Investigating the Effect of Anxiety,Uncertainty and Ethnocentrism on Willingness to Interact in an Intercultural Communication

Abstract

The current research aims to explore the effect of low or high perceived predictability and anxiety on willingness to interact with a person from another cultural group. How differences in anxiety, both individual (trait anxiety) and intergroup (anxiety specifically related to an interaction), influence willingness to engage within an intercultural communication is investigated. In addition, the contribution of ethnocentrism to willingness to interact is explored. Anxiety, uncertainty, and ethnocentrism are all important factors that negatively affect willingness to interact in an intercultural communication. Yet to date, anxiety and uncertainty have been examined separately to ethnocentrism in the literature. The current study found that an anxiety-provoking intercultural interaction has a negative impact on willingness to interact with an intercultural interaction partner. Perceived predictability alone was not found to affect willingness to interact, contrary to previous research, with results suggesting that intergroup anxiety may be a better predictor of willingness to interact than perceived predictability and trait anxiety. Overall results indicate that anxiety and uncertainty are separate but related constructs in support of current theoretical models. Ethnocentrism was shown to uniquely contribute to willingness to interact. This novel finding indicates the importance of cultural factors on willingness to engage in intercultural communication and points to the need for further research to explore the impact of cultural values on these relationships.

Keywords

communication cultural psychology clinical/abnormal intergroup relations

Increasing cultural diversity in many societies has resulted in a growing need to better understand the processes that promote and inhibit intercultural communication (Van Oudenhoven, Ward, & Masgoret, 2006). Research suggests that the activity of new cultural contact and learning can often be accompanied by stress and intense emotion, manifested as uncertainty, confusion, and anxiety (Kim, 2008). Consequently, anxiety and uncertainty have been listed as factors that hinder cross-cultural engagement and lead to avoidance (Duronto, Nishida, & Nakayama, 2005).

During an intercultural communication, many individuals experience anxiety and related negative affect (W. G. Stephan & Stephan, 1985), thought to arise from the perception of threat or fear of negative appraisal, such as being embarrassed, ridiculed, or rejected (W. G. Stephan et al., 2002). Intergroup anxiety theory proposes that the anxiety experienced during intercultural interactions gives rise to both behavioral consequences, in that the anxiety will lead people to avoid or leave the interaction as soon as possible, and emotional consequences, such as intense feelings of discomfort, fear, resentment, or even indignation and disgust (W. G. Stephan & Stephan, 1985).

Uncertainty, defined here as the inability to predict or explain the behavior of others, is strongly linked to the expression of anxiety (Neuliep & Ryan, 1998). An intolerance of uncertainty, where ambiguous situations are perceived as threatening, is regarded as a significant risk factor for the expression of clinical anxiety (Gosselin et al., 2008; Norton, 2005). It is not surprising then that anxiety and uncertainty are regarded as related constructs during intercultural communication. Research suggests that when someone is confronted with an interaction with another person of a different ethnic or cultural background, they will experience communication apprehension (MacIntyre, Baker, Clément, & Donovan, 2002; Neuliep & Ryan, 1998). Specifically, when the uncertainty associated with an intercultural communication increases, and the interaction partner is viewed as less predictable, anxiety is thought to increase (Duronto et al., 2005; Greco & Roger, 2003).

Uncertainty reduction theory (Berger & Calabrese, 1975) states that one will try to minimize the aversive state of uncertainty brought about by initial communications with a stranger, through the utilization of key relationship variables, such as reciprocity and self-disclosure. Berger provides several axioms and theorems to explain how the level of uncertainty during initial interactions will either increase or decrease depending on the amount of interpersonal communication or key relationship variables that are utilized. Uncertainty reduction theory has been more recently adapted to apply to intercultural interactions specifically.

Gudykunst’s Anxiety Uncertainty Management (AUM) theory similarly postulates that managing levels of both anxiety and uncertainty within an intercultural communication is central to the effectiveness of the communication (Gudykunst & Nishida, 2001) and intercultural adjustment (Gao & Gudykunst, 1990). AUM theory argues that if anxiety and uncertainty are either too high or low, communication will not occur. High levels of anxiety and uncertainty are thought to lead to avoidance of intercultural communication or aversive communication when avoidance is not possible (Gudykunst, 2005). This notion is supported by evidence that high levels of both anxiety and uncertainty predict avoidance within an intercultural communication (Duronto et al., 2005) or less effective communication (Hubbert, Gudykunst, & Guerrero, 1999). However, lower levels of anxiety and uncertainty may lead to a lack of motivation to engage.

More recently, research exploring avoidance, or unwillingness to communicate, has separated anxiety into two categories: incidental anxiety, the individual’s affective experience of anxiety (or trait anxiety), and integral anxiety, the anxiety experienced within an intergroup situation specifically (Samochowiec & Florack, 2010). Samochowiec and Florack (2010) manipulated both the perceived predictability of an intercultural interaction partner, as a measure of uncertainty, and the emotion state of the participants, as a measure of incidental anxiety. The researchers found that participants generally reported less willingness to engage with an interaction partner when the interaction partner was perceived as unpredictable, and that those who experienced more anxiety reported less willingness to interact with an unpredictable interaction partner specifically. No significant relationship was reported between willingness to interact and integral (or intergroup) anxiety; however, it is worth noting that within the study design, the measure of integral anxiety did not refer to the specific experimental interaction but relied on a measure based on each participant’s recall of anxiety felt during a previous intercultural experience.

