Trust2

Int. J. of Human Resource Management 14:1 February 2003 117-127

Trust me, I'm your boss: trust and power in supervisor-supervisee communication Michael WUemyns, Cynthia Gallois and Victor J. CaUan Abstract This study examined employees' perceptions of trust, power and mentoring in manager-employee relationships in a variety of sectors, including health care, education, hospitality and retail. The main theoretical frameworks used were communication accommoidation theory and social identity theoiy, in examining the manager-employee relationships from an in-group/out-group perspective. Computer-aided content analyses revealed a number of emergent communication and relationship themes that impact upon the level of 'in-groupness' and therefore trust in supervisor-supervisee relationships. While it may be illusory to believe that any organization can enjoy complete tnist among its workforce, it is clear that ceitain conmiunication characteristics can result in greater trust in manager-employee relationships, even within the context of organizational constraints. It is argued that the results of the study could be used to inform human resource management academics of key aspects of managerial communication that should be further researehed, and also provide insights into the main communication skills that managers should focus upon to improve trust in the workplace. Kejrwords

Trust; communication; workplace; power.

Introduction Trust is an elusive concept, especially in the current organizational climate of job insecurity, reduced resources and the decline of unions (McCune, 1998). In terms of supervisor-supervisee communication, McCune argues, it is difficult to trust someone who has a distinct advantage over you. Recent studies have found that 43 per cent of employees believe their managers cheat and lie to them, and 68 per cent of employees do not trust their managers (see Davis and Landa, 1999). Such findings cleariy have important implications for oiganizations and the individuals who woik for them. For example, Davis et al. (2000) found that trust was significantly related to sales, profit and turnover, and to employees' perceptions of their managers' integrity and competence. While it is acknowledged that factors such as organizational culture, structure and availability of resources are important determinants of trust in the workplace, this paper focuses on the more micro, but important issue of how supervisors and supervisees relate to each other, and the implications for trust and mistmst. This paper reports the results of a thematic content analysis of employees' perceptions of their supervisors' or managers' communication style, where issues relating to trust and power emerged as Michael Willemyns, College of Business Sciences, Zayed University, PO Box 19282, Dubai, UAE (e-mail: [email protected]): Cynthia Gallois, School of Psychology, Faculty of Social and Behavioural Sciences, The University of Queensland, Brisbane, Australia (e-mail): [email protected]); Victor J. Callan, UQ Business School, University of Queensland, Australia 4072 (e-mail: [email protected]). The Inlematiomil Journal o / Human Rexource Managemenl ISSN 0958-5192 prim/ISSN 14«64399 online 6 2003 Taylor & Francis Ltd http:/Avww.landf.oo.uk/jounials

