master’s (n = 63; M = 28.33, SD = 24.54) and doc- toral programs (n = 76; M = 26.08, SD = 27.86), t(139) = 0.50, p > .05. In relation to marital status, analyses were performed with only married and single participants because, like program of study, they represented 99% of the sample (n = 154). No significant differences in degree of culture shock were found between married (n = 76; M = 30.13, SD = 25.04) and single participants (n = 64; M = 23.36, SD = 27.54), t(138) = 1.52, p > .05. Finally, the sample was divided into two broad age categories because 94% of the

FIGURE 2. Alternative Social Skills and Culture Learning Model of Culture Shock

MARCH/APRIL 2004 ◆ VOL 45 NO 2 179

Culture Shock

sample (n = 147) ranged from 25 to 44 years of age. The first age category included participants who were 34 years or younger (n = 105; M = 25.44, SD = 25.46) and the second category consisted of participants who were 35 years or older (n = 36; M = 31.53, SD = 28.31). An independent t test revealed no significant differences in degree of culture shock between age groups, t(139) = –1.20, p > .05.

Test of Hypothesized Social Skills Model of Culture Shock

The second major hypothesis of this study relates to the testing of the expanded Social Skills and Culture Learning Model of Culture Shock (see Figure 2). Scale cor- relations and standard deviations were converted to covariance matrices for path analysis using EQS (Bentler, 1995). A number of criteria were used to assess the goodness of fit of the model (Bentler, 1995; Loehlin, 1992). Good fit is indicated when the χ2 statistic is nonsignificant or when the ratio of χ2 to the degrees of freedom is less than 2. However, because the χ2 statistic is sensitive to sample size and to assumptions underlying the χ2distribution, the χ2 may not always provide a reliable fit (Tabachnick & Fidell, 1996). Good fit can be more accur- ately assessed when the root mean square error of approximation (RMSEA) does not exceed .10 and the standardized root mean square residual is less than .05 (SRMR) (Tabachnick & Fidell). Furthermore, models provide a good fit to the data when either the comparative fit index (CFI) (Bentler) or the adjusted goodness-of-fit index (AGFI) (Jörskog & Sörbom, 1996) exceeds .90. These latter indices account for model degrees of freedom and represent the propor- tion of variance explained by the evaluated model.

The feasibility of the hypothesized model was tested using a maximum likeli- hood estimation procedure. Results showed that the proposed model did not provide a good fit for the data: χ2(9, N = 156) = 42.98, p < .001; CFI = 0.670, AGFI = 0.786, RMSEA = 0.175, SRMR = 0.137. These results indicated the presence of some misspecification in the hypothesized model because the fit indices were not acceptable. Post hoc model modifications were per- formed to determine the source of the misspecification. Along with an examination of the standardized and nonstandardized path coefficients and measurement equations, both Wald and LaGrange Multiplier tests (Bentler, 1995) were conducted to guide the deletion and addition of paths, respectively. Two nonsignificant paths were dropped from the model. Specifically, the path from cross- cultural differences to culture shock was dropped, as was the path from previous cross-cultural experience to culture shock. In addition, two covariance paths were added to the model: There was a path linking cross- cultural differences to size of co-national group and another path linking size of co- national group to previous cross-cultural experience. Evaluation of this alternative model provided a significantly improved fit: χ2(9, N = 156) = 12.26, p = .19; CFI = 0.968, AGFI = 0.927, RMSEA = 0.055, SRMR = 0.065. The revised model, with all path coefficients, R2, and error terms is shown in Figure 2.

Four out of 6 hypothesized path co- efficients received empirical support in the study. Specifically, lower levels of social interaction with hosts were associated with of higher levels of culture shock (Ζ = –5.48, p < .001). In addition, higher levels of cross- cultural differences in social interaction were associated with lower levels of social

180 Journal of College Student Development

Chapdelaine & Alexitch

interaction with hosts (Ζ = –3.58, p < .01). Being accompanied by partners and/or children in the host country was linked to lower levels of social interaction with hosts (Ζ = –2.55, p < .05). Finally, belonging to a large co-national group was also related to lower levels of social interaction with hosts (Ζ = –4.32, p < .001).

According to the model tested, social interaction with hosts was the sole significant direct predictor of culture shock accounting for a fair amount of variance (R2 = .20). In addition, cross-cultural differences, size of co-national group, and family status signi- ficantly predicted social interaction with hosts, explaining a reasonable proportion of variance (R2 = .29). Taken together, these results suggest that social interaction with hosts acts as a mediating variable between culture shock and (a) cross-cultural differ- ences in social interaction, (b) size of co- national group, and (c) family status in the host country.