To further define the process of anxiety and uncertainty in intercultural settings, a number of changes could be made to Samochowiec and Florack’s (2010) experiment. As incidental anxiety is arguably a measure of disposition to experience anxiety, it may be better encapsulated by a clinically validated measure of anxiety. Other research has used measures of individual disposition, such as personality, and found that when an intercultural communication was manipulated to appear more stressful, individuals with high levels of adaptation and openness, defined by the authors as intercultural traits, experienced less negative affect (van der Zee, van Oudenhoven, & de Grijs, 2004). Therefore, incidental anxiety, measured as trait anxiety, may show a similar relationship with an individual’s responses to a stressful intercultural communication. Similarly, one could also measure the amount of uncertainty that an individual brings to a situation, through a trait measure of intolerance to uncertainty, to see whether a similar relationship is held.

It is also possible to measure anxiety within an intercultural communication and its impact on willingness to engage in two ways—first, as the affective experience of anxiety in an intergroup situation (via self-report) and second, by manipulating the degree of stress or anxiety associated with that interaction (via experimental manipulation). Moreover, by also manipulating the perceived predictability of this interaction partner, as has previously (Samochowiec & Florack, 2010) and more recently (Rohmann, Florack, Samochowiec, & Simonett, 2014) been implemented, one could then determine how perceived predictability, as a measure of uncertainty, affects the experience of anxiety and willingness to interact. Research has yet to investigate how different forms of anxiety, such as trait or intergroup, are individually related to perceived predictability or how they may uniquely affect willingness to engage in intercultural communication.

In considering this, it should be noted that research on willingness to engage in intercultural interactions has indicated that even when intercultural anxiety and uncertainty are low, communication may still be ineffective, due in part to the importance of other factors, such as ethnocentrism (Neuliep & Ryan, 1998). Ethnocentrism, which can be conceptualized as a preference for one’s own cultural standing or values (Lin, Rancer, & Trimbitas, 2005), can adversely affect harmonious intergroup relations (Bozumic, Duckitt, Papadic, Dru, & Krauss, 2009), has been linked to lower cultural sensitivity (Hammer, Bennett, & Wiseman, 2003; Lin, Rancer, & Lim, 2003), and can increase out-group negativity (Keles, 2013).

Ethnocentrism appears to be related to anxiety and uncertainty in intercultural communications. Ethnocentrism has been associated with increased anxiety in an intercultural setting (C. W. Stephan & Stephan, 1992), with the experience of anxiety and uncertainty arising from the ethnocentric sentiment of the communicators (Awang-Rozaimie, Sahari, & Ali, 2012). Furthermore, higher degrees of both ethnocentrism and communication apprehension are related to lower levels of willingness to engage in an intercultural communication (Lin & Rancer, 2003). More recent research has analyzed the direct relationship between ethnocentrism and intolerance of uncertainty in intercultural interactions and found that these constructs are associated (Cargile & Bolkan, 2013). Given there is a significant relationship between ethnocentrism, anxiety, and uncertainty, ethnocentrism is an important factor which must be considered when investigating willingness to engage in intercultural communication.

To date, the majority of research documenting interracial interactions and avoidance of intercultural contact has been conducted in America, comparing Anglo-American with African American participants (Shelton & Richeson, 2006). Little research has been conducted utilizing an Australian demographic, despite the fact that almost one quarter of Australian residents are born overseas (Van Oudenhoven et al., 2006). Given the high degree of cultural diversity, it is important to further explore intercultural communication within an Australian sample, exploring the factors that might influence willingness to engage in intercultural interactions.

The current study aims to add to previous research by assessing (a) the effect of a low- or high-anxiety-provoking intercultural interaction on willingness to interact; (b) the effect of low or high perceived predictability of an intercultural interaction partner on the willingness to interact; (c) the unique contribution that both individual differences in anxiety (trait), and the experience of anxiety within a cultural communication (intergroup) have on the willingness to interact, alongside perceived predictability; and (d) the contribution of ethnocentrism alongside trait anxiety and uncertainty measures on the willingness to interact with an intercultural communication partner.

Based on the current literature, we hypothesize the following:

Participants will be less willing to interact with the intercultural interaction partner when they have been assigned to the low versus high predictability condition and high- versus low-anxiety condition, with those assigned to both the low predictability and high-anxiety conditions being the least willing to interact.

Intergroup anxiety (self-reported) and perceived predictability will each make a unique contribution to willingness to interact.

Trait anxiety and perceived predictability will each make a unique contribution to willingness to interact.

Ethnocentrism, trait anxiety, and intolerance of uncertainty will each make a unique contribution, after controlling for intergroup anxiety, to willingness to interact with someone from a different cultural background.

