نقش جنسیت در مشاوره : مفاهیمی برای روابط مشاوره متنوع و یکنواخت

The present study examined the effects of gender composition of mentoring relationships on protégés' perceptions of the degree of role modeling and psychosocial and career development mentoring functions received. Data from 200 mentor/protégé dyads composed of working professionals from a variety of industries were analyzed using ANCOVA with planned comparisons. Results indicated that mentoring relationships involving female mentors in either homogeneous or diversified relationships provided more role modeling and less career development than relationships involving male mentors. Unexpectedly, male mentors in homogeneous relationships were associated with lower levels of role modeling than female mentors in either homogeneous or diversified relationships. Homogeneous male relationships also offered less psychosocial support than female mentors in diversified relationships with male protégés. Male mentors in diversified relationships with female protégés were associated with more career development than any other gender combination of mentoring relationship.

An increasing number of men and women in organizations are recognizing the benefits associated with mentoring relationships, in which individuals with advanced experience and knowledge (i.e., mentors) provide support and facilitate the upward mobility of junior organizational members (i.e., prote´ge´s) (Ragins, 1997). These benefits include increased self-esteem at work (Koberg, Boss, & Goodman, 1998), increased job satisfaction and decreased work alienation (Koberg, Boss, Chappell, & Ringer, 1994), effective socialization of young employees (Schein, 1978), promotions and increased compensation (Dreher & Ash, 1990; Dreher & Cox, 1996), career mobility and advancement (Scandura, 1992; Wilson & Elman, 1990), career satisfaction (Fagenson, 1989), career commitment (Colarelli & Bishop, 1990), job satisfaction (Bahniuk, Dobos, & Hill, 1990), and reduced turnover intentions (e.g., Koberg et al., 1998). In addition, Ragins (1989) identified mentoring as important for men but essential for women, since mentors can buffer them from discriminatory selection and treatment.While mentoring has been identified as a key career resource for men and women in organizations (Bowen, 1986; Dreher & Ash, 1990; Ragins & Cotton, 1991), the mentoring literature remains unclear regarding the role of gender in mentoring relationship processes and outcomes. Some research (Maniero, 1994; Ragins, 1989) has linked mentoring (primarily male mentor) relationships to advancement of female prote´ge´s, whereas other research (e.g., Dreher & Ash, 1990; Ensher & Murphy, 1997; Fagenson, 1989; Ragins & McFarlin, 1990) has failed to find associations between gender and mentoring relationship processes and outcomes. These conflicting results have prompted numerous writers (e.g., Allen, Russell, & Rush, 1994; Burke & McKeen, 1990; Ragins & McFarlin, 1990; Tharenou, Conroy, & Latimer, 1994) to call for research to expand what we know about the role of gender composition of mentoring dyad relationships and career advancement. As Ragins (1997) noted, most prior research has examined independent main effects of the mentor’s or prote´ge´’s gender without considering the gender composition of the mentoring relationship. She argued that the composition of the mentoring relationship leads to unique behavioral and perceptual processes. As such, a dyadic approach is methodologically superior to previous designs and should be used to explore the influence of the gender composition of the relationship on mentoring processes and outcomes (Ragins).A few studies have investigated the effects of gender composition of mentoring relationship on mentoring processes and outcomes, and these efforts have produced inconsistent results. Koberg et al. (1998) found that characteristics of the mentoring dyad (e.g., gender and race) and prote´ge´ characteristics (e.g., education and ethnicity) had significant effects on psychosocial mentoring. They found same-gender dyads provided more psychosocial support to the prote´ge´ than diversified dyads. However, Ensher and Murphy (1997) failed to find such results. Burke, McKeen, and McKenna (1990) found female mentors provided more friendship, counseling, personal support, and sponsorship in same-gender dyads than did any other gender composition. Female mentors provide more exposure, protection, counseling, and promotional opportunities (Burke et al., 1990) and are more willing to mentor others (Allen, Russell, & Maetzke, 1997). Yet, Burke and McKeen (1997) found that female prote´ge´s with female mentors reported greater intentions to leave an organization. Ragins and McFarlin (1990) noted that prote´ge´s in cross-gender dyads reported less role modeling and social role modeling functions received than prote´ge´s in same-gender dyads.These ubiquitous results may be a function of methodological issues confounding meaningful comparisons across studies. Several studies had unequal gender compositions (e.g., Burke et al., 1990, 1993; Koberg et al., 1998), industry-specific samples (e.g., Koberg et al., 1998), or tested for main effects of gender while ignoring gender composition of the dyad (e.g., Allen et al., 1997). All had uneven mentor/prote´ge´ homogeneous or diversified dyad cell sizes. Accordingly, the present study’s objectives were to investigate the effects of the gender composition on mentoring relationships using a dyadic approach and to attempt to increase the cell sizes by identifying a relatively equal sample of men and women. To this end, we examined mentoring functions received (career development, psychosocial support, and role modeling) in same-gender (homogeneous) and cross-gender (diversified) mentor/prote´ge´ dyads composed of working professionals from a variety of industries. Career development functions include sponsorship, protection, challenging assignments, exposure, and visibility. Psychosocial support functions include acceptance, coaching, counseling, and role modeling (Kram, 1985). Scandura (1992) found role modeling to represent a third distinct function. Role modeling functions involve behaviors and attributions in which prote´ge´s identify with and emulate mentors, who are trusted and respected, possess much referent power, and hold high standards (Godshalk & Sosik, 1998). As such, the present study can be positioned as a contribution to Ragins’ (1997) argument about the importance of gender and composition of the mentoring relationship in perceptions of mentoring processes and outcomes.

