Balancing the Double-Edged Sword of Political Correctness: The Role of Politically Correct Communication Behaviour
in Ethnically and Culturally Diverse Teams
Siri H. Hollekim Haaland 10602240
Master Thesis
Graduate School of Communication
Communication Science: Corporate Communication
Supervisor: dr. Joep Hofhuis
Abstract
When teams are characterized by ethnic and cultural diversity, work processes may suffer due to the dilemma of political correctness, which captures the difficult balance between being culturally sensitive, and being overly culturally sensitive. The current study developed and tested two scales to assess politically correct (PC) communication behaviour on an individual level, in interactions with ethnically and culturally dissimilar colleagues, i.e. out-group members, and in interactions with ethnically and culturally similar colleagues, i.e. in-group members. Based on quantitative data provided by employees working in Norway (N = 221), three communication characteristics with corresponding subscales were identified within both out-group and in-group interactions; normative behaviour, topic avoidance and disagreement avoidance. In the case of out-group PC communication behaviour, results show that pro-diversity employees increasingly adopt normative behaviour, while both employees with negative diversity beliefs and employees with positive diversity beliefs adopt disagreement avoidance. Among pro-diversity employees, disagreement avoidance is increasingly adopted by low cultural empathy individuals, and decreasingly adopted by high cultural empathy individuals. Employees who display topic avoidance and disagreement avoidance perceive decreasing quality of communication in interactions with out-group members. In in-group interactions, positive diversity beliefs increase topic avoidance and normative behaviour, while negative diversity beliefs negatively influence normative behaviour. In the case of topic avoidance, the positive association is stronger for low cultural empathy individuals.
Disagreement avoidance is also in in-group interactions increasingly adopted by both pro-diversity employees and employees holding negative pro-diversity beliefs. In-group disagreement avoidance positively predicts in-group perceptions of person-related and task-related
communication, while in-group topic avoidance positively influences in-group task-related communication.
Balancing the Double-Edged Sword of Political Correctness: The Role of Politically Correct Communication Behaviour in Ethnically and Culturally Diverse Teams
A diverse workforce is by many organisations regarded as a competitive advantage, as a broadened pool of experiences and perspectives stimulates creativity and innovation (Pless & Maak, 2004). However, when diversity is salient it may stifle important communication processes, as people perceive a need to regulate their communication behaviour in
heterogeneous settings (Chatman, 2010). The current study posits that members of ethnically and culturally diverse teams struggle with the dilemma of political correctness, which
captures the difficult balance between being sensitive and being overly sensitive to different others (Chatman, Goncalo, Kennedy & Duguid, 2012). Ely, Meyerson and Davidson (2006) dubbed political correctness a double-edged sword, as it promotes sensitivity and inclusion, but can potentially hinder development of effective relationships. Despite its potential salience in diverse settings, the complex nature of political correctness is, to this study’s knowledge, yet to be explored as individual-level communication behaviour in ethnically and culturally heterogeneous teams. The present study thus aims to bring new insights into the interpersonal dynamics of workplace communication, by exploring the communication
characteristics of political correctness and present a self-report instrument aimed at measuring individuals’ politically correct (PC) communication behaviour in interactions with both demographically dissimilar, and demographically similar, colleagues.
Previous research has emphasised prejudice attitudes as important antecedents to communication behaviour in diverse settings (e.g. Dunton & Fazio, 1997). However, diversity beliefs may function as a stronger predictor of communication behaviour in diverse teams as it captures employees’ perceptions of diversity as instrumental or detrimental to the social and professional environment (Hofhuis, van der Zee & Otten, 2013). However, at this stage, research has not given attention to diversity beliefs as influential of specific communication
behaviours, and the current study investigates the influence of diversity beliefs on PC communication behaviour. With respect to this relationship, the moderating role of ability to empathise with culturally different others, i.e. cultural empathy, is examined.
PC communication behaviour is further explored as a predictor of members’
perceptions of communication in their team. As an established norm of cultural sensitivity, political correctness has been found to facilitate free exchange of ideas by reducing
uncertainty (Goncalo, Chatman, Deguid and Kennedy, 2014). However, there is a lack of knowledge on how employees deal with the dilemma of political correctness and how subsequent communication behaviour affect their involvement in communication processes.
This paper follows the course of the presented research. The theoretical framework offers relevant insights into interpersonal communication in diverse environments, and a literature review of political correctness and related theoretical concepts (e.g. Chatman et al., 2012; Andrews, 1996), which underpins the current concept of PC communication behaviour. Hypotheses are presented based on literature and research on e.g. communication behaviour in interethnic interactions (e.g. Dunton & Fazio, 1997), employees’ diversity beliefs (e.g. Hofhuis et al., 2013), and workplace communication processes (e.g. Dinsbach, Feij & de
Vries, 2007). Hypotheses are tested with quantitative data provided by employees in Norway.
Social Categorization in Diverse Teams
When faced with diversity, individuals assign social categories based on immutable characteristics, such as gender and ethnicity. Through social categorization, people distinguish between own in-group, consisting of members perceived as similar to the self, and out-groups, consisting of members perceived as dissimilar to the self (Tajfel, 1974). The differentiation between out-group and in-group members is especially pertinent in ethnically diverse groups, as ethnicity is one of the most automatic dimensions people use to categorize others (Ito & Urland, 2003). Social categorization is argued as a major challenge for diverse teams, as
individuals tend to trust in-group members more than out-group members (Brewer, 1979), and find intergroup interactions ambiguous, causing uncertainty and anxiety (Stephan, 2014), and even hostility (Hentschel, Shemla, Wegge and Kearney, 2013). Consequently, team processes suffer as the negative consequences of categorization increase the risk of conflicts (Hentschel et al., 2013), and may impede members’ ability to communicate efficiently (Chatman, 2010). However, the negative effects of social categorization are not inevitable, but may depend on people’s attitudes toward the presence of demographically different others, and efforts to facilitate positive relations (Hentschel et al., 2013). Specifically, individuals may regulate their communication behaviour so to express sensitivity to differences and limit the risk of coming across as hostile (Chatman, 2010). However, as captured by the political correctness term, individuals often struggle with finding a balance between the positive intention of being sensitive, and being overly sensitive, which can potentially impede a person’s ability to communicate freely around both out-group and in-group members (Chatman et. al, 2012).
Political Correctness
The potential to engage in politically correct behaviour emerges when there are demographic differences in a group, and members perceive a need to regulate own communication so to not offend, or appear excluding towards, those they identify as out-group members (Chatman et al., 2012). Politically correct behaviour can also surface between ethnically and culturally similar individuals, as they worry about breaking perceived social norms, or feel a responsibility to promote inclusion of out-group members (Crandall,
Eshleman & O’Brien, 2002). In order to explore the role of political correctness in ethnically and culturally diverse teams, the first aim of this study is to establish an understanding of its specific communication behavioural characteristics, i.e. politically correct (PC)
communication behaviour, and develop instruments that can quantitatively measure the concept on an individual level in out-group and in-group interpersonal interactions.