118

The International Jounuil oj Hunttiit Re.sourvc Maiuifieineni

highly salient themes in the study. The paper takes an in-group versus out-jiroup approach in examining power and trust in supervisory communication. Thi.s is an important perspective, as much research has found that hoth upward and downward communication in organizations is significantly distorted or withheld as a function oi' interactants' group identity, power dynamics and level ot° trust or mistrust towards each other (Morand. 1996). Indeed, Morand argue.s that the more an employt-e perceives u manager in 'in-group' terms, the more the manager is trusted. The main theoretical frameworks used are communication accommodation theory (CAT) and social identity theory (SIT). First, these will be briefly outlined and discussed in terms of their relevance to trust in supervi.sor-supervisee relationships. This will be followed by an overview of the results of the study's findings in relation to trust and power. The paper concludes with a critical discussion of the implications of supervisor-supervisee communication in terms ofthe so-called 'quick fix' approach to trust in supervisor-supervisee relationships, as well as implications for trust beyond the quick fix. Communication accommodation theory (CAT) This overview of CAT is selective, with a focus on issues of power and trust in supervisor-supervisee communication. For a more detailed discussion of CAT concepts and empirical findings, see Jones et al. (1999). CAT is a robust iiamework for examining interpersonal and in-group/out-group communication, as well as interactants' motivations, goals and communication strategies. Such motivations and goals include seeking the approval of the other person or signalling in-group or out-group membership (affiliation or social distance and power). CAT proposes that interactanLs draw upon a wide range of communication strategies, including approximation, interpersonal control, discourse management and relational strategies, to achieve such goals. While labelled 'strategies' in CAT. interactanLs are not always conscious of using such communication strategies (Jones et al.. 1999; Willemyns et al.. 2(X)0). The strategies will now be briefly outlined. Approximation strategies refer to interdctants adjusting their communication style to sound more like or less like the other person (termed convergence and divergence, respectively). Interactants have been found to modify communication characteristics such as vocabulary, jargon, accent and non-verbal behaviours. CAT proposes that individuals converge in order to signal affinity with, or seek the approval of, the other person. For example, in a study of communication in employment interviews, Willemyns et al. (1997) found that interviewees converged in speech style towards that of their interviewers. Conversely, individuals have been found to diverge in order to signal interpersonal or social distance or disapproval. CAT draws upon similarityattraction theoiy (Byrne, 1971), which contends that individuals are likely to be attracted to people who are similar to themselves, in terms of personal characteristics or group memberships. In relation to the present study, individuals are more likely to trust in-group members than out-group members (Morand, 1996). Interpersonal control strategies refer to the speaker's communication strategy of positioning him- or herself, and/or the other person, in a particular role or power position (Jones et al.. 1999). For instance, in interactions with an employee, a manager may explicitly and implicitiy communicate their superior status in the relationship. Conversely, supervisors may reduce perceived power differences by referring to their supervisees as their 'fellow team members', or by referring to themselves in terms of a nurturin^i, mentoring role.

Willemyns et al.: Trust me, I'm your boss

119

Discourse management strategies are manifested in a more discouise-oriented, but equally powerful fonn. For instance, a long tradition of power and communication research has shown that higher-status individuals are more likely than their lower-status counterparts to display behaviours such as intemipting, dominating the conversation, controlling the choice of topic and the use of directives, and are less likely to use an informal tone or self-disclosure. Conversely, powerless communication is characlerized by a higher incidence of indirectness, disqualifications, hedges, hesitations and tag questions (see Jones et al., 1999). Relational strategies are a recent addition to CAT, with a focus on communication behaviours that indicate support, empathy, inclusion, valuing the other person and face issues. For example, in relation to face issues, positive face is manifest in behaviours that allow the recipient to save face, such as a manager not reptinunding an employee for making a mistake, or at least providing negative feedback privately and in a tactful manner. Conversely, face threat or attack is manifest in speakers challenging or embarrassing the listener (Morand, 1996). Social identity theory Social identity was defined by Tajfel (1974) as an individual's knowledge that he or she belongs to certain social or status groups, together with some emotional and value significance of the group membership. The more a person identifies with his or her ingroup (e.g. manager), the more he or she will feel distinct fiom out-group members (e.g. employees). When one's social identity is salient, so too are out-group dynamics. Drawing upon social identity theory, CAT proposes that an interactant's communication style contains social markers that convey infonnation about the speaker, such as personality, social identity, status and power (Callan et al., 1983). Gallois and Giles (1998) contend that, in some interactions, the relationship is mostiy interpersonal (e.g. friendships) and the interactants perceive each other as individuals, while, in other situations, people interact primarily in terms of group-based identities or stereotypes (e.g. formal supervisee-supervisor relationships). Hogg and Abrams (1988) argue that communication is more often a function of the relative status or power of the interactants than of their personality. Thus, most communication in supervisorsupervisee relationships is seen as a function of the interactants' status or role. [t is argued here that a pattern of positive interactions with an out-group member may lead to 'breaking through' the inter-group barrier and, thereby, to an increase in tmst. In terms of the present study, this would mean a lower likelihood of the supervisor being perceived and related to as a member of a status out-group by their supervisee, and vice versa. Again, this has implications for trust in the woikplace, as we are more likely to identify with and trust in-group members than out-group members (Morand, 19%). Communication, and trust and mentoring Not all supervisors are mentors; however. Bell (1996) argues that, when supervisors take on a mentoring role (formally or infonnally), the individuals as well as the organization benefit. House (1981) identified four dimensions of mentoring: emotional support (e.g. trust, concern, listening and esteem); appraisal support (e.g. affirmation, feedback); informational support (e.g. advice, suggestions, directives and information); and instrumental support (e.g. money, labour and time). These factors are clearly related to issues of trust in supervisor-supervisee relationships.

