International Journal of Intercultural Relations

26 (2002) 609–631

Attitudes toward the culturally different:the role of intercultural communicationbarriers, affective responses, consensual

stereotypes, and perceived threat

Julie Spencer-Rodgersa,*, Timothy McGovernb

aDepartment of Psychology, University of California, Berkeley, Berkeley, CA 94720, USAbUniversity of California, Santa Barbara, USA

Abstract

The psychological impact of intercultural communication barriers on intergroup attitudes

was examined by testing a model of global attitudes toward the culturally different. The

prejudice literature has largely overlooked the role of intercultural communication and

intercultural communication in determining people’s evaluative orientation toward ethno-

linguistic outgroups. Intercultural communication emotions (negative affect associated with

perceived linguistic and cultural barriers) were investigated as determinants of prejudice, in

conjunction with causal factors that are widely recognized as central to intergroup judgments

(consensual stereotypes, intergroup anxiety, and realistic and symbolic/cultural threats

[Stephan & Stephan Int. J. Intercultural Relations 20 (1996) 409]). Regression analyses

indicated that intercultural communication emotions were strongly and uniquely related to

prejudice toward a culturally diverse outgroup: foreign students. Consistent with the contact

hypothesis [Allport (1954) The nature of prejudice, Addison-Wesley], moderated regression

analyses indicated that the structure of intergroup attitudes was modified by social contact

with the international community. Implications for intergroup relations and international

educational exchange are discussed.

r 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Intercultural communication; Intergroup relations; Intergroup emotions; Stereotyped atti-

tudes; Prejudice; International students; Foreign students; Contact hypothesis; Intergroup anxiety;

Communication apprehension

*Corresponding author. Tel.: +1-510-527-7194.

E-mail address: rodgers@socrates.berkeley.edu (J. Spencer-Rodgers).

0147-1767/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved.

PII: S 0 1 4 7 - 1 7 6 7 ( 0 2 ) 0 0 0 3 8 - X

1. Introduction

Effective intercultural communication is critical to the establishment andmaintenance of favorable intergroup relations (Dodd, 1995; Gudykunst, 1986; Hall,1976; Kim, 1986; Martin, 1993). Factors that have been identified as central tointercultural communication competence, such as cultural knowledge and aware-ness, communication skills, and tolerance for ambiguity, strongly impact thefavorability of intergroup contacts (Giles & Johnson, 1981; Gudykunst, 1986; Lustig& Koester, 1996). As an area of inquiry, intercultural communication has beenstudied extensively in fields such as cultural anthropology, sociolinguistics, sociologyof language, and communication science (Gudykunst, 1986; Harman & Briggs, 1991;Kim, 1986). Within psychology, research on intercultural communication hascontributed greatly to our understanding of ethnolinguistic identity, languageattitudes, speech accommodation, and the significance of language in stereotyping(Giles & Johnson, 1981; Giles & Robinson, 1990). Intercultural communicationscholars and educators have called for more research on intercultural communica-tion within the discipline of psychology (Harman & Briggs, 1991). Relativelyunexplored within social psychology, for example, is the relationship betweenintercultural communication barriers and intergroup attitudes (Gudykunst &Hammer, 1988; Wiseman, Hammer, & Nishida, 1989).Intercultural communication barriers arise from group differences in cognition

(e.g., fundamental epistemologies, values, norms, etc.), affect (e.g., types and levelsof emotional expressivity), and patterns of behavior (e.g., language, customs,communication styles, etc.). Effective intercultural communication requires cogni-tive, affective, and behavioral (including linguistic) adaptations that can be arduousand troublesome to participants in an intergroup encounter (Dodd, 1995;Gudykunst, 1986; Lustig & Koester, 1996). Several decades of research onintercultural communication points to the relative difficulty of achieving effectiveand satisfying communication between ethnolinguistic outgroups (Dodd, 1995; Hall,1976; Kim, 1986; Martin, 1993). Individuals must meet the challenges of languagebarriers, unfamiliar customs and practices, and cultural variations in verbal and non-verbal communication styles in order to achieve successful intercultural under-standing (Dodd, 1995; Gudykunst & Hammer, 1988; Kim, 1986; Wiseman &Koester, 1993). As a result, linguistic and cultural barriers often carry evaluative andaffective consequences for interactants in an intercultural context.Communication with the culturally different is frequently associated with adverse

emotional responses (Gudykunst & Hammer, 1988; Neuliep & McCroskey, 1997;Stephan & Stephan, 1985; Yook & Albert, 1999). To illustrate, individuals may feelawkward and anxious when interacting with culturally different others (Stephan &Stephan, 1985), in part, because of communication obstacles. Members of adominant ethnolinguistic group may experience feelings of impatience andfrustration when communicating with non-native speakers of a language (Dodd,1995; Giles & Robinson, 1990; Wiseman & Koester, 1993). Although accentedspeech is sometimes viewed as socially attractive, processing accented speech iscognitively and emotionally taxing (Yook & Albert, 1999), and non-native speakers

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631610

of a language are rated less favorable than native speakers on a wide range ofattributes, including competence and trustworthiness (Edwards, 1982). Interculturalencounters may also be confusing due to group differences in emotional expressivityand non-verbal communication styles (Kim, 1986; Gudykunst & Hammer, 1988),and cultural variations in values, norms, and customs may lead to culturalmisunderstandings and instances of communication breakdown that are stressfuland unpleasant (Giles & Robinson, 1990; Gudykunst, 1986; Wiseman & Koester,1993). Ultimately, repeated communication failures and emotionally laden culturalmisunderstandings can give rise to a negative evaluative orientation toward theculturally different.

2. Intercultural communication and prejudice

Since the Cognitive Revolution in social psychology in the 1970s, research onintergroup attitudes has largely emphasized cognitive causal factors (e.g., stereo-types) to the neglect of affective determinants (Eagly & Chaiken, 1993; Stangor &Lange, 1994; Stephan & Stephan, 1985). Social psychologists are increasinglyrecognizing the significance of emotions in determining people’s evaluativeorientation toward outgroups (Dijker, 1987; Esses, Haddock, & Zanna, 1993;Stangor, Sullivan, & Ford, 1991). Seminal work on intergroup anxiety (Stephan &Stephan, 1985) demonstrated that negative affect—generalized feelings of awkward-ness, anxiety, and apprehension—elicited during intergroup contact stronglyinfluences people’s attitudes and behaviors (Stephan, Ybarra, & Bachman, 1999;Stephan, Ybarra, Martinez, Schwarzwald, & Tur-Kaspa, 1998). Anxiety andapprehension directly associated with communication barriers has also been shownto predict inimical attitudes toward ethnolinguistic outgroups (e.g., non-nativeteaching assistants; McCroskey, 1998).The psychological literature has also largely overlooked intercultural commu-

nication difficulties as a potential causal factor underlying prejudice anddiscrimination. Whereas social psychologists have extensively documented theimpact of stereotypes on intergroup attitudes (Dovidio, Brigham, Johnson, &Gaertner, 1996; Hamilton & Sherman, 1994; Stangor & Lange, 1994), interculturalcommunication scholars have investigated diverse variables that influence commu-nication effectiveness. In addition to affecting communication quality, constructssuch as anxiety/uncertainty (Gudykunst & Hammer, 1988), cultural competence(Lustig & Koester, 1996), and communication apprehension (Neuliep & McCroskey,1997) impact the favorability of intergroup attitudes and relations. Somecorrelational and experimental evidence points to a connection between interculturalcommunication barriers and prejudice. Much of this research has examinedAmerican college students’ judgments of international teaching assistants (Mesten-hauser, 1983; Paige, 1990). Communication difficulties between US nationals andforeign student-instructors are a significant source of intergroup conflict andhostility (Brown, 1988; McCroskey, 1998; Rubin & Smith, 1990).