DISCUSSION

The current data support Furnham and Bochner’s (1982) proposition that culture shock occurs in the domain of social en- counters between sojourners and members of the host society. As predicted, inter- national students reported higher degrees of social difficulty in Canada than in their countries of origin. The correlations and path model evaluated in the current study sup- ported many of the hypotheses and expanded upon the findings of previous studies. Specifically, findings revealed that as the degree of cross-cultural differences increases between international students’ countries of origin and the host society, the degree of interaction with hosts decreases. Similarly, results indicated that as the size of co-

national group increases, the degree of interaction with hosts decreases. Taken together, these findings corroborate Gallois et al.’s (1995) proposition that higher amounts of adjustment are required in the interaction of culturally different individuals. Compared to international students reporting lower degrees of cross-cultural differences, international students reporting higher degrees of cross-cultural differences may not feel as comfortable interacting with host members. Likewise, international students belonging to large co-national groups may find it easier to interact with co-nationals. In both cases, the amount of cross-cultural training that international students could receive from host members is limited, thus leading to higher degrees of culture shock.

The hypothesized link between family status in the host country and degree of social interaction with hosts was also supported. International students who came to the host country with their partners and/or children were less likely to interact with host mem- bers. This proposition, which was derived from a pilot study with international graduate students (Chapdelaine & Alexitch, 1995), seems to be quite reasonable if one considers that international graduate students are enrolled in demanding academic programs and have a limited amount of time for socializing. Students who are accompanied by their partners and/or children may want to spend their free time with their families, thus limiting the amount of interaction they have with hosts. Also, families may make additional demands on the time of the international students.

The data supported the hypothesized link between social interaction with hosts and culture shock. As the degree of social interaction with hosts decreases, the degree of culture shock increases. These findings

MARCH/APRIL 2004 ◆ VOL 45 NO 2 181

Culture Shock

support Furnham and Bochner’s (1982) argument that without meaningful social interactions with hosts, international students are unlikely to learn and develop the culture- specific social skills that would enable effective cross-cultural interactions. In addition, results are consistent with previous research associating social interaction with hosts with sojourner adjustment (e.g., Klineberg & Hull, 1979; Mallinckrodt & Leong, 1992; Rohrlich & Martin, 1991; Ward & Searle, 1991; Westwood & Barker, 1990). Social interaction with hosts plays an important role in explaining culture shock as it acts as a mediating variable between culture shock and (a) cross-cultural differ- ences, (b) size of co-national group, and (c) family status in the host country. These findings suggest that by engaging in higher degrees of interaction with hosts, inter- national students will experience lower degrees of culture shock, even if they have highly different cultural backgrounds or have a high number of co-nationals or family members in the host country.

Contrary to expectations, results did not support the hypothesized links between culture shock and cross-cultural differences and previous cross-cultural experience. It was predicted that (a) higher levels of cross- cultural differences would be associated with culture shock, and (b) that previous cross- cultural experience would be negatively correlated with culture shock. There are two potential explanations for these unexpected findings. First, cross-cultural differences and previous cross-cultural experience have been operationalized and measured in distinct manners in different studies. Furnham and Bochner (1982), for example, did not measure cross-cultural differences directly; instead they used a theoretical framework to classify the cultural distance between

international students’ home countries and the British society. Previous cross-cultural experience could be also operationalized as degree of knowledge regarding the values, customs, and norms of the host culture. In this study, the previous cross-cultural experience was solely operationalized as length of time spent in the host country and in other foreign countries, given that we could not find any instruments in the literature measuring the variable. Spending time in the host country may not guarantee that one will be willing, or have the means, to learn the social rules that regulate the host culture. In future studies, the measurement of previous cross-cultural experience should be expanded to include degree of cultural learning. Further, future researchers should also examine how previous cross-cultural experience relates to some of variables examined in this study, such as family status in the host country, size of co-national group, and degree of social interaction with hosts.

The second explanation for the un- expected findings is that, contrary to pre- dictions, both cross-cultural differences and previous cross-cultural experience do not actually have a direct effect on culture shock. If this assertion is true, the alternative model is the most adequate model of culture shock. According to this model, the relationship between these determinants and culture shock would be indirect, being mediated by social interaction with hosts. However, several studies have supported the latter association between the variables. A signi- ficant link between the variables may not have been found because the measure of previous cross-cultural experience used in this study may have been limited in scope.

The alternative model indicated that two covariance paths should be added to the model. Specifically, post hoc analyses

182 Journal of College Student Development

Chapdelaine & Alexitch

indicated that (a) cross-cultural differences should be correlated with size of co-national group and (b) size of co-national group should be linked to previous cross-cultural experiences. One could argue that the correlation between cross-cultural differ- ences and size of co-national group is feasible because higher levels of cross- cultural differences demand greater cross- cultural adjustment (Gallois et al., 1995). Thus, international students from highly different cultural backgrounds may favor to a greater extent social interactions with their own co-nationals—especially if the size of their co-national group is large.