120

Till' International Jounuil
In a major study of management style in large Australian corporations. Evans (1 found that a number of features are related to employees' perceptions of their managers' trustworthiness. These features included the employee's perceptions of thf supervisor's appreciation of the employee's worth, as manifest in praise, compliments, etc. A trusted supervisor was also seen as one who cuuld he relied on to care for the employee's personal and professional welfare: thai is. as one who takes a mentoring approach. Openness in communication was seen as highly important, as gatekeeping of information, or keeping employees in ignorance, creates uncertainty, tear and distrust. Essentially, supervisors were most likely to be trusted if they were seen to take a caring, mentoring approach with their supervisees, while still being regarded as competent and deserving of their authority. By contrast, managers perceived as untrustworthy were seen as self-serving, failing to give recognition, stifiing the employee's potential, quick to blame and criticize, and perceived as incompetent. It can be concluded that a trusted supervisor not only manages the task responsibilities of his or her role, but manages the relationship and power differences positively at the same time. Sodal identity and mentoring McManus and Russell argued that 'obtaining in-group status may be a pre-requisite for employees to receive mentoring from their supervisors' (1997: 147). They found that in-group members received more attention and support from their supervisor than out-group members, while out-group members experienced a more formal relationship with their supervisor. In mentoring-style supervisorsupervisee relationships, in-group and out-group dynamics are salient due to the paradoxical combination of the affiliative nature of the mentoring relationship and the inherent power differences in it. Thus, an understanding of how communication influences in-group/out-group perceptions (and vice versa) is important to understanding trust in mentoring relationships. General hypotheses. Based on CAT and other supervisor-supervisee communication research, it was expected that issues of trust and power would emerge as significant categories in the content analyses. More specifically, it was expected that, in positively perceived conversations, managers' communication would be perceived in terms of a mentoring, caring, in-group and trusting relationship. For example, it was expected that managers' communication would be characterized in terms of minimal power distance ('ingroupness'), personal similarities, self-disclosure, active listening, suppon, empathy, inclusive communication and positive face. Conversely, it was expected that salient themes in negatively perceived interactions would include managers' communication of dominance, distancing, lack of willingness to listen, lack of support or empathy and face threat. Method Participants The panicipants were 1S7 working psychology students (ninety females and sixty-seven males, ranging in age firom 17 to 58; mean age 22.12 years) who had been in full-time employment for at least six months, or part-time for at least twelve months. Their occupations and places of work covered a wide spectnim, including the service sector

Willemyns et al: Trust me, I'm your boss 121 (retail outlets,restaunmts),education (e.g. teachers) and the health sector (nurses, health practitioners or assistants). Procedure

Questioniiains Participants received two questionnaires and were asked to return them anonymously by mail or via a survey-retum box within a week. The questionnaires asked participants to describe a satisfactory and an unsatisfactory interaction with a supervisor or manager. Participants wrote up to one page (responses ranged from ISO to 2S0 words) describing the conversation in as much detail as they could recall, including specific statements made by themselves and their manager. The questionnaire also obtained brief responses (one or two sentences) to open-ended probe questions (e.g. 'How important was the manager's personality [or status] to the way he/ she communicated? Please provide an example'). The questionnaires also obtained structured quantitative data about the participants, their managers and the actual interactions, using six-point Likert-type scales (e.g. 'How distancing is your manager?'. 'How considerate was your manager during the interaction?').' Development of the coding scheme The development of the coding scheme was conducted using NUD*IST4 qualitative research software (NIJD*IST4, 1998). Transcripts of employees' descriptions of the interactions were content-coded using a combination of a grounded theory approach (e.g. Strauss, 1987), where salient concepts emerge from the data over several readings and iteratively refined re-categorizations and a substantive theory-based approach, where statements relevant to CAT strategies were coded. Thus the coding was both data- and theory-driven. The text units were coded at the micro-level of phrases or simple sentences. Results Six thousand and fifty-three text units (e.g. phrases) were coded into thirty-five lowerorder coding categories (e.g. 'non-dominating', 'self-disclosure', 'supportive'), within six second-order categories (e.g. 'interpersonal control', 'discourse nuuiagement' and 'relational' strategies) and two higher-order categories, 'accommodating' (in-group communication) and 'counteraccommodating' (out-group communication). Table 1 shows the categories, their operational definition, typical text units and the frequencies of text units pertaining to trust and power. Discussion Overview The results indicate a clear pattem of communication behaviours and characteristics salient for employees in terms of in-group/out-group relations with their managers, and implications of these communication behaviours for trust. The content analysis yielded a number of categories that were conceptually related to CAT's well-established strategies, interpersonal control and discourse management. This was expected, as the manager-employee communication context is very much an inter-group one, and both strategies relate strongly to in-group/out-group and power dynamics. The results were also encouraging in that they provided empirical support for the recently theorized concept of 'relational communication' as a CAT strategy, with implications for perceptions of in-group/out-group membership and, therefore, trust.