Before presenting the current conceptualisation of PC communication behaviour, it is necessary to provide relevant insights into the equivocal nature of the arguments affiliated with the political correctness term. The modern version of so-called “political correctness” gained attention in North American media in the early 1990s, and was linked to activities on university campuses, and feminist, anti-racist and gay rights activist groups striving for social change. Consequently, it developed into an umbrella term associated with the left end of the political spectrum (Lalonde, Doan and Patterson, 2000). Academic literature offers multiple definitions, as the political correctness concept is applied differently in a diverse range of analyses and articles within different fields of research. A common theme is an emphasis on political correctness as the use of language or action that counteract intolerance that was or is prevailing towards certain societal groups (Halmari, 2011). Meaning, the use of language or action that intends to not cause any discrimination, exclusion, marginalization or violation towards gendered, ethnic or cultural groups (Andrews, 1996; Lalonde et al. 2000; Halmari, 2012), but promotes language or actions that “include, or will not offend, the broadest array of relevant identity groups” (Chatman et al., 2012, p. 163). However, as a norm, political
correctness has raised concerns relating to censorship, or as Hughes (2010) stated: “political correctness inculcates a sense of obligation or conformity in areas which should be (or are) matters of choice” (as cited in Hughes, 2012, p. 4). The political correctness debate is thus characterized by the balance between political correctness as a norm that discourage prejudice and promotes inclusion, and as a form of censorship that due to sensitivity to demographic differences can stifle the exchange of ideas and perspectives (Chatman et al., 2012).
PC Communication Behaviour in a Diverse Workplace
Being politically correct, i.e. adopting PC communication behaviour, is influenced by what the individual believes will not discriminate, exclude, marginalize or violate another person based on ethnic and cultural background (Andrews, 1996; Lalonde et al. 2000;
Chatman et al., 2012). In an ethnically diverse workplace, PC communication behaviour can emerge as a result of the positive intentions of sensitivity towards different others (Chatman et al. 2012), a motivation to avoid appearing prejudiced (e.g. Olson & Fazio, 2010), and a desire to ease intergroup interactions (Shelton, Richeson, Salvatore & Trawalter, 2005). However, as PC communication behaviour is the result of what an individual perceives to be appropriate, it may also result in overreaching the goal of sensitivity and inhibit the individual from acting and speaking freely around others (Chatman et al., 2012).
Internal and external motivations to adopt PC communication behaviour. In accordance with Plant and Devine’s (1998) research on an individual’s efforts to not appear prejudiced, PC communication behaviour is motivated by internal and external factors. When PC communication behaviour is internally motivated, it emerges due to a person’s
internalized standards, based on values and beliefs such as egalitarian and anti-bias ideals (Plant & Devine, 1998). In a diverse environment, internal motives trigger a commitment to communicate in a way that is in line with perceived anti-bias norms and the facilitation of an inclusive environment (Plant and Devine, 1998). In interactions with ethnically and culturally different colleagues, i.e. out-group interactions, PC communication behaviour is thus adopted to communicate in a way that is perceived as culturally sensitive and non-excluding (Crandall & Eshleman, 2003). In interactions with people perceived as ethnically similar, i.e. in-group interactions, the person may similarly feel a commitment to communicate in line with anti-bias values, and promote a non-prejudiced and non-excluding communication behaviour.
External motivations to adopt PC communication behaviour are conceived from the social environment due to perceived social norms, and concerns about one’s image and negative outcomes in interpersonal interactions (Crandall & Eshleman, 2003; Plant & Devine, 1998; Stephan, 2014). In out-group interactions, PC communication behaviour is thus adopted so to not appear prejudice (Major, Sawyer & Kunstman, 2013), and ease interactions by
limiting the risk of violating out-group members (Devine, Monteith & Zuwering, 1991). In in-group interactions, the individual may likewise be preoccupied with not coming across as prejudice, or violate the environment’s perceived anti-bias norms (Crandall et al., 2002).
The behavioural characteristics of PC communication behaviour. Existing
literature does not offer an unequivocal account of the behavioural consequences when people attempt to strike a balance between being sensitive, and being overly sensitive, in ethnically and culturally diverse environments. However, a literature review of empirical studies and theoretical articles, within the fields of e.g. organisational psychology and prejudice research, emphasised four central aspects of the political correctness norm (e.g. Halmari, 2011),
culturally sensitive language in the workplace (e.g. Ely et al., 2006) and communication behavioural tendencies in interethnic interactions (e.g. Harber, 1998). These four aspects form the conceptual framework on which the operationalization of out-group and in-group PC communication behaviour is based, and are as follows: (a) normative behaviour; (b) topic avoidance; (c) ignoring ethnic and cultural differences; and (d) disagreement avoidance.
Normative behaviour falls in line with political correctness as an anti-bias norm (Chatman et al., 2012), and includes refraining from using words, expressions and humour that can be considered culturally insensitive. Normative behaviour thus indicates an
individual’s self-censoring of words and expressions (Halmari, 2011; Andrews, 1996), and humour that relies on e.g. ethnic stereotypes, which has the perceived potential to violate ethnically and culturally different others (Leap & Smeltzer, 1984).
Ely et al. (2006) stressed the polarizing potential of certain topics in diverse
workplaces. The second behavioural characteristic of PC communication behaviour is thus topic avoidance. In out-group interactions, topic avoidance pertains to refraining from discussing controversial topics related to the interaction partner’s ethnicity and culture. In in-group interactions, topic avoidance concerns a general avoidance of controversial issues
concerning other ethnicities and cultures.
The third PC communication behaviour characteristic is the tendency to ignore ethnic and cultural differences, which is based on the concept of colour-blindness, defined by Sue (2007) as a propensity to “not acknowledge race or deny a person of colour’s racialized experience” (as cited in Neville, Awad, Brooks, Flores & Bluemel, 2013, p. 459). The current concept of ignoring ethnic and cultural differences centres on the colour-evasion aspect of colour-blindness, which is a “denial of potential racial differences by emphasizing sameness” so to not appear prejudiced (Neville et al., 2013, p. 457). In both out-group and in-group interactions, ignoring ethnic and cultural differences pertains to refraining from referring to a person’s ethnicity (Norton et al., 2006), and making mutual ethnic and cultural differences the foci in conversations.
Finally, in line with what Chatman et al. (2012) dubbed overly deliberate actions, disagreement avoidance is characterised by a tendency to refrain from opposing and criticising out-group members (Dunton and Fazio, 1997), and exhibit positive bias, i.e. preferential treatment (Harber, 1998). Disagreement avoidance can potentially be motivated by fear of disputes, or internalized egalitarian values motivating the individual to inhibit or promote certain behaviours so to make an out-group member not perceive hostility (Olson & Fazio, 2010). In in-group interactions, an individual may refrain from discussing disputes with out-group members, or present a positive bias towards their work performance.