Method

Participants and Design

The study was a 2 (predictability manipulation) × 2 (anxiety manipulation) design. Dependent variables were perceived predictability and willingness to interact. The Beck Anxiety Inventory (BAI; Beck & Steer, 1990) was used as a screening tool to stratify participants by low (scores falling within the mild–minimal range) or high anxiety (scores within the moderate–severe range), to ensure an equal distribution of anxiety symptoms across each of the four experimental conditions. Participants were then randomized to condition (within both the low- and high-anxiety groups) through the use of sequentially numbered opaque sealed envelopes (Doig & Simpson, 2005).

A total of 143 participants commenced the study. Three participants did not successfully complete all the manipulations, as they did not adequately follow instructions, and were excluded from further analyses, with one participant completing the trial but omitting some demographic information. The resulting 140 participants (94 females [67%]; M_age = 20 SD = 3.9) were undergraduate psychology students at the University of Sydney who received course credit for participation. Country of birth groupings (Australian Bureau of Statistics, 2011) indicated that participants represented a wide range of cultural or ethnic groups, including 81 from Australia, New Zealand, or Pacific Islands (termed Oceania); 28 Northeast Asian; 15 Southeast Asian; 2 Southern and Central Asian, 3 North African and Middle Eastern; 4 European; 2 Sub-Saharan African; and 5 people of the Americas. Of these participants, 92% (n = 132) had previously traveled overseas, 76% (n = 108) had at least one parent born outside of Australia, and 64% (n = 91) reported that English was their first language. The Human Research Ethics Committee at the University of Sydney granted ethics approval for the study. Exclusion criteria were current treatment for anxiety or disturbance in mood.

Dependent Measures

Intergroup anxiety

The modified version of the Intergroup Anxiety Scale (W. G. Stephan & Stephan, 1985) is a six-item self-report questionnaire measuring how participants feel when interacting with a member from another cultural group, with each item rated on a 10-point Likert-type scale (0 = not at all; 10 = extremely). The anxiety-related terms assessed by the items are as follows: comfortable, uncertain, confident, awkward, anxious, and at ease. These items are reverse-scored where necessary and summed to generate a single score, with higher scores indicating increased anxiety. This measure reports high internal consistency (Cronbach’s α = .83-.92) in previous samples (W. G. Stephan et al., 2002) and within the current sample (Cronbach’s α = .81).

Trait anxiety

The State Trait Anxiety Inventory (STAI) for Adults trait measure (Spielberger, 1983) is a 20-item self-report questionnaire that measures trait levels of anxiety (e.g., “I feel nervous and restless”), rated on a 4-point Likert-type scale (1 = not at all; 4 = very much so). This measure has previously demonstrated high internal consistency (average Cronbach’s α > .89) and good test–retest reliability (average r = .88) in previous samples (Gros, Antony, Simms, & McCabe, 2007). Within the current sample, internal consistency was also high (Cronbach’s α = .87).

Intolerance of uncertainty

The Intolerance of Uncertainty Scale (IUS) is a 27-item self-report questionnaire (Freeston, Rhéaume, Letarte, Dugas, & Ladouceur, 1994), which assesses the degree to which one experiences uncertainty as intolerable (e.g., “Uncertainty stops me from living a full life), rated on a 5-point Likert-type scale (1 = not at all true; 5 = extremely true). This measure has shown excellent internal consistency (Cronbach’s α = .94) and good test–retest reliability (r = .74) within a previous sample (Buhr & Dugas, 2006). In the current sample, internal consistency was high (Cronbach’s α = .94).

Ethnocentrism

The revised Generalized Ethnocentrism Scale (GENE) assesses the degree to which participants judge another culture solely on the values and standards of their own culture (Neuliep & McCroskey, 1997). This self-report questionnaire contains 22 items, 7 of which act as filler items and are dropped, and 3 are reverse-scored, leaving a remaining 15 items which are balanced in the number of positively (e.g., “Lifestyles in other cultures are just as valid as those in my culture”) and negatively (e.g., “I have little respect for the values and customs of other cultures”) worded questions. The items are rated on a 5-point Likert scale (1 = disagree; 5 = strongly agree), with higher scores indicating higher ethnocentrism. This scale reports high internal consistency (Cronbach’s α = .84-.94) across previous samples (Neuliep, 2002) and adequate consistency within the current sample (Cronbach’s α = .77).

Willingness to interact

Willingness to interact with someone from another cultural background generally (at Time 1) and with an assigned intercultural interaction partner specifically (named “Yon” at Time 2 and 3) was assessed by four questions (“Would you be willing to spend half a day with this person/Yon?” “Could you imagine learning more about this person/Yon?” “If this person/Yon sought contact with you, would you respond?” “How much would you be interested in a dialogue with this person/Yon?”) rated on an 11-point Likert-type scale (0 = not at all; 10 = very much), which were then averaged to create a single score, with higher scores indicating more willingness to interact. This measure has high internal consistency (Cronbach’s α = .85) in previous research (Samochowiec & Florack, 2010) and within the current sample (Cronbach’s α = .90).