122

The IntemutUmal Journut of Hunum Resoune Mtiiitigenifiii

1 Content coding categories pertaining to trust und power, openuioi.al definitions, typical text units and the frequencies of te.\t unit.s Iin ptirentheses) ACCOMMODATIVE (in-gmiip) code.i

COVNTERACCOMMODATIVH (out-grtrnp) (odes

INTERPERSONAL CONTROL Non-dotninating/equistatus (159) References to the manager communicating with the employee as an equal; manager perceived as non-dominating, e.g. 'She didn't make you feel like manageremployee - she got in and worked with you'.

Dominating (291) Manager referred to as behaving in a dominating way. e.g. '. . . left me in a nowin situation, as she is in the higher position telling me what to do'.

Non-manager role refs (75) References to manager in a nonmanagerial role. e.g. 'He said being a family man himself he could understand my problem'.

Coercion {138) Communication of ability to sack employee, reduce or increase hours or change conditions, etc., e.g. 'He told me to hurry up and get my act together or else I wouldn't have a job'.

FriendsMp role references {51) References to manager as a friend, e.g. 'She is very approachable and treats me as a friend, not an inferior'. Similarities {interpersonal similarities, similar values) {101). e.g. ' When she told me she was in the same union as I was. I saw her in a different way'. DISCOURSE MANAGEMENT Willing to discuss/listen {104) References to the manager being willing to discuss an issue or to listen to employee's concerns - 'He listened intently and pointed out . . .'.

Small talk {226) References to the manager speaking about non-work topics, chatting, pleasant conversation - 'We gave examples of what sports we had played, or friends had played'. Self-disclosure {112) Where the manager discloses relatively personal information about themselves, or their feelings about issues or other people - 'We saw a side of him that we didn't realize even existed - he iqmlogized to us for not giving more positive feedback, and explained he'd been brought up in a

Unwilling to discuss/Usten {118) References to the manager not being willing to discuss an issue that the employee wanted discussed or not willing to listen to employee or hear employee's side of the story - 'He would not let me explain. He would walk off while I was talking. He cut me off. He would not give me an answer'.

Willemyns et at.: Trust me, I'm your boss

123

Tbiiie 1 Continued ACCOMMODATIVE {in-group) codes

COUNTERACCOMMODATIVE {out-group) codes

household where compliments weren't given much'. RELATIONAL COMMUNICATION Supportive (e.g. helpful, giving advice) (392) Where the manager is referred to as helpful, giving advice or teaching, or supportive (either emotionally supportive including 'caring', or instrumentally supportive regarding work issues/ resources) - 'Ofliered opinions/advice without being overbearing'.

Lack of stipport/cmnpromise (210) References to the manager being unsupportive, uncaring or unwilling to compromise - 'To point out to me that what I was wanting was not really appreciated, and that I was making things difficult for her'.

Empathetic {147) Includes references to 'understanding' the employee's situation - 'She acknowledged my feelings of frustration'.

Unempathetic (79) References to the manager lacking empathy, sympathy or understanding - 'He didn't understand that I was finding it hard - I was trying my best'.

Trustworthy {60) References to manager being trustworthy, fair, just, honest, etc. - 'I douht I would have raised the topic with any other superior - I trusted her'.