Employees’ Attitudes Towards Diversity as Predictor of PC Communication Behaviour Empirical studies have found that both low and high prejudice individuals inhibit or promote certain behaviours to ease interethnic interactions (e.g. Olson & Fazio, 2010). However, high prejudice individuals tend to show stronger motivations to control prejudiced reactions and overcompensate so to not reveal their negativity (Dunton & Fazio, 1997). Although prejudice serves as an important predictor of interethnic communication behaviour,
diversity beliefs may constitute a more influential antecedent to PC communication behaviour in work teams, as it captures opinions about the presence of heterogeneous others in everyday work life (van Knippenberg, van Ginkel & Homan, 2013). Conceptually, diversity beliefs differ from prejudice by implying the instrumentality of diversity (Kauff, Ismmer & Nau, 2013). Pro-diversity employees regard diversity as beneficial to the social environment and instrumental to reaching organisational and group goals. In contrast, negative diversity beliefs imply concerns about out-group members as detrimental to own status in the workplace, existing organisational culture and organisational goals (Hofhuis et al., 2013). Although diversity beliefs are yet to be explored as antecedents to explicit communication behaviour in diverse work teams, it is posited that pro-diversity employees will display increasing PC communication behaviour in both out-group and in-group interactions, while the opposite will be true for employees with negative diversity beliefs.
Employees differ on whether diversity functions as an asset or as a disadvantage for the social and professional environment (Hofhuis et al., 2013). Pro-diversity employees recognize the benefits of diversity, and are thus likely more preoccupied with facilitating an inclusive environment in which colleagues with multiple backgrounds can work effectively together and reach their performance potential (Pless & Maak, 2004). In contrast, employees who perceive diversity as a threat to status quo feel discomfort and are less accepting of ethnically different others as co-workers (Montei, Adams & Eggers, 1996). Perceived instrumentality of diversity thus represents an external motivation to express inclusion
towards out-group members and promote inclusion among in-group members (Pless & Maak, 2004). Pro-diversity employees will likely increasingly adopt PC communication behaviour characteristics that emphasise sameness, i.e. ignoring ethnic and cultural differences, and are perceived as culturally sensitive, i.e. normative behaviour; non-excluding; e.g. topic
heterogeneous others as colleagues, will likely make less of an effort to communicate
inclusion through PC communication behaviour in both out-group and in-group interactions. Pro-diversity individuals feel personally and professionally stimulated in diverse environments (van Oudenhoven-van der Zee, Paulus, Vos & Parthasarathy, 2009), report increasing interest in intergroup interactions (Tropp and Bianchi, 2006), and emphasise the importance of intergroup contact (Adesokan, Ullrich, van Dick & Tropp, 2011). Pro-diversity employees will thus likely increasingly display PC communication behaviour characteristics, as it is perceived to facilitate positive interethnic interactions and relations. In contrast,
employees with negative diversity beliefs favour in-group members and may attempt to evade interactions with out-group members (van Oudenhoven-van der Zee et al., 2009; Stephan, 2014). Diversity thus assumes a polarizing effect, reducing attachment to work teams and increasing the likelihood of relationship conflicts (Hentschel et. al, 2013). This indicates that negative diversity beliefs decrease efforts to adopt PC communication behaviour so to avoid disputes and facilitate positive relations with out-group members. In in-group interactions, assumptions of deep-level similarities lead the employee to believe that he can speak freely and disregard the potential of conflicts over not adopting sensitized communication behaviour towards out-group members (Chatman, 2010).
Finally, positive diversity beliefs have previously been found to decrease hostility and discriminatory behavioural intentions toward out-group members on a societal level (Kauff and Wagner, 2012). Furthermore, Yoder (1991) argued that employees, who perceive out-group members as a threat, increasingly display discriminatory communication behaviour. The present study posits that employees with negative diversity beliefs likely make less of an effort to adopt a perceived culturally sensitive communication behaviour, while pro-diversity employees will be increasingly preoccupied with adopting perceived culturally sensitive communication behaviour.
H1a: Positive diversity beliefs positively predict out-group and in-group PC communication behaviour.
H1b: Negative diversity beliefs negatively predict out-group and in-group PC communication behaviour.
The Moderating Role of Cultural Empathy
People who are able to empathize with culturally different others’ behaviours and feelings, increasingly view diversity as a challenge rather than a threat and are thus better equipped to adjust in diverse settings (Hofhuis et al., 2013; Yakunina, Weigold, Weigold, Hercegovac & Elsayed, 2012). The ability to empathize with out-group members is captured by cultural empathy, defined by Ruben (1974) as “the capacity to clearly project an interest in others, as well as to obtain and to reflect a reasonably complete and accurate sense of
another’s thought, feelings and/or experiences” (as cited in van der Zee & van Oudenhoven, 2000, p. 293). This study has contended that negative diversity beliefs reduce individuals’ PC communication behaviour. However, people’s approach to different others is strongly
influenced by deep-level personality traits, such as cultural empathy (Hentschel et al., 2013). Pless and Maak (2004) argued that workplaces characterised by empathy is increasingly inclusive of multiculturalism. The current study posits that people who display strong cultural empathy are more inclined to pursue inclusion and sensitivity towards out-group members. Cultural empathy thus moderates the influence diversity beliefs have on PC communication behaviour. A pro-diversity employee’s PC communication behaviour will increase further if he displays strong cultural empathy. For employees who perceive diversity as a threat, strong cultural empathy will increase the probability of PC communication behaviour.
H2a: The positive relationships between positive diversity beliefs and out-group and in-group PC communication behaviour are moderated by cultural empathy -
the relationships will be stronger when employees display high rather than low cultural empathy.
H2b: The negative relationships between negative diversity beliefs and out-group and in-out-group PC communication behaviour are moderated by cultural empathy - the relationships will be weaker when employees display high rather than low cultural empathy.
PC Communication Behaviour and Content and Quality of Communication The instrumentality of workplace diversity lays in its broad pool of experiences, and perspectives, which can positively stimulate communication processes in both social and task-related settings (Pless & Maak, 2008). Team members’ participation in these communication processes is crucial to workplace socialization and team outcomes (Chao, O’Leary-Kelly, Wolf, Klein & Gardner, 1994). However, the negative consequences of social categorization, such as uncertainty in interethnic interactions, can potentially deter employees’ ability to communicate effectively in the exchange of task-related and person-related communication content (Dinsbach et al., 2007; Chatman, 2010). Categorization of out-group and in-group memberships can create social barriers across team members’ demographic differences, and consequently impede their ability to exchange important task-related information, and develop a shared understanding of its content through opinion sharing and quick feedback (Dinsbach et al., 2007; Dennis & Kinney, 1998). Social barriers between demographically different colleagues may further impede the sharing of personal issues, which negatively affects the development of interpersonal relationships (Dinsbach et al., 2007).
Pless and Maak (2004) stressed established norms as guidelines that can reduce uncertainty in interethnic workplace interactions. The positive effects of political correctness as a culturally sensitive norm were demonstrated by Goncalo, Chatman, Deguid and Kennedy (2014), who found that diverse teams produced significantly more ideas when this norm was
made salient, as it reduced diversity-related uncertainty. However, as communication behaviour, political correctness is induced by what people perceives to be appropriate based on multiple internal and external factors (Plant & Devine, 1998). Although PC
communication behaviour expresses the positive intention of cultural sensitivity, it can manifest itself as overly deliberate behaviour, which can inhibit an individual’s ability to participate in communication processes (Chatman et al., 2012).