Perceived predictability

The perceived predictability of the intercultural interaction generally (at Time 1) and to an assigned intercultural interaction partner (at Times 2 and 3) was assessed by two questions (“How well can you predict this person’s/Yon’s behavior in different situations?” “How well can you predict this person’s/Yon’s reactions to your own behavior?”), rated on an 11-point Likert-type scale (0 = not at all; 10 = very much). These two scores were averaged to create a single score for perceived predictability, with high values indicating that Yon was perceived as more predictable. This measure has high internal consistency (Cronbach’s α = .81) in previous research (Samochowiec & Florack, 2010) and within the current sample (Cronbach’s α = .91).

Control variables

The Positive and Negative Affect Schedule (PANAS) was used to assess for current mood state across all conditions (Watson, Clark, & Tellegan, 1988). The PANAS consists of two 10-item self-report mood scales that measure positive (PA; for example, “Inspired”) and negative (NA; for example, “Distressed”) affect states, rated on a 5-point Likert-type scale (1 = very slightly or not at all; 5 = very much). The PANAS reports high internal consistency for both the PA (Cronbach’s α = .89) and NA (Cronbach’s α = .85) subscales (Crawford & Henry, 2004), and these were also high within the current sample (PA Cronbach’s α = .90; NA Cronbach’s α = .88).

Previous intercultural interaction information was also attained to adequately capture the sample characteristics and check for group differences. Participants were asked to report their level of contact, how likely they were to interact, and willingness to interact with someone from a different cultural background, both in the past and currently. All intercultural interaction variables were measured on an 11-point Likert-type scale (0 = none/not at all; 10 = a lot/very much).

Procedure

Recruitment and demographic information

Participants were recruited through an advertisement on an online student recruitment website. They were informed at the commencement of the study that they may interact with someone from a different cultural background and were given a paper-based copy of the BAI to be stratified for random assignment to an experimental condition. They then completed all self-report measures on a single computer in a survey format (Time 1; see Figure 1 for study procedure).

Figure 1.

A diagrammatic representation of measures and tasks completed by all participants during the three main time points within the study procedure.

Predictability manipulation

Participants were then informed that based on the demographic information they had provided, they would be matched with an intercultural interaction partner, named Yon. They were instructed that this matching process would take place while they completed a paper-based copy of the GENE. Participants were then sent a hyperlink for one of the four experimental conditions via student email. Participants were presented with 20 scenarios and asked to guess how they believed Yon would respond to each of these, or the behavior that Yon would expect from them, by choosing from a list of four responses. After completing all scenarios, participants were given feedback on whether their answers were right or wrong. In the low predictability condition, participants were informed that their responses were correct 20% of the time. In the high predictability condition, participants were informed that their responses were correct 80% of the time. To control for the impact of perceived performance on participant’s motivation or self-esteem, participants were told that the mean response for the low predictability condition was 25% ± 15% and that the mean response for the high predictability condition was 75% ± 15%. This procedure has been used in previous research (Samochowiec & Florack, 2010) and has been shown to successfully produce significant differences in perceived predictability between two groups.

Directly following the predictability manipulation, participants completed the perceived predictability, willingness to interact, and intergroup anxiety measures the second time (Time 2), using the same computer-based survey format.

Anxiety manipulation

Participants were assigned to either a high- or low-anxiety-provoking condition. Participants were informed that they were about to interact with Yon over the Internet, via live chat. They were then provided with written statements and told that these were direct feedback from previous participants who had interacted with the same person. In the high-anxiety condition, the interaction was defined as difficult and anxiety provoking, while in the low-anxiety condition the interaction was defined as easy and pleasant (see the appendix). These descriptions were based on research, indicating that specific factors of an interaction between cultural groups increase anxiety, including the knowledge of subjective culture, feared negative consequences, perception of difference, expectations, and amount of structure (W. G. Stephan & Stephan, 1985).

Directly following the anxiety manipulation, participants then completed the perceived predictability, willingness to interact, and intergroup anxiety measures the third time using the survey format (Time 3). Finally, all participants were informed that the study had used deception and that they would not be interacting with this person.

Results

Preliminary Analyses

Analysis of the results was conducted using SPSS program version 21. The significance level of all tests was set at p < .05. Data were initially analyzed to detect potential univariate outliers and assumptions of normality using the Kolmogorov–Smirnov statistic. A logarithm transformation was performed on the IUS variable to correct for significant skewness at p < .001. Chi-square analysis and ANOVA conducted to assess for condition differences on baseline measures reported no significant findings.

Manipulation check

Independent-samples t tests were run on both perceived predictability and intergroup anxiety measures to assess the impact of experimental manipulations. A significant condition difference directly following the predictability manipulation was found on the measure of perceived predictability at Time 2, t(138) = 10.43, p < .001; participants assigned to the low predictability condition reported their interaction partner to be less predictable (M = 3.17, SD = 1.88) compared with participants assigned to the high predictability condition (M = 6.23, SD = 1.59). A significant condition difference was also seen directly following the anxiety manipulation on intergroup anxiety at Time 3, t(138) = −3.64, p < .001, with participants assigned to a high-anxiety condition reporting more anxiety (M = 28.63, SD = 9.83) compared with those assigned to the low-anxiety condition (M = 23.01, SD = 8.34).