Untrustworthy (188) References to the manager heing untrustworthy or unfair, unjust, biased, etc. - 'We went to her with a problem and we were the ones got into trouble - after that we didn't trust her with our complaints'.

Praise/valtied {271) Praise, encouragement, thanks, other explicit statements of valuing employee 'He said "Great work Jenny you have done a fantastic job - 1 couldn't have done a better job myself".'

Criticism (466) Any inference of or reference to being blamed, accused, or criticized unjustly 'He deliberately attempted to publicly embarrass me (in front of my co-workers) so that by the time he had finished I felt two inches tall'.

Positive face (141) Where the manager communicates in such a way as to help employee save face; for example, not becoming angry when the employee makes a mistake, speaking to the employee privately when discussing sensitive issues or giving feedback - '. . . and .she didn't get really angry even though I gave her such short notice'.

Face threat (e.g. embarrassment, challenges) (313) References indicating the manager has put the employee in a position of having to defend him or herself, or that the employee felt 'imposed upon' or embanassed - 'He wanted to find out the truth, to see if I would cover up for the other co-worker'.

Inclusion {164) Providing the employee widi opportunity to have input into workplace issues (planning, decisions, suggestions, etc.). or providing infonnation to the employee that the employee would not otherwise have obtained - 'She was able to give me extra infonnation she had been told and I hadn't'.

124

The tntcniiitioiiiil Jnuriuil of Hunum Re\oun-f

Intcrpcrsoiitil cmitrnl The out-gn)up interpersonal control themes were highly salient in employees' descriptions of negative interaction.s with their managers. For example, (he eatejiorx 'dominance' was one of the largest categories that emerged from the analyses. Combined with 'cuercive power', these categories highlight the negatively perceived, power-marked inter-group dynamics in many manager-employee interact ion.s. Clearly, a manager's u.se of a domineering or coercive communication style is antithetical to manager-employee tru.st. By contrast, the in-group 'interpersonal i''.)nir(>r cixles reflected communication behaviours thai would reduce perceptions of power differences, emphasize interpersonal similarities and position the manager more us an individual, rather than simply as a member of a higher-status out-group. Again, individualizing a manager breaks down employees' stereotypes of their manager, decreases perceptions of power and maximizes perceptions of trust. Di.scoiirse immurement At the discourse level, the out-group categories were indicative of managers' lack of willingness to listen or communicate, the use of directives and negatively perceived control of conversation patterns. These discourse behaviours were clearly indicative of power and role distance, which directly and indirectly reduce employees' trust in their managers. By contrast, in-group categories were indicative of two-way communication, openness and plea.sant interactions. Again, managers were described more in terms of individual characteristics and personality rather than as stereotypical members of a higher-status out-group. Active listening is a communication skill that has long been known to indicate that the .speaker is t^en seriously and that the listener cares. Selfdisclosure is a powerful form of communication in terms of breaking through the outgroup barrier and personalizing oneself. Small talk, while not us revealing us self-disclosure, can also facilitate in-group perceptions (e.g. fans of the same football team, type of movie, etc.). Over time, such positive discourse management would lead to an increase in perceptions of in-group membership and trust. Relational communication Communication behaviours coded within the 'relational communication' category accounted for 41 per cent of the coded text units. While relational communication is a relatively new and untested concept in CAT, recent theorizing of this concept has emphasized the face and emotional needs of the interactants (Gallois and Giles. 1998). The accommodative or in-group relational codes that emerged in this study were strongly related to issues such as positive face (e.g. the nunager not being critical or angry when the employee made a mistake), being supportive, empathetic. perceived as trustworthy and conveying that the employee was valued (e.g. through praise and inclusion in decision-making proces.ses). As noted earlier, the.se ure important communication behaviours that lead to positive relationships and increased tmst between managers and employees (Evans. 1996). By contrast, lack of support, empathy, trust or recognition was a salient aspect of out-group relational communication. Face threat was also a salient issue in the negative interactions. Face threat is defined by Morand (1996) as communication that is perceived as diminishing the value or worth of the recipient, and includes issues of criticism, blame and embarrassment. Face threat was evident in the present study in references to employees being embarrassed, imposed uptin or criticized. The data suggest that handled poorly, negative feedback (especially