PC communication behaviour and in-group communication. PC communication behaviour is likely more salient in out-group interactions than in in-group interactions, as people expect to be able to speak freely among those they perceive as like-minded (Chatman, 2010). However, in a diverse team, members may also regulate communication behaviour in in-group interactions, due to internalized egalitarian values, or perceived social norms (Crandall and Eshleman, 2003). Goncalo et al. (2014) found that while established political correctness norms facilitated idea exchange in diverse teams, it stifled the same process in homogeneous teams, as expectations of being able to communicate freely were violated. This study posits that the tendency of in-group PC communication behaviour may be more
indicative of reduced subgroup tensions between team members, as PC communication behaviour reduces perceived insensitive language, and negatively connoted topics about ethnicity and culture, and out-group members. In in-group interactions, team members who adopt PC communication behaviour, such as topic avoidance and disagreement avoidance, will likely perceive communication as increasingly effective and communication content as high, as PC communication behaviour reduces presence of non-relevant issues.
PC communication behaviour on out-group communication. While PC
communication behaviour is argued as beneficial for in-group communication, it is contended that the opposite is true for content and quality of communication between heterogeneous
team members. Contrary to the aforementioned findings by Goncalo et al. (2014), Ely and Thomas (2001) argued that enforcing culturally sensitive communication behaviour, e.g. normative behaviour, on employees is not in itself an instrumental link between diversity and work processes. This study argues that the dilemma of political correctness is especially difficult in out-group interactions, and it is likely the manifestation of political correctness as overly deliberate communication that will be most influential of communication processes between ethnically and culturally different colleagues (Barker, 2007).
Hofhuis, van der Zee and Otten (2012) found that a strong diversity climate reduces diversity-related conflicts, such as relationship conflicts and miscommunication. Diversity climate is characterised by the degree to which the “organisational climate facilitates the presence of cultural differences, and views this diversity as a positive asset” (Hofhuis et al., 2012, p. 969). A strong diversity climate promotes openness to diversity, specifically in regards to demographic differences and the problems these may cause. Employees are thus forced to recognize the perspectives of different colleagues, which can potentially reduce the negative effects of social categorization and intergroup barriers, such as diversity-related conflicts (Hofhuis et al., 2012). Diversity climate stands in contrast to the PC communication behavioural tendency of ignoring ethnic and cultural differences, which diverse companies have reported to promote so to limit the risk of relationship conflicts (Ely & Thomas, 2001). However, in interethnic interactions, appearing to not notice a person’s ethnicity has been found to impede task performance, as individuals who are reluctant to address ethnicity experience increasing uncertainty, which negatively affects their nonverbal communication behaviour (Norton et al., 2006). The findings by Hofhuis et al. (2012) and Norton et al. (2006) emphasise the potential detrimental effect an individual’s tendency to ignore ethnic and cultural differences can have on communication processes in out-group interactions. Ethnic and cultural differences and the potential problems they may cause become an unspoken
matter, which can reinforce social barriers between heterogeneous colleagues and thus impede quality of communication and the sharing of task-related and person-related content.
Finally, disagreement avoidance may best capture what Chatman et al. (2012) referred to as overly deliberate actions. Whether it is motivated by fear of conflict or a perceived need to appear inclusive, disagreement avoidance can obstruct communication processes,
especially quality of communication and the sharing of task-related content. If communication processes are characterised by reluctance to criticise or disagree with, and positive bias
towards, ethnically and culturally different colleagues, important aspects such as opinion sharing and efficient communication stand to suffer (Olson & Fazio, 2010). Furthermore, such overly deliberate behaviour may deter out-group members’ ability to efficiently participate in intergroup communication processes, as they perceive the behaviour to be disingenuous, causing uncertainty about their interaction partners’ motives (Major et al., 2013).
In sum, people who adopt PC communication behaviour in out-group interactions will perceive content and quality of communication as low, as PC communication behaviour impedes effective participation in these processes. Individuals, who adopt PC communication behaviour in in-group interactions, will perceive content and quality of communication as high, as PC communication behaviour reduces the presence of non-relevant issues.
H3a: Out-group PC communication behaviour negatively predicts out-group content and quality of communication.
H3b: In-group PC communication behaviour positively predicts in-group content and quality of communication.
As shown in figure 1, the present study explores the role of PC communication behaviour in out-group and in-group team interactions, by examining the influence of
diversity beliefs moderated by cultural empathy. PC communication behaviour is investigated as predictor of content and quality of communication with out-group and in-group members.
Methods Participants and Procedure
The current study was conducted by cross-sectional design, administered as an online questionnaire over a period of two weeks. Respondents were employees in Norway, recruited through various methods of sampling. First, respondents were approached via personal and professional networks, and snowball sampling. Second, a link to the questionnaire was posted in relevant communities and groups in social media, such as Facebook, LinkedIn and online forums. Third, several organisations were approached as potential providers of respondents.
After eliminating response sets and non-response, final sample consisted of 221 respondents. As reported in table 1, 60.4% were female, average age was 38.12, and the majority worked in public sector (59%). Respondents were highly educated with 49.8% having finished a master’s degree. 81.4% reported being born in Norway to Norwegian born parents. In regards to non-western minority members, defined in accordance with Statistics Norway (Høydahl, 2008), 4.1% had emigrated from a non-western country, while 2.3% were first generation Norwegians. 20.8% of the respondents reported that they had no ethnic or cultural diversity in their team. This constituted a substantial portion of the sample, and initial analyses were conducted to compare the results when excluding and including these
Figure 1. Relationships between diversity beliefs and out-group and in-group PC
communication behaviour, moderated by cultural empathy, and between out-group and in-group PC communication and perceived content and quality of communication.
respondents. As the analyses did not report significant differences, it was decided to include the respondents who reported no team diversity in the final sample (N = 221).
Developing the questionnaire. PC communication behaviour and three dependent communication variables were to be measured in both out-group and in-group interactions. Efforts were thus put into making the questionnaire as structured as possible, so to ensure that respondents understood the questionnaire and did not drop out before finishing.
Scales adopted from existing literature and the PC communication behaviour items were originally formulated in English. In accordance with Brislin (1970), the researcher translated the scales into Norwegian and recruited third parties, with high language
proficiency in both Norwegian and English, to conduct back-translations. Comparisons of the back-translations and the original English formulations resulted in minor changes to the final scale formulations, and the validity of the Norwegian translations was satisfactory. A pilot study was conducted to further test the quality of the questionnaire (Bryman, 2012). Results indicated that respondents understood instructions, questions and answer options. Based on feedbacks changes were made to certain formulations and design features. The original codebook in English, and the final Norwegian version, are included in Appendix A.
Table 1: Demographics of sample.