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 611

In an attempt to integrate further intergroup emotion, intercultural communica-tion, and prejudice research, the present article examined a special class of intergroupemotions related to intercultural communication experiences, and more specifically,intercultural communication difficulties. Rather than stemming from strongly heldbeliefs about the negative attributes of a group (Dovidio et al., 1996; Hamilton &Sherman, 1994; Stangor & Lange, 1994), we submit that prejudice toward certainethnolinguistic outgroups may derive from adverse emotions related to linguistic andcultural barriers. This antecedent of outgroup attitudes, labeled intercultural

communication affect, should be especially relevant in international settings, as wellas multicultural and linguistically diverse societies, such as the United States, whereintercultural communication is ‘‘virtually unavoidable’’ (Neuliep & McCroskey,1997, p. 147). To our knowledge, the present study is the first to examine thecontribution of intercultural communication emotions, relative to factors such asstereotypic beliefs, in predicting prejudice toward the culturally different.

3. The relative contribution of intercultural communication emotions

The psychological impact of intercultural communication barriers on prejudicialattitudes can be examined in relation to factors that are widely accepted by socialpsychologists as germane to intergroup judgments. Intergroup attitudes have beenconceptualized as a global evaluation of an attitude object that is based on multiplesources of information (Eagly & Chaiken, 1993; Zanna & Rempel, 1988).Accordingly, much empirical work has investigated the relative contribution ofvarious factors—such as realistic group conflict, values, stereotypes, and affectiveresponses—in determining attitudes toward a wide range of social groups (Eagly &Chaiken, 1993; Esses et al., 1993; Zanna & Rempel, 1988). The Integrated ThreatTheory (ITT) of Prejudice (Stephan & Stephan, 1996a) posits that there are fourbasic causal factors or classes of threat that lead social perceivers to be biased againstoutgroups: negative stereotypes (cognitive beliefs), intergroup anxiety (affectiveresponses), realistic threats (economic and physical concerns), and symbolic/culturalthreats (perceived cultural differences and norm violations). Because communicationbetween ethnolinguistic outgroups may be experienced as a type of intergroup threat(Gudykunst & Hammer, 1988; Stephan & Stephan, 1985), intercultural commu-nication affect was examined within the context of this model. The ITT model alsoprovided a useful framework for examining the relative contribution of interculturalcommunication emotions in predicting intergroup hostility.Although intercultural communication barriers operate in most, if not all,

intergroup situations (Dodd, 1995; Gudykunst, 1986; Kim, 1986), this source ofattitudes is highly prevalent in international contexts where linguistic and culturaldissimilarities between group members are pronounced. International educationalexchange—the movement of students and scholars across national boundaries—presents a unique opportunity for studying intercultural communication. Represent-ing over 185 countries, and many more distinct cultures (Davis, 1998), foreignstudents vary tremendously with respect to national, racial/ethnic, and linguistic

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631612

background. Furthermore, it is widely recognized that members of this group arevulnerable to social isolation, prejudice, and discrimination (Leong & Chou, 1996;Paige, 1990; Pedersen, 1991; Sodowsky & Plake, 1992; Spencer-Rodgers, 2001).Foreign students rank negative attitudes and a lack of cultural sensitivity among USnationals as the greatest perceived barriers to successful intergroup relations (Lee,Abd-Ellah, & Burks, 1981).An additional goal of this study was to investigate attitudes toward the

international student community. Although a vast literature exists concerning thisgroup (for reviews, see Leong & Chou, 1996; Mestenhauser, 1983; Paige, 1990;Pedersen, 1991), more research is needed on the nature and structure of Americans’attitudes toward their foreign guests. There is ample evidence indicating thatunfavorable relations with host nationals have serious consequences for thepsychological well-being of international students (Leong & Chou, 1996; Paige,1990; Pedersen, 1991). Because social contact has been identified as a significantmoderator of intergroup attitudes and has been studied extensively in internationalsettings (Allport, 1954; Amir, 1976; Stephan et al., 1998), prejudicial attitudes werealso examined among domestic students who had experienced differing levels ofsocial contact with the international community.

4. The structure of attitudes toward foreign students

The ITT (Stephan & Stephan, 1996a) posits that there are four causal factors thatgive rise to prejudice: negative stereotypes, intergroup anxiety, realistic threats, andsymbolic/cultural threats. Stereotypic beliefs are a well-established source of inimicalattitudes toward the culturally different, especially where there has been minimalprior intergroup and interpersonal contact (Dovidio et al., 1996; Stangor & Lange,1994). The international student literature indicates that a narrow set of negativeattributes is commonly ascribed to this group as a whole (Paige, 1990; Pedersen,1991; Spencer-Rodgers, 2001). To illustrate, a prevalent view exists of foreignstudents as outsiders who are culturally maladjusted, na.ıve, and confused. They areseen as psychologically unbalanced individuals who suffer from a ‘‘foreign studentsyndrome’’, a controversial condition characterized by a disheveled appearance, apassive and withdrawn interpersonal style, and a multitude of psychosomaticailments (Paige, 1990; Pedersen, 1991). Consistent with the ITT, individualstereotypic beliefs are associated with prejudice toward international students(Spencer-Rodgers, 2001).Emotions are another fundamental source of intergroup judgments (Dijker, 1987;

Esses et al., 1993; Stephan & Stephan, 1985, 1996a). Host nationals may view foreignstudents, especially those from developing nations, as a low-status or inferior groupand they may feel contempt or disdain for the group. Highly ethnocentric individualsmay feel suspicious, defensive, and hostile toward the international community.Intergroup anxiety—the apprehension individuals feel when anticipating orexperiencing social contact with an outgroup—is a highly prevalent emotion inintercultural contexts (Stephan et al., 1998). On the other hand, international

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 613

students may also evoke a number of positive emotions among members of thereceiver-nation. Some research suggests that domestic students feel curious,interested, and inspired by their foreign guests (Spencer-Rodgers, 2001). Becauseboth positive and negative emotions have been shown to predict intergroupevaluations (Esses et al., 1993; Stangor et al., 1991), a measure of general (positiveand negative) affective responses toward foreign students was included in the study.Intergroup competition represents a third type of intergroup threat (Stephan &

Stephan, 1996a) underlying prejudice. Realistic threats are related to group conflict,competition for scarce resources, or threats to the physical well-being of an ingroup(Levine & Campbell, 1972; Stephan & Stephan, 1996a). Realistic threat is a causalfactor underlying hostility toward immigrants groups (Stephan et al., 1998, 1999)and may be a determinant of prejudice toward foreign students. Although mostinternationals (77%) pay for their education with funds from family members andoverseas agencies (Davis, 1998), a popular belief exists that US institutions financethe education of foreign students (Paige, 1990; Rubin, 1997). Foreign students maybe regarded as illegitimate competitors who are depriving domestic students ofvaluable educational and material resources (e.g., admission to competitive academicprograms, housing services, financial aid, employment opportunities, etc.). As aresult, US nationals may oppose institutional policies and programs designed tobenefit international students and perceptions of realistic threat may contribute tointergroup hostility.Threats associated with value-laden beliefs and perceived group differences in

cultural norms constitute a fourth determinant of prejudice (Esses et al., 1993;Rokeach, 1968; Stephan & Stephan, 1996a). Symbolic threat is experienced when aningroup believes that its sociocultural system is being obstructed, undermined, orviolated by an outgroup. Although international students may be regarded asvaluable cultural and intellectual resources that enrich the university community,they also possess values, norms, and patterns of behavior that conflict with those ofdomestic students (Mestenhauser, 1983; Paige, 1990; Pedersen, 1991). Foreignstudents can express opinions that challenge or threaten the worldview of domesticstudents and they may represent social, religious, and political systems that areunpopular in the receiver-nation (Paige, 1990). Where the values, beliefs, andcultural norms of ethnolinguistic groups are greatly dissimilar, symbolic threatsshould impact the favorability of intergroup judgments.