Tests of Hypotheses

Hypothesis 1

To test Hypothesis 1, two-way ANOVAs were run to explore the impact of the predictability and anxiety manipulations on willingness to interact (Time 3). Dummy coding was used to assign participants to both the predictability (0 = high predictability, 1 = low predictability) and anxiety (0 = low anxiety; 1 = high anxiety) manipulation conditions. In partial support of Hypothesis 1, a significant main effect was seen for anxiety condition on willingness to interact (Table 1), with participants assigned to both the low predictability and high-anxiety conditions reporting least willingness, as predicted (means displayed in Figure 2). However, contrary to expectations, the main effect between predictability condition and willingness to interact (Table 1) was not significant.

Table 1.

Two-Way ANOVA for Anxiety and Predictability Conditions and Standardized Regression Coefficients on Willingness to Interact.

	df	F	p	$η_{p}^{2}$
Anxiety and predictability conditions
Anxiety	(3, 136)	4.46	.036	.032
Predictability		0.81	.370	.006
Anxiety × Predictability		1.08	.302	.008
		β	t	p
Intergroup anxiety and perceived predictability
Intergroup anxiety		−.68	−9.09	<.001
Perceived predictability		−.09	−1.18	.241
Intergroup anxiety × Perceived predictability		.06	0.83	.406
Trait anxiety and perceived predictability
Trait anxiety		−.14	−1.69	.94
Perceived predictability		.21	2.59	.011
Trait anxiety × Perceived predictability		.14	1.70	.092
Ethnocentrism, trait anxiety, and intolerance of uncertainty
Step 1
Intergroup anxiety		−.61	−8.98	<.001
Step 2
Intergroup anxiety		−.57	−8.52	<.001
Ethnocentrism		−.32	−4.71	<.001
Trait anxiety		−.01	−0.05	.959
Intolerance of uncertainty		.02	0.22	.838
Ethnocentrism × Intergroup anxiety		−.13	−2.01	.047

Figure 2.

Willingness to interact with an intercultural interaction partner by anxiety and predictability conditions.

To explore this finding, the data were split by low or high predictability condition and independent t tests were run to test for differences between the high- and low-anxiety conditions, within each predictability condition, on willingness to interact. Analyses revealed a significant difference between anxiety conditions within the low predictability manipulation, t(68) = −2.15, p = .035, but not between anxiety conditions within the high predictability condition, t(68) = −.79, p = .433, on willingness to interact.

Hypothesis 2

Multiple regression was used to assess whether intergroup anxiety (Time 3) and perceived predictability (Time 3) made unique contributions to willingness to interact (Time 3), with all measures converted to standardized scores. The regression model significantly predicted willingness to interact, R² = .42, F(3, 136) = 32.75, p < .001; however, intergroup anxiety alone made a significant contribution (see Table 1), providing partial support for Hypothesis 2.

Hypothesis 3

Multiple regression was run to assess whether trait anxiety and perceived predictability (Time 3) made unique contributions to willingness to interact (Time 3), with all measures converted to standardized scores. The regression model significantly predicted willingness to interact, R² = .08, F(3, 136) = 4.385, p < .001, with perceived predictability alone making a significant contribution (see Table 1), providing partial support for Hypothesis 3.

Hypothesis 4

Hierarchical multiple regression was used to assess the contribution of ethnocentrism, trait anxiety, and intolerance of uncertainty on willingness to interact (Time 3), after controlling for the influence of intergroup anxiety (Time 3), with all measures converted to standardized scores. Intergroup anxiety (Time 3) was entered at Step 1, making a significant contribution to willingness to interact (Time 3), R² = .38, F(3, 133) = 80.62, p < .001. After entering ethnocentrism, trait anxiety, and intolerance of uncertainty at Step 2, the model remained significant, R² = .47, F(3, 129) = 24.67, p < .001. Intergroup anxiety (Time 3), ethnocentrism, and the interaction between ethnocentrism and intergroup anxiety (Time 3) each made a significant contribution to the model (see Table 1) providing partial support for Hypothesis 4.

Discussion

The current study aimed to explore how perceived predictability of an intercultural interaction partner and an anxiety-provoking intercultural interaction affect participant’s willingness to engage in a dialogue with someone from a different cultural background. This is the first study to experimentally manipulate both the predictability and anxiety associated with an intercultural interaction within an experimental paradigm. The study also aimed to explore the relationship between two forms of anxiety (trait and intergroup), intolerance of uncertainty, perceived predictability, and willingness to interact. This study adds to previous research by further exploring the effect of different aspects of anxiety on intercultural interactions. In addition, the study aimed to examine the unique contribution of ethnocentrism relative to anxiety and uncertainty to add to current theoretical models of intercultural contact.

In support of Hypothesis 1, participants assigned to a high-anxiety-provoking condition reported less willingness to interact. However, contrary to expectations, those assigned to a low predictability condition did not report less willingness to interact compared with those assigned to a high predictability condition. Post hoc analyses revealed a significant difference in willingness to interact between high- and low-anxiety conditions only when considering those assigned to the low predictability condition, indicating that anxiety was more important in predicting willingness to interact when predictability was low compared with when it was high. This post hoc finding suggests that with a larger sample size, a significant interaction effect between anxiety and predictability on willingness to interact may have been found.