Willemyns et al: Trust me, I'm your boss 125 in public) is not soon forgotten by employees, and can be a major source of face threat and distrast. Applied impUcations The results of this study indicate the importance of managers' awareness and use of ingroup communication behaviours for building and maintaining a bond of trust with their employees. A manager may maintain appropriate role, authority and status without necessarily resorting to negative power strategies, such as domineering or coercive communication. The importance of providing appropriate feedback to employees while allowing them to maintain face is also crucial to positive employee perceptions, and therefore employee trust. Finally, the results indicate the importance of employees being able to relate to their manager not only as a member of a higher-status out-group, but also as an individual and a fellow human being. The quick fix? It would be easy - as a number of papers examining trust in the workplace do - to conclude with a list of prescriptive behaviours that managers should employ to build and maintain trust. Indeed, the coding scheme developed in the content analysis could be read as such a list, with almost Machiavellian overtones (e.g. use self-disclosure to gain trust, emphasize similarities, use praise regularly, ensure you spend time listening to your employees, do not behave in too domineering a manner, etc). An insincere manager who attempts to use these 'strategies' is likely to be unsuccessful at carrying out the charade for long, which would result only in greater mistrust of the manager in the long term. Managers who genuinely value their employees, however, will tend to use an ingroup and trust-maintaining communication style naturally. All of the in-group communication categories that emerged in this study can be distilled into the core theme of communicating that the employees are valued as members of the managers' workforce and as fellow human beings. Thus, while status and power differences may exist structurally in the workplace, perceptions of power differentials can be minimized, and trust maximized, by managers relating to their employees as fellow human beings worthy of respect. Beyond the quick fix The findings of this study support Morand's (1996) argument that communicating with employees at an in-group level appears to be a prerequisite for trust in manageremployee relationships. Further, as noted earlier, trust is strongly linked to workplace morale and productivity. Why, then, do many managers find it difficult to be pereeived as trustworthy by their employees? While this is a complex issue, two main factors will be briefly addressed here: interpersonal and organizational factors. Interpersonal factors Many managers who are sincere and trustworthy may not have the persona or communication style to convey these traits. More importantly perhaps, many managers are not aware of the amplification of the power gap oflen inherent in being a 'subordinate' or a comparatively powerless out-group member. Much research shows that managers believe they communicate well and communicate often with their employees, while their employees rate the same managers significantly lower in terms of communication quality and quantity (Bell. 1996). An analogous perceptual gap

12b

The Intemational Jounuil of Humun Resoun t Miuiu;^enicni

iiccurs in terms of power. That is. many supervisors ure unaware that perceptions ot the p*)wer gap iirc often more salient to their supervisees than to themselves. For example, in u studv ot postgraduate student-supervisor communication. Willemyns ,-i ul. i I9<)6i found that many superx'isors of postgraduate students confidentially rated themselves in positive, mentoring terms, but their postgraduate students rated them in significantly less positive and even negative terms. This 'power-perception gap' cKists t«i some degree in any supervisor-supervisee relationship, but is particularly amplified in negative relationships (Gallois and Giles. 1998). In sum, managers need to be aware of their employees' amplified perception of a power gap. and the implications of this gap for trust. Thus, due to the inherent power differential in supervisor-supervisee relationships, supervisors or managers may need to work harder than their employees' co-workers in building and maintaining trust. Organizational factors The second major factor affecting trust in the workplace relates to organizational factors, such as an Ingrained culture of mistrust, scarcity of resources, job security, political alliances and rivalries, etc. Such factors may force even the most trustworthy of managers to behave in ways that undermine employees' trust in them. While discussion of organizational issues is beyond the scope of this paper, they are acknowledged as important determinants of trust in the workplace. Implications The findings of this study clearly have applied implications for HRM practitioners, as they highlight the importance of developing managers' communication skills to maximize employees' perceptions of being 'in-group' and being valued, and thereby maximizing trust in the workplace. As noted earlier, this is beneficial both to the individuals in the organization, as well as to the organization as a whole, in terms of productivity. The findings also have implications for researchers examining trust in the workplace, as there is a dearth of researeh empirically examining the links between communication and trust in the workplace. The present study examined trust and communication, with working students as participants. Further research needs to be conducted examining trust and communication in broader, more generalizable samples. Conclnsion The content analysis highlighted a pattern of interpersonal control, discourse management and relational conununication characteristics of managers who were described in positive, in-groiq) and, therefore, trustworthy terms by their employees. Conversely, it highlighted a pattern of communication behaviours of managers who were described in negative, out-group and untrustworthy terms. While improved communication skills may increase trast in manager-employee relationships, numagers also need to be aware of the magnification of the 'power-perception gap' by their employees, and the implications of this gap for trust and mistrust. Finally, while it may be illusory ui believe that any organization can enjoy complete trust between managers and employees, it is clear that certain communication characteristics can result in greater trust, even, to some degree, within the constraints of organizational factors. Note I Results of statistical analyses involving the quantitative data are available firom the authors upon request.