5. Hypotheses

We hypothesized that intercultural communication emotions would constitute aunique and potent source of attitudes toward ethnolinguistic outgroups. Based onthe sizable correlation found between stereotypic beliefs and judgments of foreignstudents in previous research (Spencer-Rodgers, 2001), consensual stereotypic beliefswere expected to be a significant, but relatively weaker, determinant of attitudes.Given that the intergroup anxiety literature indicates a consistent associationbetween intergroup emotions and prejudice, we predicted that general affective

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631614

responses would be moderately related to host attitudes. Realistic and symbolicthreats are generally less salient sources of prejudice toward subordinate groups(Stephan et al., 1998, 1999). Because foreign students are unlikely to challengeseriously the economic position and sociocultural system of American students,weak positive relations were expected between both realistic and symbolic threatsand intergroup judgments.The quantity and quality of social contact experienced with members of an

outgroup should moderate the structure of intergroup attitudes (Allport,1954; Amir, 1976; Stephan et al., 1998). Accordingly, intercultural communicationemotions were expected to be more strongly related to prejudice among individualswho have had less direct experience with foreign students. Stereotypic beliefsand affective reactions were hypothesized to be more potent predictors ofprejudice among individuals who have experienced minimal contact with theinternational community. In contrast, realistic and symbolic threats would bestronger predictors of prejudice at higher levels of social contact with internationalstudents (Stephan et al., 1998). To test these hypotheses, the structure of attitudestoward foreign students was examined among low-contact and high-contact hostnationals.

6. Summary of hypotheses

1. Intercultural communication emotions would be uniquely and strongly related toattitudes toward foreign students. Stereotypes and general affect would bemoderate predictors, and realistic and symbolic threats would be weak predictors,of prejudice.

2. Social contact would moderate (decrease) the association between interculturalcommunication emotions, stereotypes, general affect, and prejudice; and increasethe association between realistic threats, symbolic threats, and prejudice.

7. Method

7.1. Participants and procedures

A diverse group of undergraduate and graduate students enrolled at two largeWest Coast universities participated in the study. Approximately 64% of thestudents were recruited through the psychology research participant pool. Theseindividuals received partial course credit for their participation in a 1-hour testingsession. In order to increase the diversity of the sample, additional participants wererecruited through classroom presentations and advertisements posted in variousacademic departments. The latter individuals were paid $5 for their participation.Nineteen individuals who were not citizens or permanent residents of the UnitedStates and three individuals who did not indicate their racial/ethnic category

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 615

membership were eliminated from the study, resulting in a usable sample of N ¼154: The demographic characteristics of the sample were as follows: 83 (54%)identified as European American, 34 (22%) identified as Asian American, 26 (17%)identified as Latino/Hispanic, and 11 (7%) identified as African American. Thepercentage of participants who were born in the United States was as follows:European American (98%), Asian American (93%), Latino/Hispanic (86%), andAfrican American (100%). The sample consisted of 86 (56%) women. Eighty-twopercent of the participants were undergraduate students and 18% were graduatestudents. The students ranged in age from 17 to 42, with a mean age of 23.The students were invited to participate in a study on ‘‘social attitudes’’. In order

to reduce demand, the written and oral instructions informed the participants thatthe researchers were interested in people’s attitudes toward a wide variety of socialgroups. They were further instructed that they would be responding to questionsabout the group ‘‘foreign students’’ in the current testing session. The term ‘‘foreignstudents’’ was defined as ‘‘college students from other countries who are studying inthe United States’’. Because the label ‘‘international students’’ may evoke sociallydesirable responding (Spencer-Rodgers, 2001), the potentially more negative (Paige,1990), but commonly used designation ‘‘foreign students’’ was selected for this study.The participants were assured that their responses to the questionnaire would bekept strictly confidential and they returned their completed questionnaires to ananonymous drop-box at the end of the testing session. They were subsequentlydebriefed and thanked for their participation.

7.2. Measures

7.2.1. Global attitudes

Scholars posit that prejudice may derive from a variety of sources, such ascognitive beliefs, affective responses, and values (Eagly & Chaiken, 1993; Esses et al,1993; Stangor et al., 1991). These factors have been shown to be differentiallyimportant in predicting attitudes toward different social groups (e.g., immigrants,homosexuals, the disabled, etc.). In order to assess global evaluations of foreignstudents, semantic differential items that are essentially ‘‘content-free’’ were selected.Content-free items allow participants to make judgments on the basis of information(e.g., cognitions, emotions, values, etc.) that is most relevant and important to them(Esses et al, 1993; Stangor et al., 1991). The participants indicated their overallattitude toward foreign students on three semantic differential scales: favorable–unfavorable, positive–negative, and good–bad. The 11-point scales ranged from 0(extremely favorable) to 100 (extremely unfavorable). Variations of this instrumenthave been shown to possess high test–retest reliability, to be strongly correlated withlonger attitude scales, and they have been used to assess attitudes toward a widevariety of groups (Esses et al., 1993). Cronbach’s alpha was 0.90. For purposes ofcomparison, the participants also rated the group ‘‘American college students’’(defined as college students who are citizens and permanent residents of the UnitedStates) on the same three-item measure. Cronbach’s alpha for American studentswas 0.93. Higher scores indicate a less favorable attitude toward the group.

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631616

7.2.2. Intercultural communication emotions

As a measure of intercultural communication affect, the participants responded toa seven-item scale developed for this study. The items were as follows: (a) ‘‘I find itunpleasant to listen to foreign students who speak with a strong accent’’, (b) ‘‘Irarely feel annoyed when talking to foreign students who have poor English skills’’(reverse-scored), (c) ‘‘I sometimes feel frustrated when interacting with foreignstudents who do not understand American customs and ways of behaving’’, (d) ‘‘Iam comfortable interacting with foreign students who have different ethnic customsand practices’’ (reverse-scored), (e) ‘‘I become impatient when listening to foreignstudents who speak English poorly’’, (f) ‘‘I find it agreeable to talk to foreignstudents who speak with a strong accent’’ (reverse-scored), and (g) ‘‘I sometimes feeluncomfortable when interacting with foreign students because of cultural barriers(i.e., cultural differences in ways of communicating)’’. The nine-point response scalewas anchored by 1 (strongly disagree) and 9 (strongly agree). Factor analysis(principal components analysis with varimax rotation) indicated that all of the itemsloaded on a single factor, which explained 51% of the variance. Cronbach’s alphawas 0.76. Higher scores correspond to greater negative affect associated withintercultural communication barriers.

7.2.3. Preliminary stereotypic attributes

A preliminary stereotype scale was composed of 40 attributes that have beenspecifically ascribed to foreign students. Thirty of the descriptors were adapted froma free-response study on international student stereotypes (Spencer-Rodgers, 2001).Based on a review of the international student literature, 10 additional traits (e.g.,competitive, homesick, etc.) were included to test specific hypotheses regarding thecontent of the consensual stereotype. Participants were asked to rate the extent towhich each of the 40 attributes was characteristic or typical of foreign students as agroup. The nine-point response format ranged from 1 (not at all characteristic) to 9(extremely characteristic). To assess the evaluative content of the attributes, theparticipants rated each of the 40 descriptors on a favorability scale, ranging from �4(extremely negative) to +4 (extremely positive).