Frequency (%) M Mo SD Min. Max. n
Gender .49 217 Male 39.6 Female 60.4 Age 38.12 28.00 12.51 20 71 217 Sector .59 217 Private sector 33.2 Public sector 59.0 Other 7.8 Team diversity 2.88 3 1.38 1 5 221
Measures
Demographics. Eight variables gathered data on demographics; gender (1 = male, 2 = female), birth year, birth country, parents’ birth country, education level, sector and team diversity. An eight-item scale based on the Norwegian Standard Classification of Education assessed education level (Statistics Norway, 2003). Ethnic and cultural team diversity was measured on a five-point scale ranging from 1 = “My team is not at all diverse” to 5 = “My team is very diverse”. Variables do deviate from normal distribution, specifically education, which is negatively skewed, and age, which is positively skewed. Respondent and parents’ birth country was included to compare based on ethnic background. However, data collection did not result in a satisfactory sample of employees with non-Norwegian ethnicity.
Diversity Beliefs. The Benefits and Threats of Diversity Scale (BTDS) by Hofhuis et al. (2013) was used to measure diversity beliefs. In accordance with this study’s scope and based on original study’s analyses, one item was removed from each subscale prior to distribution of questionnaire.
Positive diversity beliefs. The BTDS includes five subscales measuring perceived benefits of diversity: (a) understanding groups in society, (b) creative potential, (c) social image, (d) job market, and (e) social environment (Hofhuis et al., 2013). In this study, subscales are analysed as one 15-item scale measuring positive diversity beliefs. A sample-item is: “Ethnic/cultural diversity…makes us more innovative”. Scale reliability is good, Cronbach’s alpha = .91.
Negative diversity beliefs. BTDS subscales measuring perceived threats of diversity includes (a) symbolic threat, (b) realistic threat, (c) intergroup anxiety, and (d) productivity loss (Hofhuis et al., 2013). Subscales are analysed as one 12-item scale measuring negative diversity beliefs. An example of an item is: “Ethnic/cultural diversity…leads to uncomfortable situations.” Scale reliability is good, Cronbach’s alpha = .91.
BTDS items were measured on a five-point Likert scale ranging from 1 = “strongly disagree” to 5 = “strongly agree”.
Cultural empathy. The eight-item short form of the cultural empathy scale by van der Zee, Oudenhoven, Ponterotto and Fietzer (2012) was included as a moderator, measuring self-reported ability to empathize with out-group members. A sample-item is: “I set others at ease.” Items were measured on a five-point Likert scale ranging from 1 = “strongly disagree” to 5 = “strongly agree”. Scale reliability is good, Cronbach’s alpha = .80.
Communication Outcomes.
Content of communication. Two scales by Dinsbach et al. (2007) measure perceived content of communication in out-group and in-group interactions. The four-item scale task-related communication measures perceptions of how often communication is about work-related topics. An out-group sample-item is: “Consider interactions with colleagues of a different ethnic/cultural background, from your own; how often would your communication at work be about...improving your skills?” Reliability of the task-related communication scales is good for both out-group (Cronbach’s alpha = .83) and in-group (Cronbach’s alpha = .87).
The four-item scale person-related communication measures employees’ perception of how often communication content is about personal issues. An example of an in-group item is: “Consider interactions with colleagues of the same ethnic/cultural background, as your own; how often would your communication at work be about…your spare time activities?” Scale for out-group (Cronbach’s alpha = .89) an in-group (Cronbach’s alpha = .90) is good.
Task- and person-related communication were measured on a five-point Likert scale ranging from 1 = “ very rarely” to 5 = “very often”.
Quality of communication. Quality of communication was assessed with a four-item scale based on a media richness construct by Dennis & Kinney (1998), measuring to what
degree employees’ perceive out-group and in-group communication as facilitating effective communication. An out-group sample-item is: “The way colleagues, of a different
ethnicity/culture, and I communicate, would allow us to...understand each other better.” Items were measured on a five-point Likert scale ranging from 1 = “strongly disagree” to 5 =
“strongly agree”. Scale reliability for out-group (Cronbach’s alpha = .88) and in-group (Cronbach’s alpha = .89) is good.
Control variables. Age and education are included as control variables, as their distributions were quite skewed. Analyses further control for team diversity, as respondents are included irrespective of team diversity. Gender is included as fourth control variable.
Out-group and In-group PC Communication Behaviour
Two 16-item scales were constructed for out-group and in-group PC communication behaviour based on the four communication characteristics; normative behaviour, topic avoidance, ignoring ethnic and cultural differences, and disagreement avoidance. Importance was put on out-group items reflecting the in-group items, and vice versa. Second, to
encourage respondents to answer based on how they would communicate, and not necessarily how they have communicated, items were formulated as hypothetical situations.
Each scale includes a total of six items measuring an individual’s normative behaviour; four were based on Dunton and Fazio’s (1997, p. 319) Motivation to Suppress Prejudiced Reactions scale, and literature addressing ethnic humour (e.g. Leap & Smeltzer, 1984). The remaining two items were based on Andrews (1996) and Halmari (2011), who stressed political correctness as censorship of perceived culturally insensitive expressions. Second, three items measures topic avoidance, inspired by Ely et al.’s (2006) article on potential triggers for conflicts in diverse work environments. Third, four items measure tendency to ignore ethnic and cultural differences. Items were based on literature and studies
on colour-blindness (e.g. Neville et al., 2013, Norton et al., 2006). Finally, three items were included to measure disagreement avoidance. Items assessing an individual’s tendency to suppress disagreement and criticism were adapted from the scale by Dunton and Fazio (1997, p. 319). The last item was based on a study by Harber (1998), which confirmed the presence of positive bias in evaluations of ethnically different others’ work performance.
Out-group PC communication behaviour. Reliability of the 16-item scale
measuring out-group PC communication behaviour is reasonable, Cronbach’s alpha = .76. A principal component analysis (PCA), with orthogonal (varimax) rotation, was conducted to explore underlying components. The analysis confirmed reversed items, and are reported with factor loadings after reversal. Four factors had eigenvalues above 1 and in combination
explained 51.32% of the variance, with a clear point of inflexion after the fourth component in the scree plot. The PCA results and Cronbach’s alphas are reported in table 2 (Appendix B).
In-group PC communication behaviour. Reliability of the 16-item in-group PC communication behaviour scale is good, Cronbach’s alpha = .83. A PCA, with orthogonal (varimax) rotation, was conducted. Reversed items were confirmed, and are reported with factor loadings after reversal. Four factors had eigenvalues above 1 and in combination
explained 58.01% of the variance, with a clear point of inflexion after the fourth component in the scree plot. PCA results and Cronbach’s alphas are reported in table 3 (Appendix B).
Out-group and in-group communication behaviour: Final scales. Based on the PCA results, it was concluded that the current study’s theoretical concepts of topic avoidance and ignoring ethnic and cultural differences capture a general tendency of avoiding ethnicity and culture as a topic in out-group and in-group interactions. The items based on the two concepts were thus combined into one subscale, supported by subsequent reliability analyses.
Normative behaviour. The first component for both out-group and in-group PC communication behaviour appears to measure normative behaviour, i.e. self-censoring of words, expressions and humour that is considered insensitive. Reliability of the final six-item scale for out-group normative behaviour is reasonable, Cronbach’s alpha = .73. Reliability of the six-item scale for in-group normative behaviour is good, Cronbach’s alpha = .84.