The results indicated that intergroup anxiety and perceived predictability significantly contributed toward willingness to interact; however, only intergroup anxiety made a unique contribution, providing partial support for Hypothesis 2. Similarly, Hypothesis 3 was partially supported, in that while both trait anxiety and perceived predictability significantly predicted willingness to interact, perceived predictability alone made a significant contribution. Finally, the study hypothesized that ethnocentrism, trait anxiety, and intolerance of uncertainty would each contribute to willingness to interact after controlling for intergroup anxiety (given this variable had been subject to manipulation). Again this finding was partially supported, with ethnocentrism, intergroup anxiety, and the interaction between ethnocentrism and intergroup anxiety each providing a unique contribution to the model.

The findings of the present study highlight that the degree to which anxiety is provoked in an intercultural situation is integral to willingness to engage in communication. This finding supports AUM theory (Gudykunst, 2005), which states that when anxiety regarding an intercultural communication is high, people will avoid intercultural communication. The current study did not find that higher degrees of uncertainty alone were directly related to willingness to interact, as AUM would suggest. However, post hoc analyses indicated that when uncertainty is high, anxiety may be a more important predictor of willingness to engage. This relationship has been supported in previous research by Samochowiec and Florack (2010). In this way, the current study highlights that both uncertainty and anxiety are important, interrelated, yet separate constructs, which must be managed successfully for engagement to occur.

A notable aspect of the current study was the way in which both trait and intergroup anxiety were analyzed separately, to better understand how different anxiety modalities affect willingness to interact. When trait anxiety was considered with predictability, predictability alone made a unique contribution to willingness to interact, suggesting that in this scenario the uncertainty of the interaction partner was more important. However, the opposite relationship was seen when considering the anxiety that is experienced within an intercultural situation alongside predictability. In this instance, intergroup anxiety was the only significant predictor of willingness to interact. These findings highlight that the anxiety regarding an intercultural communication specifically may be more influential in predicting willingness to interact, over and above the anxiety an individual may bring with them to a situation in the form of a trait or temperament.

Another unique aspect of the study was to investigate the contribution of ethnocentrism to willingness to interact in an intercultural interaction. After controlling for the effect of intergroup anxiety, ethnocentrism made a significant contribution to the model investigating willingness to interact, while both trait anxiety and intolerance of uncertainty did not. This finding is consistent with literature that highlights the importance of ethnocentrism on engagement in intercultural contact (Lin & Rancer, 2003).

Despite the significant findings, the current study has a number of limitations that need mention. As the study used an anticipated interaction, caution must be exercised when comparing these findings with an interaction that is currently happening or one that has previously occurred. In addition, as the study utilized volunteer participants, a sampling bias may have occurred, where only those who were generally willing to interact chose to participate. The willingness to interact with a person from a different culture may also have been high in the youthful university sample. When utilizing a convenience sample, any attempts to apply findings to the general population should be done with caution.

The study was designed in such a way as to measure the perceived predictability initially, followed by the anxiety of the same intercultural interaction partner. Although participants were prompted to respond 3 times throughout the study on perceived predictability and intergroup anxiety, there was a low likelihood that participants were motivated to provide purely consistent responses, given the significant pre- and post-manipulation effects. However, it is likely that the later anxiety manipulation was influenced by predictability. As anxiety and predictability are related constructs, it is possible that within this experimental design a relatively pure predictability manipulation was followed by one that manipulated both anxiety and predictability. As such, the anxiety manipulation regarding the interaction partner may have had a direct effect on the perceived predictability of the interaction. However, despite these difficulties, the findings do provide partial support for AUM theory, in that the addition of heightened anxiety within an intercultural communication can lead to less willingness to interact.

Another possible limitation is the way in which the intercultural interaction partner was utilized as a manipulation within the study. The majority of the current sample was female; however, the interaction partner was male. Although previous research has shown no significant gender effects while also reporting a female majority sample and utilizing a male interaction partner (Samochowiec & Florack, 2010), other research suggests that gender influences the way in which people respond to initial interactions (Lin et al., 2003; MacIntyre et al., 2002). Consequently, more recent research (Rohmann et al., 2014) has adapted the predictability manipulation to match the gender of each participant to rectify this possible confound. Future research may benefit from similarly adapting this predictability task to match the gender of participants or make the interaction partner gender neutral to avoid this difficulty.

Given that significant differences in intergroup anxiety are shown to influence uncertainty (measured by perceived predictability) and willingness to interact culturally within an experimental setting, future research should aim to investigate these factors in a real-world setting. Given the increasing multiculturalism seen globally, it is important to further explore how anxiety and uncertainty affect real-world intercultural interactions. In addition, given that ethnocentrism has been reported to be a defining factor of willingness to interact, further research should be undertaken to better understand the influence of cultural values on engagement within intercultural communication.