7.2.4. Consensual stereotype measure

The impact of stereotypes on intergroup attitudes is related to both the strengthand the valence of the components of a stereotype (Eagly & Mladinic, 1989; Esseset al., 1993; Stephan & Stephan, 1996a). Because valenced attributes that arestrongly associated with a group have greater predictive utility and validity, wecreated a composite (strength� evaluation) stereotype index (Eagly & Mladinic,1989). For each of the 40 attributes, the strength rating of the trait (1–9) wasmultiplied by the valence rating (�4 to +4). The mean (strength� evaluation) scorefor each of the 40 descriptors was then calculated and the distribution of means wasanalyzed. Attributes that were rated as fairly to extremely negative (or positive) andas moderately to extremely characteristic of foreign students were selected asstereotypic attributes for the group. Table 1 presents the initial pool of 21 traits and

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 617

the mean (strength� valence) rating and standard deviation for each of theattributes.To create a reliable consensual stereotype index, the 21 attribute ratings were

factor analyzed (principal components analysis with varimax rotation). Attributesthat loaded on the principal negative stereotype factor (eigenvalue>1.0; e.g.,culturally maladjusted, socially awkward, etc.) and traits that loaded on the positivestereotype factor (eigenvalue>1.0; e.g., hardworking, determined, etc.) wereincluded in the final consensual stereotype measure. These 13 attributes areidentified by the letter d in Table 1. Next, each participant’s (strength� evaluation)rating was summed across the 13 stereotypic traits and divided by the number ofattributes (Eagly & Mladinic, 1989). To create a measure of negative stereotypicbeliefs, the scale was reverse-scored. Coefficient alpha was 0.81. Higher scores on theconsensual stereotype measure indicate that participants hold predominantlynegative stereotypic beliefs about foreign students.

7.2.5. General affective responses

To assess general affective reactions associated with foreign students, theparticipants indicated the extent or degree to which they felt various emotions inresponse to the group (adapted from Stephan & Stephan, 1985). The nine-pointresponse scale was anchored by 1 (not at all) and 9 (extremely). The eight negativeemotions included were: awkward, anxious, uneasy, self-conscious, defensive,suspicious, hostile, and superior. The eight positive emotions were: admiration,respectful, happy, comfortable, confident, interested, curious, and inspired. To assessthe evaluative content of the affective responses, the participants rated each of the 16

Table 1

The consensual foreign student stereotype among American host nationals

Positive attributes Compositea

(strengthb� valencec)

mean (SD)

Negative attributes Compositea

(strengthb� valencec)

mean (SD)

1. Multilingual 22.59 (10.01) 1. Culturally maladjustedd �11.97 (8.41)2. Intelligentd 21.86 (7.46) 2. Clannish �9.93 (10.78)

3. Eager to learnd 19.72 (9.33) 3. Socially awkwardd �9.71 (9.69)

4. Friendly 19.18 (9.07) 4. Frightened �8.69 (11.57)

5. Hardworkingd 17.31 (8.30) 5. Sad/depressedd �8.03 (7.09)

6. Determinedd 14.74 (9.49) 6. Confused/lostd �7.96 (9.22)

7. Talentedd 13.29 (9.61) 7. Anxious/stressedd �7.57 (8.45)

8. Open-minded 11.43 (10.35) 8. Lonelyd �6.45 (7.01)

9. Worldly 10.88 (9.74) 9. Speak English poorlyd �6.18 (7.86)

10. Brave/adventurous 8.91 (8.96) 10. Arrogant �6.01 (9.17)

11. Studiousd 6.63 (10.27)

aNote: Composite=mean (strength rating� valence rating) across all participants.bStrength scores range from 1 (not at all characteristic) to 9 (extremely characteristic).cValence scores range from –4 (extremely negative) to +4 (extremely positive).dAttributes included in the final consensual stereotype scale.

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631618

emotions on a favorability scale that ranged from �4 (extremely negative) to +4(extremely positive).For each of the 16 emotions, the strength rating (1–9) was multiplied by the

favorability rating (�4 to +4) to create a (strength� evaluation) index. Emotionsthat were rated as fairly to extremely negative (or positive) and that were moderatelyto strongly associated with the group were included in the initial pool of affectiveresponses. Factor analysis of the affect scale (principal components analysiswith varimax rotation) yielded two emotion categories: positive emotions (eigen-value>1.0; e.g., interested, inspired, curious, etc.) and intergroup anxiety (eigen-value>1.0; e.g., awkward, anxious, uneasy, etc.). Each participant’s(strength� evaluation) rating was then summed across the 12 affective responsesand divided by the number of emotions. To create a measure of negative generalaffect, the scale was reverse-scored. Cronbach’s alpha was 0.79. Higher scores on thecomposite index indicate that participants primarily associate negative emotionswith foreign students.

7.3. Realistic threats

Perceptions of realistic threat were assessed by six items (adapted from Stephan &Stephan, 1996a) including: (a) ‘‘Foreign students take jobs away from Americanstudents (e.g., on-campus employment, teaching/research assistantships)’’, (b)‘‘Foreign students pay their fair share for the education and services thatthey receive at US universities’’ (reverse-scored), (c) ‘‘Foreign students takevaluable educational resources away from American students (e.g., financialaid, university housing, etc.)’’, and (d) ‘‘American colleges and universities arepaying too much to finance the education of foreign students’’. The items wererated on a nine-point scale with endpoints 1 (strongly disagree) and 9 (strongly

agree). Factor analysis (principal components analysis with varimax rotation)indicated that all six items loaded on one factor, which explained 61% of thevariance. Cronbach’s alpha was 0.78. Higher scores correspond to greater perceivedrealistic threat.

7.3.1. Symbolic/cultural threats

As a measure of symbolic threat, the participants responded to six statements(adapted from Stephan & Stephan, 1996a), which included: (a) ‘‘Some Americancolleges and universities are losing their ‘American’ character because of increasingforeign student enrollments’’, (b) ‘‘Foreign students contribute positively to theethnic mix at American universities’’ (reverse-scored), and (c) ‘‘Cherished Americannorms and traditions are threatened somewhat by increasing foreign studentenrollment on US campuses’’. The statements were rated on a nine-point scalewith the continuum ranging from 1 (strongly disagree) to 9 (strongly agree).Factor analysis (principal components analysis with varimax rotation) indicated thatall six items loaded on a single factor, which explained 53% of the variance.Cronbach’s alpha was 0.72. Higher scores indicate greater perceived symbolic/cultural threat.

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 619

7.3.2. Social contact

To assess social contact with the international community, the participantsresponded to three items (adapted from Biernat & Crandall, 1994): (a) ‘‘Howoften do you talk to and engage in informal conversations with foreign students?’’,(b) ‘‘How often do you study or do other class work with foreign students?’’,(c) ‘‘How often do you do things socially with foreign students? (This includesthings like sharing meals, going to movies and parties, etc.)’’. The items wererated on a nine-point scale that ranged from 1 (never) to 9 (all the time).Factor analysis (principal components analysis with varimax rotation) indicatedthat the items loaded on one factor, which explained 79% of the variance.Cronbach’s alpha was 0.93. Higher scores correspond to greater social contact withthe group.The order of presentation of each of the scales was counterbalanced across all

participants, with the following exceptions: (1) the strength ratings preceded thefavorability ratings for both the stereotypic attribute and affective response measuresand (2) the measure of attitudes toward ‘‘American college students’’ was presentedat the end of the questionnaire.

8. Results

A multivariate analysis of variance (MANOVA) was conducted on the principalstudy variables (global attitudes, intercultural communication emotions, consensualstereotypes, general affect, and realistic and symbolic threats), using gender,ethnicity, and educational level (undergraduate vs. graduate) as the factors. Therewere no main effects of gender. However, there was a significant (gender�educational level) interaction on global attitudes, F (1, 139)=5.54, po0:05; andgeneral affect, F (1, 139)=4.77, po0:05: Male graduate students reportedsignificantly greater prejudice (M ¼ 46:67) and less positive affect (M ¼ 0:39)toward foreign students than did female graduate students (M ¼ 29:11 and �4:64;respectively). There were no other interactions involving gender. There was a maineffect of ethnicity on the following measures: global attitudes, F (3, 139)=3.93,po0:01; general affect, F (3, 139)=2.89, po0:05; and symbolic threats, F (3, 139)=3.05, po0:05: However, post-hoc multiple comparisons (using the Games Howelltest statistic for unequal cell sizes) indicated that group-level differences in globalattitudes and general affect were not statistically significant. European Americans(M ¼ 3:48) perceived significantly greater symbolic/cultural threats than did Latinos(M ¼ 2:57). There was a main effect of educational level on realistic threats,F (1, 139)=5.57, po0:05: Graduate students perceived significantly greater realisticthreat (M ¼ 5:12) than did undergraduate students (M ¼ 3:97). In addition, weexamined the intercorrelations between age and the principal study variables. Olderstudents perceived greater symbolic/cultural threat (r ¼ 0:17; po0:05) and theytended to perceive greater realistic threat (r ¼ 0:13; ns) than younger students. Thesubsequent regression analyses were conducted controlling for gender, ethnicity, age,and the educational level of the participants.