Topic avoidance. The second PC communication behaviour subscale is topic avoidance. The scale includes seven items from the theoretical concepts of topic avoidance and ignoring ethnic and cultural differences, and measures the tendency of avoiding ethnicity and culture as a topic, including mutual differences. The scale reliability for out-group topic avoidance is quite weak (Cronbach’s alpha = .58), but is retained for further analysis. The scale reliability for in-group topic avoidance is reasonable, Cronbach’s alpha = .68.
Disagreement avoidance. Items from the theoretical construct of disagreement avoidance loaded strongly on one component in both analyses, and the subscales measure tendency to supress criticism and disagreement with, and exhibit positive bias towards, out-group members. The reliability of the 3-item scales for out-out-group disagreement avoidance (Cronbach’s alpha = .69) and in-group disagreement avoidance (Cronbach’s alpha = .62) is reasonable.
Final PC communication behaviour subscales with items are reported in table 4 (Appendix B). Items are measured on a five-point Likert scale ranging from 1 = “strongly disagree” to 5 = “strongly agree”.
Model variables’ descriptive statistics. Table 5 (Appendix B) reports descriptive statistics for model’s variables. Normal distributions are generally satisfactory, with some exceptions. First, variables measuring quality of communication in out-group and in-group interactions, were quite positively skewed, with respondents reporting high quality of communication. Second, variables measuring out-group and in-group topic avoidance were
quite negatively skewed and maximum score was 4 (range 1-5). Despite the four variables’ problematic distributions, it was concluded that they were satisfactory for further analysis.
Results Intercorrelations
Table 6 (Appendix B) reports intercorrelations between PC communication behaviour subscales, and control variables. Cultural empathy and diversity beliefs are included to check for intercorrelations between variables in the moderation. The subscales display strong positive correlations between groups, especially between out-group and in-group normative behaviour (r = .74, p > .001), and between out-group and in-group disagreement avoidance (r = .56, p > .001). Topic avoidance and normative behaviour correlate moderately strong within out-group (r = .42, p > .001) and strong within in-group (r = .52, p > .001). Cultural empathy correlates moderately strong with positive diversity beliefs (r = .22, p = .001), but has a weaker correlation with negative diversity beliefs (r = -.16, p = .017). The moderator additionally has relatively strong correlations with several of the PC communication
subscales, most notably with in-group normative behaviour (r = .30, p > .001) and in-group topic avoidance (r = .32, p > .001). Gender correlates moderately strong with out-group normative behaviour (r = .22, p = .001) and in-group normative behaviour (r = .28, p > .001). Age correlates moderately strong with out-group disagreement avoidance (r = .22, p > .001).
Diversity Beliefs on PC Communication Behaviour - Moderated by Cultural Empathy The analyses first tested the associations between diversity beliefs and PC
communication behaviour (H1a-b), and the moderating effect of cultural empathy (H2a-b). Continuous variables were standardised and interaction effects between diversity beliefs, and cultural empathy, were constructed. Multiple regressions by forced entry method including control variables were run to test the relationships.
H1a stated that positive diversity beliefs would positively predict out-group and in-group PC communication behaviour. Table 7 reports that positive diversity beliefs has significant positive associations with out-group normative behaviour and out-group
disagreement avoidance. The association with out-group topic avoidance is not significant. As shown in table 8, positive diversity beliefs has significant positive associations with all in-group PC communication behaviour subscales: normative behaviour, topic avoidance and disagreement avoidance. H1a is supported.
Table 7: Separate regression analyses of associations between positive diversity (div.) beliefs and out-group PC communication behaviour subscales, moderated by cultural empathy. Controlled for gender, age, education and team diversity. N = 217.
Out-group PC
communication behaviour
Normative behaviour Topic avoidance Disagreement avoidance
b SE b SE b SE
Constant 3.11*** .15 2.43*** .12 2.28*** .16
Positive div. beliefs (PD) .16*** .04 .06 .04 .10* .05
Cultural empathy (CE) .10* .04 .00 .04 .00 .04
PD x CE -.05 .03 -.04 .02 -.10*** .03
R2 .18 .06 .12
F (df) 6.66*** (7, 209) 2.08* (7, 209) 3.93*** (7, 209) Note. *p < .05, ** p < .01, *** p < .001.
Table 8: Separate regression analyses of associations between positive diversity (div.) beliefs and in-group normative behaviour (n = 217), topic avoidance (n = 217) and disagreement avoidance (n = 215), moderated by cultural empathy. Controlled for gender, age, education and team diversity. In-group PC
communication behaviour
Normative behaviour Topic avoidance Disagreement avoidance
b SE b SE b SE
Constant 2.79*** .17 2.39*** .12 2.55*** .15
Positive div. beliefs (PD) .25*** .05 .15*** .03 .17*** .04
Cultural empathy (CE) .15*** .05 .12*** .03 .14*** .04
PD x CE -.02 .03 -.05* .02 -.05 .03
R2 .27 .19 .18
F (df) 11.24*** (7, 209) 6.81*** (7, 209) 6.63*** (7, 207) Note. *p < .05, ** p < .01, *** p < .001.
H1b stated that negative diversity beliefs would negatively predict out-group and in-group PC communication behaviour. As reported in table 9, negative diversity beliefs has a significant positive association with out-group disagreement avoidance, but has no significant associations with out-group normative behaviour and topic avoidance. In the case of in-group
PC communication behaviour (table 10), negative diversity beliefs has a significant negative association with group normative behaviour, and a significant positive association with in-group disagreement avoidance. Negative diversity beliefs has no significant association with in-group topic avoidance. H1b is not supported.
Table 9: Separate regression analyses of associations between negative diversity (div.) beliefs and out-group PC communication behaviour subscales, moderated by cultural empathy. Controlled for gender, age, education and team diversity. N = 217.
Out-group PC
communication behaviour
Normative behaviour Topic avoidance Disagreement avoidance
b SE b SE b SE
Constant 3.07*** .15 2.36*** .12 2.11*** .16
Negative div. beliefs (ND) -.08 .05 .05 .04 .18*** .05
Cultural empathy (CE) .12** .04 .03 .04 .08 .05
ND x CE -.02 .04 .01 .03 .06 .04
R2 .15 .05 .14
F (df) 5.21*** (7, 209) 1.75 (7, 209) 4.94*** (7, 209)
Note. *p < .05, ** p < .01, *** p < .001.
Table 10: Separate regression analyses of associations between negative diversity (div.) beliefs and in-group normative behaviour (n = 217), topic avoidance (n = 217) and disagreement avoidance (n = 215), moderated by cultural empathy. Controlled for gender, age, education and team diversity.
In-group PC
communication behaviour Normative behaviour b SE Topic avoidance b SE Disagreement avoidance b SE
Constant 2.78*** .17 2.31*** .12 2.38*** .15
Negative div. beliefs (ND) -.20*** .05 .02 .04 .19*** .04
Cultural empathy (CE) .17*** .05 .17*** .04 .22*** .04
ND x CE -.03 .04 .01 .03 .03 .04
R2 .24 .11 .20
F (df) 9.63*** (7, 209) 3.79*** (7, 209) 7.58*** (7, 207) Note. *p < .05, ** p < .01, *** p < .001.