The covariance path between size of co- national group and previous cross-cultural experience is somewhat unclear. Specifically, there is no theoretical formulation that could explain how previous cross-cultural experi- ence (i.e., length of time spent in Canada and in other foreign countries) relates to one’s size of co-national group in the host country. Given Pedhazur’s (1982) recommendation that path models should not be determined by the data—but rather by sound theory— one could argue that the link between size of co-national group and previous cross- cultural experience should be disregarded. On this basis, the variable previous cross- cultural experience could also be deleted from the model. However, given that previ- ous studies (e.g., Adler, 1975; Church, 1982; Rohrlich & Martin, 1991) have supported the link between previous cross-cultural experi- ence and culture shock, the former variable should remain part of the model and future studies should investigate this relationship.

IMPLICATIONS This study has practical value for inter- national student advisors, cross-cultural trainers, and researchers. International

student advisors may use information regarding the patterns of interactions be- tween international students and hosts when planning programs. Given the potential benefits of the interaction between inter- national students and hosts, advisors should promote programs that provide opportunities for such interaction. Advisors should also consider offering cross-cultural training programs for international students and for the university staff members who work with these students (e.g., faculty). These programs should be aimed at increasing participants’ awareness of how different cultural frame- works may cause “well-meaning clashes” in cross-cultural encounters. Finally, cross- cultural trainers and applied researchers may use the revised version of Furnham and Bochner’s (1982) RSSQ as a tool for designing training sessions. Through the RSSQ, these professionals can identify situations that have the potential for most difficulty for students and tailor their programs to address the specified needs.

LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH

One of the limitations of the study concerns the generalization of the results. Findings are restricted to male graduate students, as females were not included in the research because they were underrepresented in the population of international students at the time of the study. Future studies including a more representative sample of international students—including undergraduate inter- national students—are needed to support current findings. Further, the operational definition of culture shock used in the study—social difficulty in cross-cultural interaction—was limited in scope. Future studies should measure the additional aspects of culture shock such as the psychological

MARCH/APRIL 2004 ◆ VOL 45 NO 2 183

Culture Shock

reactions individuals experience as a result of being in a new culture (e.g., depression, loneliness) as well as racism. There are also some limitations to the instrument assessing culture shock in this study—the RSSQ. Despite significant differences between degree of social difficulty in Canada and in the countries of origin, results showed that the overall social difficulty ratings for Canada indicated a small amount of diffi- culty (M = 2.59) compared to almost no dif- ficulty at all in the home country (M = 1.79). Although these results seem reasonable when considering the home country, one would expect higher overall mean ratings for social difficulty in the host country. Perhaps only students experiencing lower levels of culture shock participated in the study. This hypothesis seems reasonable if one considers the literature indicating that international students who have lived in the host culture for longer periods of time (e.g., more than one year) experience less culture shock than those who lived in the host culture for shorter periods of time (Church, 1982). In this study,

the average length of time spent in Canada was approximately 3 years, and only 29 students (18%) had lived in Canada for less than a year.

Concerning the model, further research is also needed to determine whether the hypothesized relationships between culture shock and (a) cross-cultural differences and (b) previous cross-cultural experience indeed exist. In addition, the relationship between cross-cultural differences and size of co- national groups should be further examined. Because error coefficients in the model were high, it is likely additional variables have been omitted. Thus, the adequacy of the alternative models should further investi- gated. Finally, future researchers could also examine how culture shock may affect degree persistence and attainment of inter- national students.

Correspondence concerning this article should be sent to R. Faria Chapdelaine, Department of Psychology, University of Saskatchewan, 9 Campus Drive, Saskatoon, SK S7N 5A5, Canada; raquelc@ shaw.ca

184 Journal of College Student Development

Chapdelaine & Alexitch

Adler, P. S. (1975). The transitional experience: An alternative view of culture shock. Journal of Humanistic Psychology, 15(4), 13-23.

Anderson, L. E. (1994). A new look at an old construct: Cross-cultural adaptation. International Journal of Intercultural Relations, 18(3), 293-328.

Argyle, M., & Henderson, M. (1985). The rules of relationships. In S. Duck & D. Perlman (Eds.), Under- standing personal relationships (pp. 63-85). London: Sage.

Association of Universities and Colleges of Canada. (1998). A warm welcome? Recruitment and admission of international students to Canadian universities: Policies, procedures, and capacity. Ottawa, Canada: AUCC.

Babiker, I. E., Cox, J. L., & Miller, P. M. (1980). The measurement of cultural distance and its relationship to medical consultations, symptomatology and examination performance of overseas students at Edinburgh University. Social Psychiatry, 15, 109-116.