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631620

On average, the participants’ overall attitude toward foreign students (M ¼ 31:17;SD ¼ 22:34) corresponds to a ‘‘somewhat’’ favorable (positive, good) evaluativeorientation. This mean evaluative score is comparable to that obtained for minorityand immigrant groups (Esses et al., 1993). Notably, judgments of internationalstudents were relatively unfavorable: the mean evaluative rating for the target group‘‘foreign students’’ was significantly less favorable than that obtained for ‘‘Americancollege students’’ (M ¼ 23:89; SD=15.92), with t(153)=6.94, po0:001:Table 1 presents the consensual stereotype of foreign students. Table 2 presents the

means and standard deviations for the intercultural communication emotions,stereotype, general affect, and threat scales. The mean score for interculturalcommunication emotions (M ¼ 4:54) indicates that many American college studentsfelt frustrated, impatient, and uncomfortable when encountering communicationobstacles with the international student community. On average, stereotypic beliefsabout foreign students were somewhat positive (M ¼ �2:73; scale reverse-scored),although there was substantial variability on this measure. Descriptive statistics forthe general affect scale suggest that emotional responses toward foreign studentswere also generally positive (M ¼ �3:43; scale reverse-scored), with considerablevariability existing in participants’ affective orientation toward the group. The meanscores for realistic threats (M ¼ 4:26) and symbolic/cultural threats (M ¼ 3:18)suggest that foreign students are generally viewed as contributing positively to theuniversity community on US campuses, although international students are viewedas illegitimate competitors by some host nationals.Simultaneous regression analyses were conducted to determine the predictive

utility of each of the attitudinal determinants. In preparation to run the analysis,correlations among the predictor variables were calculated and are presented inTable 2. All of the predictors were positively intercorrelated, which is indicative ofintra-attitudinal consistency (Eagly & Chaiken, 1993; Zanna & Rempel, 1988).Several of the predictors (e.g., realistic threats and symbolic/cultural threats) werehighly correlated; however, the percentage of shared variance did not exceed 18% forany intercorrelation. This pattern of intercorrelations suggests that the attitudinal

Table 2

Means, standard deviations, and intercorrelations among predictor variables and global attitudes toward

foreign students

N ¼ 154 Mean (SD) Intercorrelations

1 2 3 4 5

1. Intercultural communication emotions 4.54 (2.31) —

2. Consensual stereotypes �2.73 (6.33)a 0.29*** —

3. General affective responses �3.43 (4.77)a 0.31*** 0.37*** —

4. Realistic threats 4.26 (1.96) 0.23** 0.26** 0.14 —

5. Symbolic/cultural threats 3.18 (2.15) 0.34*** 0.41*** 0.25** 0.42*** —

6. Global attitudes 31.17 (22.34) 0.48*** 0.39*** 0.43*** 0.16* 0.30***

aNote: The stereotype and affect indexes were reverse-scored, such that negative mean scores indicate

overall positive stereotypic beliefs and emotional responses.

*po0:05; ** po0:01; *** po0:001:

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 621

determinants are related, but distinct, sources of prejudice toward foreign students.The zero-order correlations between the predictor variables and the criterionmeasure of global attitudes are presented in the last row of Table 2. The fiveattitudinal determinants were entered simultaneously as predictors in the regressionanalysis in order to determine the unique contribution of each variable to overallattitudes.1 The results of the analysis are presented in Fig. 1.The simultaneous regression analysis indicates that intercultural communication

emotions (B ¼ 0:34; po0:001) were strongly and uniquely related to attitudes towardforeign students. General affective responses (B ¼ 0:26; po0:001) and consensualstereotypic beliefs (B ¼ 0:19; po0:01) were unique, but less potent, predictors ofprejudice. The standardized regression coefficient for symbolic/cultural threats wasnot significant (B ¼ 0:10; ns). Particularly low was the association of intergroupattitudes with perceived realistic threats (B ¼ 0:02; ns). The model including all fivepredictor variables accounted for 41% of the variance in global evaluations offoreign students.The determinants of intergroup attitudes were also examined among individuals

who had experienced differing levels of social contact with the target group.

Fig. 1. Predictors of global attitudes toward foreign students among American host nationals. N ¼ 154:Note: ICEs=intercultural communication emotions (negative affect associated with perceived linguistic

and cultural barriers).

1Evans (1991) has cautioned against the use of multiplicative composites in regression analyses. The

strength ratings, evaluative ratings, and (strength� evaluative) indices were entered into a hierarchical

regression analysis separately. Similar results emerged. Hence, the multiplicative composite did not inflate

the predictive utility of the stereotype and general affect measures.

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631622

Moderated regression analyses, using social contact as the moderator variable, wereconducted separately for each of the five predictor variables.2 Consistent with ourhypotheses, consensual stereotypic beliefs (interaction B ¼ �0:09; ns) and generalaffective responses (interaction B ¼ �0:10; ns) tended to be less potent uniquepredictors of prejudice at higher levels of social contact with the internationalcommunity. Contrary to prediction, social contact had no appreciable effect on therelation between intercultural communication emotions and intergroup judgments(B ¼ �0:01; ns). That is, intercultural communication emotions were equally, andstrongly, related to intergroup attitudes for both low-contact and high-contact hostnationals. As hypothesized, realistic threats (B ¼ 0:11; ns) and symbolic/culturalthreats (B ¼ 0:14; ns) tended to be more potent predictors of prejudice amongindividuals who had interacted more frequently with foreign students.

9. Discussion

In order to gain greater understanding of the psychological impact of interculturalcommunication barriers on attitudes toward the culturally different, we tested amodel of global attitudes (Stephan & Stephan, 1996a) toward foreign students.Based on previous intergroup emotion and intercultural communication research(Esses et al., 1993; Gudykunst & Hammer, 1988; McCroskey, 1998; Stephan &Stephan, 1985), we hypothesized that prejudice toward ethnolinguistic outgroupsmay stem from intercultural communication emotions, or negative affect associatedwith perceived linguistic and cultural obstacles. This source of attitudes wasexamined within the context of the Integrated Threat Theory of Prejudice (Stephan& Stephan, 1996a). The predictive utility of intercultural communication emotionswas examined in relation to four causal factors or classes of threat that are widelyacknowledged by social psychologists as relevant to intergroup judgments:stereotypic beliefs (Hamilton & Sherman, 1994), affective responses (Dijker, 1987;Esses et al., 1993; Stephan & Stephan, 1985), realistic threats (Levine & Campbell,1972), and symbolic/cultural threats (Esses et al., 1993; Rokeach, 1968).Intercultural communication emotions were the strongest unique predictors of

attitudes toward foreign students. This antecedent of prejudice may be particularlysalient in international contexts, where language and cultural differences can greatlyimpair communication between ethnolinguistic outgroups (Lustig & Koester, 1996;Wiseman & Koester, 1993). Many American college students reported feelinguncomfortable, impatient, and frustrated when encountering communicationdifficulties with the international student population on their campuses. Factorssuch as accented speech, cultural differences in non-verbal communication styles,

2The unique effect of each of the predictor variables was assessed by regressing the criterion of global

attitudes onto each of the five predictors separately (e.g., intercultural communication emotions), while

controlling for the remaining four predictor variables (e.g., stereotypes, affective responses, and realistic

and symbolic threats). The resulting residual scores were retained. Next, interaction terms (residualized

predictor variables�moderator) were computed. The centered main effects and interaction terms were

subsequently entered into separate moderated regression analyses.