H2a stated that the positive relationships between positive diversity beliefs and out-group and in-out-group PC communication behaviour would be stronger for high cultural empathy employees. As shown in table 7, cultural empathy does not significantly moderate the associations between positive diversity beliefs, and out-group normative behaviour and topic avoidance. It does however significantly moderate the association between positive diversity beliefs and out-group disagreement avoidance. As shown in figure 2, the association is somewhat negative for high cultural empathy employees. Meanwhile, for low cultural
empathy employees the positive association between positive diversity beliefs and out-group disagreement avoidance becomes stronger.
Fig. 2. Regression of positive diversity beliefs on out-group disagreement avoidance.
As reported in table 8, cultural empathy does not significantly moderate the associations between positive diversity beliefs, and in-group normative behaviour and disagreement avoidance. There is however a significant moderation of the association between positive diversity beliefs and in-group disagreement avoidance. As shown in figure 3, the association is positive for both high and low cultural empathy individuals. However, it appears that the association is somewhat stronger for low cultural empathy individuals. It is concluded that H2a is not supported.
H2b stated that the negative relationships between negative diversity beliefs and out-group and in-out-group PC communication behaviour would be stronger for high cultural empathy employees. However, as reported in tables 9 and 10, cultural empathy did not significantly moderate the associations between negative diversity beliefs and the out-group and in-group PC communication behaviour subscales. The results do not support H2b.
1 1,5 2 2,5 3 3,5
Low positive diversity beliefs High positive diversity beliefs
O u t-gr ou p d is agr ee me n t avoi d an ce Low Cultural empathy High Cultural empathy
Figure 3. Regression of positive diversity beliefs on in-group topic-avoidance.
PC Communication Behaviour on Content and Quality of Communication
Associations between out-group and in-group PC communication behaviour and content and quality of communication were tested with multiple regression analyses, by forced entry method including control variables. H3a stated that out-group PC communication behaviour negatively predicts out-group content and quality of communication. Out-group normative behaviour does not significantly predict any of the communication outcomes in out-group interactions (table 11). As reported in table 12, out-group topic avoidance does not
significantly predict content of communication, but has a significant negative association with out-group quality of communication. Out-group disagreement avoidance (table 13) has a significant negative association with out-group quality of communication, but does not significantly predict out-group content of communication. H3a is partially supported.
Table 11: Separate regression analyses of out-group normative behaviour predicting out-gr. content and quality of communication. Controlled for gender, age, education and team diversity. N = 217. Out-group content and
quality of communication
Task-related comm. Person-related comm. Quality of comm.
b SE b SE b SE
Constant 2.42*** .52 2.39*** .54 2.98*** .47
Out-gr. normative behaviour .08 .08 .16 .08 .04 .07
R2 .05 .04 .03 F (df) 2.09 (5, 211) 1.80 (5, 211) 1.52 (5, 211) Note. *p < .05, ** p < .01, *** p < .001. 1 1,5 2 2,5 3 3,5
Low positive diversity beliefs High positive diversity beliefs
In -gr ou p top ic avoi d an ce Low Cultural empathy High Cultural empathy
Table 12: Separate regression analyses of out-group topic avoidance predicting out-gr. content and quality of communication. Controlled for gender, age, education and team diversity. N = 217. Out-group content and
quality of communication
Task-related comm. Person-related comm. Quality of comm.
b SE b SE b SE
Constant 2.39*** .54 3.32*** .56 4.03*** .47
Out-gr. topic avoidance .10 .10 -.14 .10 -.32*** .09
R2 .05 .03 .09
F (df) 2.07 (5, 211) 1.42 (5, 211) 4.24*** (5, 211)
Note. *p < .05, ** p < .01, *** p < .001.
Table 13: Separate regression analyses of out-gr. disagreement avoidance predicting out-gr. content and quality of communication. Controlled for gender, age, education and team diversity. N = 217. Out-group content and
quality of communication
Task-related comm. Person-related comm. Quality of comm.
b SE b SE b SE
Constant 2.73*** .48 2.73*** .50 3.47*** .34
Out-gr. disagreement avoid. -.03 .08 .08 .08 -.17* .07
R2 .04 .03 .06
F (df) 1.89 (5, 211) 1.22 (5, 211) 2.73* (5, 211)
Note. *p < .05, ** p < .01, *** p < .001.
H3b stated that group PC communication behaviour would positively predict in-group content and quality of communication. In-in-group normative behaviour does not significantly predict any of the in-group communication outcomes (table 14). However, although not significant on the < .05 level, results suggest that a positive relationship may exist with out-group person-related communication (p = .055). As reported in table 15, in-group topic avoidance has a significant positive association with in-in-group task-related communication, but does not significantly predict person-related communication and quality of communication. In-group disagreement avoidance (table 16) has significant positive associations with in-group task-related and person-related communication. Although not significant at the < .05 level, it is worth noting that the results suggest that a negative relationship may exist between in-group disagreement avoidance and in-group quality of communication (p = .052). H3b is partially supported.
Table 14: Separate regression analyses of in-group normative behaviour predicting in-group content and quality of communication. Controlled for gender, age, education and team diversity. N = 217. Out-group content and
quality of communication
Task-related comm. Person-related comm. Quality of comm.
b SE b SE b SE
Constant 2.55*** .49 3.18*** .47 3.44*** .35
In-gr. normative behaviour .00 .07 .10 .07 -.01 .05
R2 .04 .09 .08
F (df) 1.89 (5, 211) 4.08*** (5, 211) 3.55** (5, 211)
Note. *p < .05, ** p < .01, *** p < .001.
Table 15: Separate regression analyses of in-group topic avoidance predicting in-group content and quality of communication. Controlled for gender, age, education and team diversity. N = 217. Out-group content and
quality of communication
Task-related comm. Person-related comm. Quality of comm.
b SE b SE b SE
Constant 1.89*** .51 3.13*** .49 3.69*** .38
In-gr. topic avoidance .30** .10 .13 .09 -.12 .07
R2 .08 .09 .09
F (df) 3.81** (5, 211) 3.95** (5, 211) 4.20*** (5, 211) Note. *p < .05, ** p < .01, *** p < .001.
Table 16: Separate regression analyses of in-gr. disagreement avoidance predicting in-gr. content and quality of communication. Controlled for gender, age, education and team diversity. N = 215. Out-group content and
quality of communication Task-related comm. Person-related comm. b SE b SE Quality of comm. b SE
Constant 2.13*** .49 2.97*** .47 3.67*** .36
In-gr. disagreement avoid. .23** .08 .18* .07 -.11 .06
R2 .08 .09 .09
F (df) 3.62** (5, 209) 4.19*** (5, 209) 4.21** (5, 209) Note. *p < .05, ** p < .01, *** p < .001.
Figure 4 reports the resulting significant relationships between diversity beliefs and PC communication behaviour subscales, and between PC communication behaviour subscales and out-group and in-group content and quality of communication. Cultural empathy significantly moderates the relationships between positive diversity beliefs and out-group disagreement avoidance, and between positive diversity beliefs and in-group topic avoidance.