Willemyns et al: Trust me, I'm your boss 127 References Bell, G.R. (1996) Managers as Mentors: Building Pannerships for Leaming. San Francisco: Bemett-Koehler. Byrne. D. (1971) T/ir Attraction Paradigm. New York: Academic Press. Callan. V.J.. Gallois. C , and Forbes. K. (1983) 'Evaluative Reactions to Accented English: Ethnicity. Sex Role and Context'. Journal of Cross-cultural P.iycht>logy, 14: 407-26. Davis. T. and Landa. M. (1999) 'The Trust Deficit'. Management Accounting, 71(10): 12-16. Davis. J.. Schoorman. D.. Mayer, R. and Tan. H.H. (2000) 'The Trusted General Manager and Business Unit Performance: Empirical Evidence of a Competitive Advantage', Strategic Management Journal, 2I(S): 563-76. Evans. J. (1996) 'Leaders in Australia: The Australian Cultural Imprint for Leadership'. unpublished paper. Gallois, C. and Giles. H. (1998) 'Accommodating Mutual Influence in Intergroup Encounters'. In Bennett. C.A. and Palmer. M.T. (eds) Progress in Communicatitui Scieru-es, Vol. 14. Stanford, CA: Ablex, pp. 135-62. Hogg. M.A. and Abrams, D.A. (1988) Social Identifications: A Social Psychology of Intergroup Rehitiims and Croup Processes. London: Routledge. House. J.S. (1981) Vi/ork Stress and Social Support. Reading, MA: Addison-Wesley. Jones, E.. Gallois, C . Callan, V.C. and Barker. M. (1999) 'Strategies of Accommodation: Development of a Coding System for Conversational Interaction', Journal of language and Social P.sychology. 81(2): 123-52. McCune. J. (1998) 'That Elusive Thing Called Trust', Management Review, 87(7): 10-14. McManus. S.E. and Russell. J.E.A. (1997) 'New Directions for Mentoring Research: An Examination of Related Constructs', Journal af Vocational Behavior, 51(1): 145-61. Morand, D. (1996) 'Dominance. Deference, and Egalitarianism in Organizational Interaction: A Socio-linguistic Analysis of Power and Politeness'. Organization Science, 7(5): 544-56. MID*IST4 (Non-numerical Unstructured Data Indexing Searching and Theorizing) (1998) Qualitative Data Analv.sis Program. Version 4. Melboume: Qualitative Solutions Research Pty. Strauss. A.L. (1987) Qualitative Aiwlysisfor StKial Scientists. New York: Cambridge University Pre.ss. Tajfel. H. (1974) 'Social Identity and Intergroup Behaviour'. Social Science litformation, 13: 65-«3. Willemyn.s. M., Gallois, C. and Callan, V.J. (19%) 'Communicating Social Identity: An Analysis of Conversations between Postgraduate Students and their Supervisors', paper presented at the 25th Annual Meeting of Australasian Social I>sychologisis, Canberra. 2-5 May. Willemyns, M.. Gallois. C Callan, V.J. and Pittam. J. (1997) 'Accent Accommodation in the Employment Interview', Jountal of Language and Social Psychology, 15(1): 3-22. Willemyns. M., Gallois. C. and Callan, V.J. (2000) 'Accommodating Power in SupervisorSupervisee Communication: A Content Analysis', Australian Jountal of Communication, 27( I): 129-42.