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 623

and cultural variations in values, norms, and customs contribute to thesecommunication problems. Intercultural interactants may also fear the negativeconsequences of communication failures and emotionally laden cultural misunder-standings (Stephan & Stephan, 1996a), such as appearing foolish to others orexperiencing rejection. In sum, negative affect elicited during communication withethnolinguistic outgroups represents an additional type of intergroup threat (Stephan& Stephan, 1996a) associated with prejudice toward the culturally different. A largecorpus of research on intergroup anxiety (Gudykunst, 1986; Stephan & Stephan,1985; Stephan et al., 1999) has established that individuals may experiencegeneralized feelings of awkwardness, anxiety, and apprehension when experiencingor anticipating social contact with an outgroup. The present research suggests that abroad range of adverse emotions, directly associated with communication betweenethnolinguistic groups, can be a potent source of intergroup hostility.Intercultural communication emotions may be especially germane to judgments of

foreign students. The most derogatory attributes elicited in a free-response study oninternational student stereotypes were related to perceived English language deficitsand intercultural communication obstacles (Spencer-Rodgers, 2001). When foreignstudents perform in the capacity of student-instructor, they may be evaluatedextremely negatively by American academic personnel, undergraduate students, andthe general public (McCroskey, 1998; Paige, 1990). Both actual and assumedcommunication barriers are thought to underlie these strongly prejudicial evalua-tions. Brown (1988) found that perceptions of communication difficulties andforeign accentedness accounted for most of the variance in college students’ ratingsof international teaching assistants. In an experimental study, Rubin and Smith(1990) manipulated the level of accentedness of a foreign student-instructor andfound that perceived—but not real—levels of accentedness predicted evaluations ofthe target. Our findings provide further evidence that intercultural communicationdifficulties underlie prejudice toward foreign students.The structure of intergroup attitudes was expected to differ for host nationals who

had experienced little or no social contact with the international community andindividuals who had interacted more extensively with students from abroad.Frequent social interaction with foreign students should facilitate interculturalcommunication and understanding. With increased intercultural exposure, socialperceivers become more accustomed to accented speech, more adept at commu-nicating with non-native speakers of a language, and more sensitive and sympathetictoward cultural differences (Giles & Robinson, 1990; Kim, 1986; Wiseman &Koester, 1993; Yook & Albert, 1999). Contrary to prediction, social contact did notmoderate the relationship between intercultural communication emotions and globalattitudes. That is, intercultural communication affect was strongly and uniquelyrelated to prejudice for both low-contact and high-contact host nationals. It isnoteworthy, however, that the social contact scale used in the present study primarilyassessed quantitative, rather than qualitative, contact with the target group (Biernat& Crandall, 1994). High levels of social contact with members of an ethnolinguisticoutgroup may have the unfortunate effect of making intercultural communicationdifficulties more psychologically salient. These data are consistent with previous

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631624

research indicating that frequent contact with members of an outgroup does notnecessarily translate into improved intergroup communication and relations (Amir,1976; Dovidio et al., 1996; Kim, 1986). It is also notable that ethnic groupmembership did not affect the mean favorability of intercultural communicationemotions and global attitudes. These findings parallel prior research indicating thatindividuals may be prejudiced toward culturally similar immigrant groups (e.g.,Hispanic and Anglo American students may hold comparably unfavorable attitudestoward Mexican immigrants; Stephan et al., 1999).A second purpose of the study was to examine the nature and structure of

American college students’ attitudes toward the international community. Thestructure of intergroup judgments has been found to vary according to the targetgroup under investigation (Eagly & Chaiken, 1993; Esses et al., 1993; Stephan et al.,1998, 1999). That is, stereotypes, emotions, and other variables are differentiallyimportant as predictors of attitudes toward different outgroups. A major strength ofthe ITT model (Stephan & Stephan, 1996a) is that it permits comparison of causalfactors that contribute to prejudice toward diverse groups. In addition tointercultural communication emotions, general affective responses and consensualstereotypic beliefs were significant and unique predictors of inimical attitudes towardforeign students. Symbolic/cultural threats were marginally related to intergroupjudgments and realistic threats were not unique determinants of intergroupevaluations. These findings are consistent with ITT research (Stephan et al., 1999)indicating that realistic and symbolic threats are less salient sources of attitudestoward subordinate cultural groups.

10. Attitudes toward foreign students

This study is among the most comprehensive empirical investigations of foreignstudent stereotypes (see also, Bond, 1986; Spencer-Rodgers, 2001). The initialstereotype index was composed of 40 group-specific stereotypic attributes. Ashypothesized by numerous cross-cultural scholars (Mestenhauser, 1983; Paige, 1990;Pedersen, 1991), many of the unfavorable attributes consensually ascribed tointernational students were related to perceived cultural and social adjustmentproblems. For example, foreign students were characterized as frightened, sad/depressed, and lonely. Moreover, these stereotypic beliefs were uniquely related tohost attitudes. A theoretical explanation for why endorsement of the consensualstereotype was significantly related to host attitudes emphasizes the function ofstereotypes as heuristic devices when social perceivers lack familiarity and experiencewith members of an outgroup (Hamilton & Sherman, 1994; Neuliep & McCroskey,1997). Stereotypic beliefs, developed during brief encounters with foreign students,or through secondary sources such as classmates, family members, or the media, cansubsequently guide evaluations and behaviors (Dovidio et al., 1996; Hamilton &Sherman, 1994; Stangor & Lange, 1994). On average, the participants in this studyhad experienced relatively little contact with the international community(M ¼ 3:89). The moderated regression analyses lend further support to this

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 625

theoretical explanation. Domestic students who had experienced less contact withthe international student population were more likely to rely on stereotypicknowledge as a basis for intergroup judgments.Although consensual stereotypes were a significant source of attitudes toward

international students, the results of the simultaneous regression analysis suggestthat emotional responses tend to be more uniquely related to evaluations of thisgroup. Affective associates have been found to be potent predictors of inimicalattitudes toward racial/ethnic minority and immigrant groups (Dijker, 1987; Esseset al., 1993; Stephan et al., 1998, 1999). A number of scholars have posited thatintergroup attitudes are formed on the basis of emotionally significant, and hence,highly memorable and personally relevant encounters with representative membersof an outgroup (Eagly & Chaiken, 1993; Stangor et al., 1991). As a result, thequantity and quality of social contact experienced with members of an ethnolin-guistic outgroup should moderate the favorability and predictive utility ofintergroup emotions. As expected, general affective responses tended to be morestrongly related to attitudes among individuals who had experienced minimal(quantitative) contact with the target group. For high-contact host nationals,frequent interaction with members of an ethnolinguistic outgroup may decreasefeelings of intergroup anxiety (Stephan & Stephan, 1985) and uncertainty(Gudykunst & Hammer, 1988), while increasing favorable emotions such assympathy, respect, and admiration.Overall, perceived symbolic/cultural threats made a marginal contribution to the

prediction of prejudice. The relatively weak predictive utility of symbolic threats maybe explained by several factors. As temporary sojourners and potential immigrants,foreign students are unlikely to obstruct, undermine, or violate the socioculturalsystem of American students. Rather, US nationals appear to value the cultural andintellectual contribution of the international student community. Foreign studentsadd to the racial/ethnic diversity of the student body and they share their knowledgeof different countries, social and political systems, and cultural customs andpractices. Nonetheless, perceptions of cultural differences have been found to predicthostility toward international teaching assistants. Brown (1988) manipulatedinformation about the demographic background of a foreign student-instructorand found that perceptions of cultural differences strongly contributed toevaluations of the target. Mean perceptions of symbolic threats were also higheramong European American students (the dominant ethnolinguistic group) thanLatino students, and symbolic/cultural threat was a stronger predictor of prejudiceat higher levels of social contact. Frequent contact with an ethnolinguistic outgroupmay increase the psychological salience of cultural dissimilarities (Stephan et al.,1998). Perceptions of cultural differences may also have an indirect, adverse effect onintergroup attitudes through increased perceptions of intercultural communicationdifficulties.Realistic threats were not uniquely related to judgments of foreign students.