Fig. 4. Significant relationships for the role of out-group and in-group PC communication behaviour characteristics. Controlled for gender, age, education and team diversity.
Conclusion and Discussion
The main aim of the present study was to explore political correctness on a
communication behavioural level, and present a self-report instrument that can measure this communication tendency in ethnically and culturally diverse work environments. Three PC communication behavioural characteristics were identified. Normative behaviour was found to be most salient in both out-group and in-group interactions. Topic avoidance, including the concepts of topic avoidance and ignoring ethnic and cultural differences, and disagreement avoidance are less salient, but appear to have the potential to emerge in both out-group and in-group interactions. The three communication behaviour characteristics, and related subscales, confirm the theoretical underpinning, which provides new insights into current perspectives
on workplace political correctness (e.g. Chatman et al., 2012; Ely et al., 2006) and individual-level communication behaviour in diverse work teams (e.g. Hentschel et al., 2013).
Diversity Beliefs and PC Communication Behaviour
The role of PC communication behaviour was investigated by examining how employees’ diversity beliefs were related to normative behaviour, topic avoidance and disagreement avoidance. Pro-diversity employees were expected to report increasing PC communication behaviour, while the opposite was expected for employees with negative diversity beliefs. However, findings reveal that people’s motivation to adopt PC
communication behaviour is not as straightforward as hypothesized.
As expected, pro-diversity employees increasingly adopt normative behaviour in interactions with both out-group and in-group members. Employees with negative diversity beliefs reported decreasing normative behaviour in in-group interactions. These findings extend the link between diversity beliefs and cultural sensitivity to a communication
behavioural level, as previous studies have mainly provided evidence of similar tendencies on a societal level (e.g. Kauff & Wagner, 2012). The findings further support Hentschel et al. (2013), who found that positive diversity beliefs can buffer the negative consequences of social categorization in diverse teams, such as communication behavioural hostility towards out-group members. The results did not support Yoder (1991), who argued perceived
diversity threats to reduce cultural sensitivity in intergroup workplace interactions. This may however be due to the inclusion of cultural empathy, as the ability to empathise with
culturally different others seem to reduce the influence of negative diversity beliefs. As hypothesized, pro-diversity employees increasingly adopt in-group topic
avoidance. However, there is a gap between high cultural empathy and low cultural empathy employees, with the latter reporting lower but increasing levels. Contrary to expectations, stronger positive diversity beliefs reduce the gap, as the positive association is stronger for
low cultural empathy employees. This finding suggests that even though pro-diversity employees appreciate the instrumentality of diversity, low cultural empathy individuals may struggle with, and thus suppress, the topic of ethnicity and culture due to a decreasing ability to adjust in diverse environments (Yakunina et al. 2012). Diversity beliefs were not found to influence out-group topic avoidance, and negative diversity beliefs were not found to
influence in-group topic avoidance. The concept of diversity climate may elucidate the current findings regarding topic avoidance (e.g. Hofhuis, et al., 2012). Addressing ethnicity, culture and associated differences in in-group interactions does not contribute to openness, but confines the topic to interactions between demographically similar colleagues. Although diversity beliefs did not influence out-group topic avoidance, in-group topic avoidance may indicate that pro-diversity employees deem diversity issues to be more appropriately
addressed in forums including all employees.
Finally, as hypothesized, disagreement avoidance is increasingly adopted by pro-diversity employees in both out-group and in-group interactions. Contrary to expectations, the same is true for employees who hold negative diversity beliefs. Previous studies have found that both positive and negative ethnic attitudes can motivate individuals to inhibit or promote certain behaviours so to avoid coming across as hostile towards ethnically different others (Olson & Fazio, 2010). The present results confirm that a similar tendency is to be found in relation to diversity beliefs. It is especially interesting that both pro-diversity employees and employees with negative diversity beliefs increasingly adopt out-group disagreement
avoidance. In the case of pro-diversity employees and contrary to expectations, the tendency of disagreement avoidance is increasingly salient for low cultural empathy individuals, while high cultural individuals display decreasing disagreement avoidance. The fact that pro-diversity employees with low cultural empathy, and employees with negative pro-diversity beliefs, increasingly adopt out-group disagreement avoidance support previous research by
Dunton and Fazio (1997), and Harber (1998), who found that disagreement avoidance characteristics in interethnic interactions is motivated by uncertainty and fear of negative outcomes in interethnic interactions.
PC Communication Behaviour and Content and Quality of Communication The consequences of PC communication behaviour in out-group and in-group interactions, was investigated in relation to employees’ perceptions of task-related and person-related content of communication, and quality of communication. Out-group PC communication behaviour was expected to negatively influence perceptions of out-group communication, while it was predicted that PC communication behaviour in in-group interactions would benefit employees’ perceptions of in-group communication.
PC Communication behaviour and out-group content and quality of
communication. Out-group normative behaviour was not found to influence an employees’ perception of content and quality of communication in interactions with out-group members. This indicates that while making norms of cultural sensitivity salient may be beneficial to heterogeneous teams’ communication processes (Gonzalo et al., 2014), individual-level culturally sensitive communication behaviour is not. The findings support Ely and Thomas (2001), who argued that promoting cultural sensitivity between employees is not an
instrumental link between diversity and work processes. However, results reflect the perceptions of majority employees, as respondents were predominantly of Norwegian ethnicity. It is not unlikely that a larger sample with minority members could offer more insight into whether and how normative behaviour perceptions of communication between ethnically and culturally diverse team members.
The current study provides additional evidence of the negative consequences of suppressing the topic of ethnicity, culture and associated individual differences (Hofhuis et
al., 2012; Norton et al., 2006), as employees who adopt topic avoidance report decreasing quality of communication in out-group interactions. These findings thus support the benefits of a strong diversity climate, as topic avoidance reportedly stifles ability to communicate efficiently. A relationship between topic avoidance and task-related communication in out-group interactions could not be confirmed. However, communication about tasks will likely indirectly suffer as topic avoidance was found to reduce quality of communication, which impedes team members’ ability develop a shared understanding of task-related content (Dennis & Kinney, 1998). Although one could assume that suppression of such immutable interpersonal differences, as ethnicity and culture, would stifle person-related interactions, topic avoidance was not found to influence person-related communication. However, Hofhuis et al. (2012) found that minorities perceive more diversity-related conflicts in weak diversity climates, and Norton et al. (2006) found that minority members find majority members less friendly when the latter refuses to acknowledge ethnicity. It is thus not unlikely that a larger sample of minority member could produce different results for the relationship between topic avoidance and the exchange of both person-related and task-related communication content in intergroup interactions.
As hypothesized, disagreement avoidance was found to reduce employees’ perceptions of quality of communication in interactions with out-group members, thus providing
empirical evidence to Chatman et al.’s (2012) argument about the inhibiting potential of overly sensitive, or overly deliberate actions. Previous studies have suggested that a person’s disagreement avoidance can cause the receiver to experience increasing uncertainty,
consequently impeding his ability to effectively participate in interethnic interactions (Major et al., 2013). The current study confirms that also the person who adopts such exaggerated communication behaviour perceives communication with ethnically and culturally different colleagues as decreasingly efficient.