Conclusion

This study has assisted in further defining the processes of engagement within an intercultural communication and provides partial support for intercultural communication models. Results indicate that when an intercultural situation appears anxiety provoking, one is less willing to interact. The higher anxiety that is experienced by an individual, regardless of how anxiety provoking a situation is, the less willing that person will be to engage in intercultural communication. Personal attributes of anxiety and uncertainty that one brings to an intercultural communication, such as trait anxiety or intolerance of uncertainty, are less predictive of engagement. However, this study has shown that ethnocentrism is an important factor that significantly affects willingness to engage within this context and needs to be further investigated.

Footnotes

Appendix

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Australian Bureau of Statistics. (2011). Australian standard classification of cultural and ethnic groups (2nd ed.). Canberra: Author.

Awang-Rozaimie

A.-S.

Sahari

S.-H.

Ali

A. J.

(2012). Toward harmonious urban communities from the inside out: Validating a Multicultural Awareness Scale. Procedia: Social and Behavioral Sciences, 36, 732-741. Retrieved from http://dx.doi.org/10.1016/j.sbspro.2012.03.080

Beck

A. T.

Steer

R. A.

(1990). Manual for the Beck Anxiety Inventory. San Antonio, TX: Psychological Corporation.

Berger

C. R.

Calabrese

R. J.

(1975). Some exploration in initial interaction and beyond: Toward a developmental theory of communication. Human Communication Research, 1, 99-112. doi:10.1111/j.1468-2958.1975.tb00258.x

Bozumic

Duckitt

Papadic

Dru

Krauss

(2009). A cross-cultural investigation into a reconceptualization of ethnocentrism. European Journal of Social Psychology, 39, 871-899. doi:10.1002/ejsp.589

Buhr

Dugas

M. J.

(2006). Investigating the construct validity of intolerance of uncertainty and its unique relationship with worry. Journal of Anxiety Disorders, 20, 222-236. doi:10.1016/j.janxdis.2004.12.004

Cargile

A. C.

Bolkan

(2013). Mitigating inter- and intra-group ethnocentrism: Comparing the effects of culture knowledge, exposure, and uncertainty intolerance. International Journal of Intercultural Relations, 37, 345-353. Retrieved from http://dx.doi.org/10.1016/j.ijintrel.2012.12.002

Crawford

J. R.

Henry

J. D.

(2004). The Positive and Negative Affect Schedule (PANAS): Construct validity, measurement and normative data in a large non-clinical sample. The British Psychological Society, 43, 245-265. doi:10.1348/0144665031752934

Doig

G. S.

Simpson

(2005). Randomization and allocation concealment: A practical guide for researchers. Journal of Critical Care, 20, 187-193. doi:10.1016/j.jcrc.2005.04.005

10.

Duronto

P. M.

Nishida

Nakayama

S.-i.

(2005). Uncertainty, anxiety, and avoidance in communication with strangers. International Journal of Intercultural Relations, 29, 549-560. doi:10.1016/j.ijintrel.2005.08.003

11.

Freeston

Rhéaume

Letarte

Dugas

M. J.

Ladouceur

(1994). Why do people worry? Personality and Individual Differences, 17, 791-802. doi:10.1016/0191-8869(94)90048-5

12.

Gao

Gudykunst

W. B.

(1990). Uncertainty, anxiety, and adaptation. International Journal of Intercultural Relations, 14, 301-317. Retrieved from http://dx.doi.org/10.1016/0147-1767(90)90017-Q

13.

Gosselin

Ladouceur

Evers

Laverdiere

Routhier

Tremblay-Picard

(2008). Evaluation of intolerance of uncertainty: Development and validation of a new self-report measure [Comparative Study Research Support, Non-U.S. Gov’t]. Journal of Anxiety Disorders, 22, 1427-1439. doi:10.1016/j.janxdis.2008.02.005

14.

Greco

Roger

(2003). Uncertainty, stress, and health. Personality and Individual Differences, 34, 1057-1068. Retrieved from http://dx.doi.org/10.1016/S0191-8869(02)00091-0

15.

Gros

D. F.

Antony

M. M.

Simms

L. J.

McCabe

R. E.

(2007). Psychometric properties of the State-Trait Inventory for Cognitive and Somatic Anxiety (STICSA): Comparison to the State-Trait Anxiety Inventory (STAI). Psychological Assessment, 19, 369-381. doi:10.1037/1040-3590.19.14.4.369

16.

Gudykunst

W. B.

(2005). An anxiety/uncertainty management (AUM) theory of effective communication: Making the mesh of the net finer. In Gudykunst

W. B.

(Ed.), Theorizing about intercultural communication (pp. 281-322). Thousand Oaks, CA: Sage.

17.

Gudykunst

W. B.

Nishida

(2001). Anxiety, uncertainty, and perceived effectiveness of communication across relationships and cultures. International Journal of Intercultural Relations, 25, 55-71. Retrieved from http://dx.doi.org/10.1016/S0147-1767(00)00042-0

18.

Hammer

M. R.

Bennett

M. J.

Wiseman

(2003). Measuring intercultural sensitivity: The intercultural development inventory. International Journal of Intercultural Relations, 27, 421-443. Retrieved from http://dx.doi.org/10.1016/S0147-1767(03)00032-4

19.