Befus, C. P. (1988). A multilevel treatment approach for culture shock experienced by sojourners. International Journal of Intercultural Relations, 12, 381-400.

Bennett, M. J. (1986). A developmental approach to training for intercultural sensitivity. International Journal of Intercultural Relations, 10, 179-196.

Bentler, P. M. (1995). EQS: Structural equations program manual [Computer program manual]. Encino, CA: Multivariate Statistical Software.

Brislin, R. W., Cushner, K., Cherrie, C., & Yong, M. (1986). Intercultural interactions: A practical guide. Newbury Park, CA: Sage.

Canadian Bureau for International Education. 1977. A question of self-interest: A statement on foreign students in Canada. Ottawa, Canada: Author.

Chapdelaine, R., & Alexitch, L. (1995). Social difficulties encountered by international students. Unpublished manuscript, {University of Saskatchewan, Saskatoon, Canada.

Charles, H., & Stewart, M. A. (1991). Academic advising of international students. Journal of Multicultural Counselling and Development, 19(4), 173-181.

Church, A. T. (1982). Sojourner adjustment. Psychological Bulletin, 91(3), 540-572.

Canadian Bureau for International Education. (1981). The integration of international students on Canadian post- secondary campuses (CBIE Research, No. 1). Ottawa, Canada: Author.

Das, A. K., Chow, S. Y., & Rutherford, B. (1986). The counseling needs of foreign students. International Journal for the Advancement of Counseling, 9, 167-174.

Day, R. C., & Hajj, F. M. (1986). Delivering counseling services to international students: The experience of the American University of Beirut. Journal of College Student Personnel, 353-357.

Furnham, A., & Bochner, S. (1982). Social difficulty in a foreign culture: An empirical analysis of culture shock. In S. Bochner (Ed.), Cultures in contact: Studies in cross- cultural interaction (pp. 161-198). Elmsford, NY: Pergamon Press.

Furnham, A., & Bochner, S. (1986). Culture shock. New York: Methuen & Co.

Gallois, C., Giles, H., Jones, E., Cargile, A. C., & Ota, H. (1995). Accommodating intercultural encounters: Elaborations and extensions. In R. L. Wiseman (Ed.), Intercultural communication theory (pp. 115-147). Thousand Oaks, CA: Sage.

Hofstede, G. (1991). Cultures and organizations: Software of the mind. New York: McGraw-Hill.

Jörskog, K. G., & Sörbom, D. (1996). LISREL 8: A guide to the program and applications [Computer program manual]. Chicago: SPSS.

Klineberg O., & Hull, F. W. (1979). At a foreign university: An international study of adaptation and coping. New York: Praeger

Loehlin, J. C. (1992). Latent variable models: An intro- duction to factor, path, and structural analyses. Hillsdale, NJ: Erlbaum.

Mallinckrodt, B., & Leong, F. T. L. (1992). International graduate students, stress, and social support. Journal of College Student Development, 33, 71-77.

Oberg, K. (1960). Culture shock: Adjustment to new cultural environments. Practical Anthropology, 7, 177-182.

Paige, M. R. (1990). International students: Cross-cultural psychological perspectives. In R. W. Brislin (Ed.), Applied cross-cultural psychology (pp. 161-185). Newbury Park, CA: Sage.

Pedhazur, E. J. (1982). Multiple regression in behavioral research: Explanation and prediction. New York: Holt, Rinehart, & Winston.

Rohrlich, B. F., & Martin, J. N. (1991). Host country and re-entry adjustment of student sojourners. International Journal of Intercultural Relations, 15, 163-182.

Schram, J. L., & Lauver, P. J. (1988). Alienation in international students. Journal of College Student Development, 29, 146-150.

Searle, W., & Ward, C. (1990). The prediction of psycho- logical and socio-cultural adjustment during cross-cultural transitions. International Journal of Intercultural Relations, 14, 449-464.

Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). New York: Harper Collins College.

Triandis, H. C. (1990). Theoretical concepts that are applicable to the analysis of ethnocentrism. In R. W. Brislin (Ed.), Applied cross-cultural psychology (pp. 34-55). Newbury Park, CA: Sage.

Ward, C., & Searle, W. (1991). The impact of value discrepancies and cultural identity on psychological and sociocultural adjustment of sojourners. International Journal of Intercultural Relations, 15, 209-225.

Westwood, M. J., & Barker, M. (1990). Academic achieve- ment and social adaptation among international students: A comparison groups study of the peer-pairing program. International Journal of Intercultural Relations, 14, 251-263.

Ying, Y. W., & Liese, L. H. (1994). The initial adjustment of Taiwanese students to the U.S.: The impact of post- arrival variables. Journal of Cross-Cultural Psychology, 25(4), 466-477.

REFERENCES