Stephan et al. (1998) posited that realistic threats are a less salient source of attitudestoward minority groups. Given that foreign students are temporary sojourners andpotential immigrants, international students are unlikely to pose a serious threat to

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631626

the economic standing of domestic students. For many US colleges and universities,international students help maintain enrollments and they represent a substantialsource of tuition revenue, teaching service, and intellectual capital (Paige, 1990;Rubin, 1997). It is noteworthy, however, that mean perceptions of realistic threatswere substantial. This finding may be explained, in part, by the sizable number ofgraduate students in the study sample. Perceptions of realistic threat weresignificantly higher among graduate than undergraduate students. Domesticgraduate students compete directly with internationals for admission to competitiveacademic programs, teaching and research assistantships, certain fellowships, andpost-graduate employment in competitive industries. Realistic threat was also astronger predictor of attitudes at higher levels of social contact. Host nationals whointeract more frequently with foreign students are more likely to compete with themfor limited educational resources.The correlational nature of the data limits the conclusions that can be drawn

from this study. A significant association between ratings on a predictor variable(e.g., an affect or stereotype measure) and an attitude scale does not demonstratethat the predictor variable produced the attitude (Eagly & Chaiken, 1993).Individuals may first retrieve a global evaluative orientation toward a target groupand then construct a response to the predictor variable that is consistent with theiroverall attitude. Maio, Esses, and Bell (1994) provided some insight into these causalrelations in an experimental study. The researchers manipulated information aboutthe valence and relevance of stereotypic beliefs, values, and affective responsestoward a fictitious immigrant group and found that the predictor variablesinfluenced overall evaluations of the group. Furthermore, intercultural trainingprograms designed to increase intercultural communication competence and todecrease anxiety/uncertainty, have been shown to improve attitudes towardethnolinguistic outgroups (Wiseman & Koester, 1993; Yook & Albert, 1999). Thecausal relation between the quality of intercultural communication and prejudicemay also be reciprocal.Social desirability bias may account for the apparent discrepancy between foreign

students’ widely held perceptions of prejudice and discrimination in the UnitedStates (Lee et al., 1981; Sodowsky & Plake, 1992) and American college students’self-reported judgments of the group. Participants may have hesitated to report theirnegative beliefs, emotions, and attitudes toward foreign students. Alternatively,scholars maintain that intergroup attitudes in contemporary US society are nolonger characterized by strong negative sentiments toward outgroups, but rather bythe absence of positive characterizations of outgroups in comparison to ingroups(Dovidio et al., 1996). Indeed, judgments of ‘‘foreign students’’ in the present studydid not compare favorably with those of ‘‘American college students’’.Our findings help to explain why attitudes toward the international student

community are relatively unfavorable. The combination of intercultural commu-nication emotions, general affect, consensual stereotypes, and perceived threatsaccounted for a significant percentage of the variance in attitudes toward foreignstudents. Factors not included in the study, such as personality traits and pastexperiences and behaviors, have also been found to predict intergroup attitudes

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 627

(Eagly & Chaiken, 1993; Zanna & Rempel, 1988). Inclusion of these, and othervariables, could significantly increase our ability to predict prejudice towardinternational students. The affective dimension of intercultural communicationbarriers may constitute a substantial source of hostility toward other ethnolinguisticgroups, such as Latinos/Hispanics and Asian Americans. Replications with othercultural and linguistic groups would help to establish the generalizability of ourfindings.

11. Implications for international student exchange

In 1998, close to 500,000 international students were attending educationalinstitutions in the United States (Davis, 1998). Presently, American colleges anduniversities host the world’s largest foreign student population and internationalenrollments are increasing. Institutions of higher education are ‘‘internationalizing’’:they are creating specialized programs and services to attract the internationalstudent market and they are aggressively recruiting students from abroad (Paige,1990; Rubin, 1997). The large influx of international students and scholars to USeducational institutions presents opportunities for intercultural contact, commu-nication, and, potentially, conflict (Paige, 1990). International educational exchangeenables foreign individuals and host nationals to acquire intercultural understandingand competence, factors that can enhance intergroup and international relations(Allport, 1954; Amir, 1976; Paige, 1990). Differences in language, cultural norms,and patterns of behavior, can also lead to serious communication and interactionproblems between domestic and foreign students (Brown, 1988; Leong & Chou,1996; Mestenhauser, 1983; Paige, 1990).Educational administrators can take into account the various sources of prejudice

toward foreign students when creating policies and programs designed to reduceinimical attitudes toward this group. In accordance with the Contact Hypothesis(Allport, 1954), the evaluative implication of negative stereotypes and general affectcan be addressed by promoting contact with foreign students. Frequent contactalone, however, may not lead to favorable intercultural communication andrelations. Increased social interaction with international students was associatedwith increased perceptions of threat. Realistic and symbolic threats could beattenuated by disseminating information about the economic, cultural, andintellectual contribution of foreign students to the university community. Onestrategy that could alter the evaluative implication of intercultural communicationbarriers would be to create structured opportunities for positive intergroupinteraction (Allport, 1954; Amir, 1976; Dovidio et al., 1996). University programssuch as integrated residential facilities and recreational programs are notableexamples. Understanding the nature and structure of attitudes toward foreignstudent visitors and attempts to alter these attitudes are worthwhile endeavors.Unfavorable relations with host nationals strongly influence the personal andcultural adjustment of international students (Leong & Chou, 1996; Paige, 1990;Pedersen, 1991). Moreover, domestic students may develop enduring negative

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631628

attitudes toward certain national and cultural groups as a result of their interactionswith foreign student visitors.More research is needed at the interface of intergroup emotion, intercultural

communication, and prejudice research. Increasing intergroup conflict in the worldtoday (Stephan & Stephan, 1996b) highlights the necessity of identifying the multiplesources of intergroup hostility. Communication difficulties and cultural barriers mayevoke adverse emotions that, in turn, give rise to prejudicial attitudes anddiscriminatory behaviors. Intercultural communication affect was the strongestunique predictor of attitudes toward foreign students. Intercultural communicationemotions may constitute a significant source of prejudice and discrimination towardother ethnolinguistic outgroups, such as racial/ethnic minority and immigrantgroups. This antecedent of intergroup attitudes is especially relevant in internationalcontexts and multicultural societies, and is likely to become increasingly importantas international migration and globalization bring more ethnolinguistic groups intocontact (Stephan & Stephan, 1996b). This study also contributes to a growing bodyof evidence indicating that emotions are central determinants of intergroup attitudes(Dijker, 1987; Esses et al., 1993; Stangor et al., 1991; Stephan & Stephan, 1985,1996a). To comprehend and address prejudice toward the culturally different,scholars need to examine the broad range of intergroup emotions that influenceprejudicial attitudes.

References

Allport, G. (1954). The nature of prejudice. Cambridge: Addison-Wesley.

Amir, Y. (1976). Contact hypothesis in ethnic relations. Psychological Bulletin, 71, 319–342.

Biernat, M., & Crandall, C. S. (1994). Creating stereotypes and capturing their content. European Journal

of Social Psychology, 26, 867–898.

Bond, M. H. (1986). Mutual stereotypes and the facilitation of interaction across cultural lines.

International Journal of Intercultural Relations, 10, 259–276.

Brown, K. (1988). Effects of perceived country of origin, educational status, and native speakerness on

American college student attitudes toward non-native instructors. Unpublished Doctoral Dissertation,

University of Minnesota, DAI: 49:7, 1710A.