Hubbert

K. N.

Gudykunst

W. B.

Guerrero

S. L.

(1999). Intergroup communication over time. International Journal of Intercultural Relations, 23, 13-46. Retrieved from http://dx.doi.org/10.1016/S0147-1767(98)00024-8

20.

Keles

(2013). What intercultural communication barriers do exchange students of Erasmus program have during their stay in Turkey, Mugla? Procedia: Social and Behavioral Sciences, 70, 1513-1524. Retrieved from http://dx.doi.org/10.1016/j.sbspro.2013.01.219

21.

Kim

Y. Y.

(2008). Intercultural personhood: Globalization and a way of being. International Journal of Intercultural Relations, 32, 359-368. Retrieved from http://dx.doi.org/10.1016/j.ijintrel.2008.04.005

22.

Lin

Rancer

(2003). Ethnocentrism, intercultural communication apprehension, intercultural willingness-to-communicate, and intentions to participate in an intercultural dialogue program: Testing a proposed model. Communication Research Reports, 20, 62-72.

23.

Lin

Rancer

Lim

(2003). Ethnocentrism and intercultural willingness to communicate: A cross-cultural comparison between Korean and American College Students. Journal of Intercultural Communication Research, 32, 117-128.

24.

Lin

Rancer

Trimbitas

(2005). Ethnocentrism and intercultural-willingness-to-communicate: A cross-cultural comparison between Romanian and US American College Students. Journal of Intercultural Communication, 34, 138-151.

25.

MacIntyre

P. D.

Baker

S. C.

Clément

Donovan

L. A.

(2002). Sex and age effects on willingness to communicate, anxiety, perceived competence, and L2 motivation among junior high school French immersion students. Language Learning, 52, 537-564. doi:10.1111/1467-9922.00194

26.

Neuliep

(2002). Assessing the reliability and validity of the generalized ethnocentrism scale. Journal of Intercultural Communication Research, 31, 201-216.

27.

Neuliep

McCroskey

J. C.

(1997). The development of a U.S. and generalized ethnocentrism scale. Communication Research Reports, 14, 385-398.

28.

Neuliep

Ryan

(1998). The influence of intercultural communication apprehension and socio-communicative orientation on uncertainty reduction during initial cross-cultural interaction. Communication Quarterly, 46, 88-99. doi:10.1080/01463379809370086

29.

Norton

P. J.

(2005). A psychometric analysis of the Intolerance of Uncertainty Scale among four racial groups [Comparative Study Evaluation Studies]. Journal of Anxiety Disorders, 19, 699-707. doi:10.1016/j.janxdis.2004.08.002

30.

Rohmann

Florack

Samochowiec

Simonett

(2014). “I’m not sure how she will react”: Predictability moderates the influence of positive contact experiences on intentions to interact with a host community member. International Journal of Intercultural Relations, 39, 103-109. Retrieved from http://dx.doi.org/10.1016/j.ijintrel.2013.10.002

31.

Samochowiec

Florack

(2010). Intercultural contact under uncertainty: The impact of predictability and anxiety on the willingness to interact with a member from an unknown cultural group. International Journal of Intercultural Relations, 34, 507-515. doi:10.1016/j.ijintrel.2010.05.003

32.

Shelton

J. N.

Richeson

J. A.

(2006). Interracial interactions: A relational approach. Advances in experimental social psychology, 38, 121-181. doi: 10.1016/S0065-2601(06)38003-3 .

33.

Spielberger

C. D.

(1983). Manual for the State-Trait Anxiety Inventory (Form Y). Palo Alto, CA: Mind Garden.

34.

Stephan

C. W.

Stephan

W. G.

(1992). Reducing intercultural anxiety through intercultural contact. International Journal of Intercultural Relations, 16, 89-106. Retrieved from http://dx.doi.org/10.1016/0147-1767(92)90007-H

35.

Stephan

W. G.

Boniecki

K. A.

Ybarra

Bettencourt

Ervin

K. S.

Jackson

L. A.

. . . Renfro

C. L.

(2002). The role of threats in the racial attitudes of Blacks and Whites. Personality and Social Psychology Bulletin, 28, 1242-1254. doi:10.1177/01461672022812009

36.

Stephan

W. G.

Stephan

C. W.

(1985). Intergroup anxiety. Journal of Social Issues, 28, 1242-1254. doi:10.1177/01461672022812009

37.

van der Zee

van Oudenhoven

J. P.

de Grijs

(2004). Personality, threat and cognitive and emotional reaction to stressful intercultural situations. Journal of Personality, 72, 1069-1096. doi:10.1111/j.0022-3506.2004.00290.x

38.

Van Oudenhoven

J. P.

Ward

Masgoret

A.-M.

(2006). Patterns of relations between immigrants and host societies. International Journal of Intercultural Relations, 30, 637-651. Retrieved from http://dx.doi.org/10.1016/j.ijintrel.2006.09.001

39.

Watson

Clark

L. A.

Tellegan

(1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063-1070. doi:10.1037/0022-3514.54.6.1063