Davis, T. (Ed.). (1998). Open doors: 1997/98 Report on international educational exchange. Institute of

International Education, New York.

Dijker, A. J. M. (1987). Emotional reactions to ethnic minorities. European Journal of Social Psychology,

17, 305–325.

Dodd, C. H. (1995). Dynamics of intercultural communication (4th ed.). Madison: Brown & Benchmark.

Dovidio, J. F., Brigham, J. C., Johnson, B. T., & Gaertner, S. L. (1996). Stereotyping, prejudice, and

discrimination: Another look. In C. N. Macrae, C. Stangor, & M. Hewstone (Eds.), Stereotypes and

stereotyping. New York: Guilford.

Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Fort Worth, TX: Harcourt Brace

Jovanovich.

Eagly, A. H., & Mladinic, A. (1989). Gender stereotypes and attitudes toward women and men.

Personality and Social Psychology Bulletin, 15, 543–01558.

Edwards, J. R. (1982). Language attitudes and their implications among English Speakers. In E. Ryan, &

H. Giles (Eds.), Attitudes towards language variation: social and applied contexts (pp. 20–33). London:

E. Arnold.

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 629

Esses, V. M., Haddock, G., & Zanna, M. P. (1993). Values, stereotypes, and emotions as determinants of

intergroup attitudes. In D. M. Mackie, & D. L. Hamilton (Eds.), Affect, cognition and stereotyping:

Interactive processes in group perception (pp. 137–166). New York: Academic Press.

Evans, M. G. (1991). The problem of analyzing multiplicative composites: Interactions revisited. American

Psychologist, 46, 6–15.

Giles, H., & Johnson, P. (1981). The role of language in ethnic group relations. In J. Turner, & H. Giles

(Eds.), Intergroup behavior (pp. 199–243). Chicago: University of Chicago Press.

Giles, H., & Robinson, W. P. (Eds.). (1990). Handbook of language and social psychology. Chichester,

England: Wiley.

Gudykunst, W. B. (Ed.). (1986). Intergroup communication. Baltimore: E. Arnold.

Gudykunst, W. B., & Hammer, M. R. (1988). Strangers and hosts: An uncertainty reduction based theory

of intercultural adaptation. In Y. Y. Kim, & W. B. Gudykunst (Eds.), Cross-cultural adaptation:

Current approaches (pp. 106–139). Newbury Park: Sage.

Hall, E. T. (1976). Beyond culture. Garden City, NJ: Anchor.

Hamilton, D. L., & Sherman, J. W. (1994). Stereotypes. In R. S. Wyer, & T. K. Srull (Eds.), Handbook of

social cognition: Basic processes (pp. 1–68). Hillsdale, NJ: Erlbaum.

Harman, R. C., & Briggs, N. E. (1991). SIETAR survey: Perceived contributions of the social sciences to

intercultural communication. International Journal of Intercultural Relations, 15, 19–28.

Kim, Y. Y. (Ed.). (1986). Interethnic communication: Current research. Newbury Park: Sage.

Lee, M. Y., Abd-Ellah, M., & Burks, L. A. (1981). Needs of foreign students from developing nations at us

colleges and universities. Washington, DC: National Association for Foreign Student Affairs.

Leong, F. T., & Chou, E. L. (1996). Counseling international students. In P. B. Pedersen, & J. G. Draguns

(Eds.), Counseling across cultures (4th ed.) (pp. 210–242). Thousand Oaks: Sage.

Levine, R. A., & Campbell, D. T. (1972). Ethnocentrism. New York: Wiley.

Lustig, M. W., & Koester, J. (1996). Intercultural competence: Interpersonal communication across cultures.

New York: Harper Collins.

Maio, G. R., Esses, V. M., & Bell, D. W. (1994). The formation of attitudes toward new immigrant

groups. Journal of Applied Social Psychology, 24, 1762–1776.

Martin, J. N. (1993). Intercultural communication competence. In R. L. Wiseman, & J. Koester (Eds.),

Intercultural communication competence (pp. 16–29). Newbury Park: Sage.

McCroskey, L. L. (1998). An examination of factors influencing US student perceptions of native and

non-native US teacher effectiveness. Paper presented at the annual meeting of the National

Communication Association, New York.

Mestenhauser, J. A. (1983). Learning from sojourners. In D. Landis, & R. W. Brislin (Eds.), Handbook of

intercultural training (pp. 153–186). Elmsford, NY: Pergamon.

Neuliep, J. W., & McCroskey, J. C. (1997). The development of intercultural and interethnic

communication apprehension scales. Communication Research Reports, 14, 145–156.

Paige, R. M. (1990). International students: Cross-cultural psychological perspectives. In R. W. Brislin

(Ed.), Applied cross-cultural psychology: Cross-cultural research and methodology series, Vol. 14

(pp. 367–382). Newbury Park: Sage.

Pedersen, P. B. (1991). Counseling international students. Counseling Psychologist, 19, 10–58.

Rokeach, M. (1968). Beliefs, attitudes, and values: A theory of organization and change. San Francisco:

Jossey-Bass.

Rubin, A. M. (1997). Intensive English programs are lucrative for universities: Attracting foreign students

to us institutions. Chronicle of Higher Education, 44, A48.

Rubin, D. L., & Smith, K. A. (1990). Effects of accent, ethnicity, and lecture topic on undergraduates’

perceptions of nonnative English-speaking teaching assistants. International Journal of Intercultural

Relations, 14, 337–48353.

Sodowsky, G. R., & Plake, B. S. (1992). A study of acculturation differences among international people

and suggestions for sensitivity to with-in group differences. Journal of Counseling and Development, 71,

53–59.

Spencer-Rodgers, J. (2001). Consensual and individual stereotypic beliefs about international students

among American host nationals. International Journal of Intercultural Relations, 25, 639–657.

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631630

Stangor, C., & Lange, J. E. (1994). Mental representations of social groups: Advances in understanding

stereotypes and stereotyping. In M. P. Zanna (Ed.), Advances in experimental social psychology, Vol. 26

(pp. 357–416). San Diego: Academic Press.

Stangor, C., Sullivan, L. A., & Ford, T. E. (1991). Affective and cognitive determinants of prejudice.

Social Cognition, 9, 359–391.

Stephan, W. G., & Stephan, C. W. (1985). Intergroup anxiety. Journal of Social Issues, 41, 157–176.

Stephan, W. G., & Stephan, C. W. (1996a). Predicting prejudice. International Journal of Intercultural

Relations, 20, 409–426.

Stephan, W. G., & Stephan, C. W. (1996b). Intergroup relations. Boulder, CO: Westview.

Stephan, W. B., Ybarra, O., & Bachman, G. (1999). Prejudice toward immigrants. Journal of Applied

Social Psychology, 29, 2221–2237.

Stephan, W. G., Ybarra, O., Martinez, C., Schwarzwald, J., & Tur-Kaspa, M. (1998). Prejudice toward

immigrants to Spain and Israel: An integrated threat theory analysis. Journal of Cross-Cultural

Psychology, 29, 559–576.

Wiseman, R. L., Hammer, M. R., & Nishida, H. (1989). Predictors of intercultural communication

competence. International Journal of Intercultural Relations, 13, 349–370.

Wiseman, R. L., & Koester, J. (Eds.). (1993). Intercultural communication competence. Newbury Park:

Sage.

Yook, L. E., & Albert, R. D. (1999). Perceptions of international teaching assistants: The interrelatedness

of intercultural training, cognition, and emotion. Communication Education, 48, 1–17.

Zanna, M. P., & Rempel, J. K. (1988). Attitudes: A new look at an old concept. In D. Bar-Tal, &

A. W. Kruglanski (Eds.), The social psychology of knowledge (pp. 315–334). Cambridge: Cambridge

University Press.

J. Spencer-Rodgers, T. McGovern / Int. J. of Intercultural Relations 26 (2002) 609–631 631