Several key study results shed some light on the role of gender composition in mentoring relationships and the nature of diversified and homogeneous mentoring relationships. First, study results indicate that male mentors were perceived to provide higher levels of career development support to their prote´ge´s than female mentors. This result parallels Ibarra’s (1992) finding that male mentors are associated with higher perceptions of instrumental or career development functions received by prote´ge´s and provides empirical support for Ragins’ (1997) proposition that mentoring relationships involving minority (e.g., female) mentors will be perceived to provide fewer career development functions than relationships involving majority (e.g., male) mentors. The shortage of female mentors (Ragins, 1989, 1997) and the perception by both genders that men hold more and different forms of power to advance the prote´ge´s’ career (Ragins & Sundstrom, 1989) may explain the present finding. In addition, this result provides support for social role theory (Bem, 1974), helping behavior theory (Eagly & Crowley, 1986), and network support theory (Ibarra, 1993), which consider instrumental forms of helping behavior (e.g., career development) as most consistent with the male gender role.Second, homogeneous mentoring relationships with majority (i.e., male) mentors were not associated with the highest levels of role modeling (i.e., idealized influence behaviors and attributes), psychosocial, and career development functions received. Contrary to expectations, male mentor/male prote´ge´ dyads were associated with lower amounts of idealized influence behaviors and psychosocial support than female mentor/male prote´ge´ dyads and lower amounts of idealized influence attributes and behaviors than female mentor/female prote´ge´ dyads. The result concerning the superiority of homogeneous female dyads over homogeneous male dyads in terms of idealized influence behaviors and attributes parallel results found by Ragins and McFarlin (1990), who reported that female prote´ge´s with female mentors were more likely to agree with the idea that their mentor served a role modeling function. Burke et al. (1990) also found similar results regarding the superiority of homogeneous female dyads over other gender dyad combinations in providing psychosocial support (including role modeling). These results corroborate Ragins’ (1997) contention that the role modeling function is particularly important for minority prote´ge´s who may perceive a minority mentor as one who has successfully overcome structural, political, and discriminatory barriers to career advancement. The result concerning the superiority of female mentor/male prote´ge´ dyads over homogeneous male dyads in terms of idealized influence behavior conflicts with prior research on identification of the prote´ge´ with the mentor (e.g., Levinson, Darrow, Klein, Levinson, & McKee, 1978; Ragins & McFarlin, 1990; Ragins, 1997). This research suggests that diversified relations are perceived to provide fewer role modeling functions than homogeneous relationships because role modeling in diversified relations may be attenuated due to non-overlapping social identities stemming from membership in dissimilar gender groups. However, a mentor’s consideration of trust, values, beliefs, and ethics through idealized influence behavior may have provided role modeling functions that are equally valued by both gender groups. Such idealized influence behavior is gender-neutral androgynous behavior (Bass, 1998) and promotes identification of the prote´ge´ with the mentor (Godshalk & Sosik, 1998). Scandura and Ragins (1994) found that individuals with androgynous sex role orientations reported receiving more mentoring functions than individuals with feminine or masculine orientations. Prior research on transformational leadership [see reviews by Bass (1998) and Avolio (1999)] has found that women leaders are likely to build trust more quickly than men by exhibiting idealized influence behavior. As such, male prote´ge´s may have perceived female mentors as role models because female mentors are more likely to build trust through the exhibition of idealized influence behavior. Such perceptions of trust may enhance the male prote´ge´’s identification with the female mentor and result in more positive ratings of idealized influence behavior (Covey, 1997). Study results provided further evidence of the efficacy of cross-gender dyads in mentoring relationships. Specifically, male mentor/female prote´ge´ dyads reported higher amounts of career development functions than any other combination of the mentoring relationship. This result provides support for Maniero (1994), Ragins (1989), and Ragins et al. (1998), who argued that career development may be perceived as essential by women. Male mentors, as members of the majority group, may be perceived as possessing and wielding more power than female mentors (Ragins & Sundstrom, 1989). Such power may be used to garner physical and political resources for career development and therefore buffer female prote´ge´s from discrimination and help shatter the glass ceiling. While male and female prote´ge´s in the present study were found to be at similar organizational ranks, the female prote´ge´s, as members of a minority group, may continue to face interpersonal and organizational obstacles to career advance ment (Allen et al., 1994; Ragins et al., 1998). Career development functions provide women with improved chances for promotions, higher income, and higher satisfaction with pay and benefits (Dreher & Ash, 1990). Thus, career development functions provided by male mentors may be more salient to female prote´ge´s wishing to advance in their careers. Third, the absence of differences between homogeneous and diversified mentoring relationships in terms of mentoring functions received is noteworthy for future research. One possible explanation for the lack of results concerns differential impacts of biological sex and gender role orientation on mentoring functions. Prior research (Scandura & Ragins, 1993) has found gender role orientation to be a stronger predictor of mentoring functions than biological sex. As such, proposed differences in mentoring functions across diversified and homogeneous mentoring relationships may be more of a function of gender role orientation than biological sex. Alternatively, prote´ge´s may respond more to a mentor’s power than a mentor’s gender (Ragins, 1991). For example, expert and referent power are important to ratings of idealized influence (Bass, 1998). Given the lack of differences in idealized influence attributes and behavior across diversified and homogeneous mentoring dyads noted in the present study, perhaps mentors derive their power from sources other than expert and referent power. For example, position power stemming from organizational rank may be more important to the career advancement of a prote´ge´. Position power may have played a part in the results of the present study because 85% of mentors were either immediate managers or supervisors of the prote´ge´s. Prote´ge´s may feel that their mentor wields significant influence in the prote´ge´s’ career progression and therefore may blindly submit to the mentor’s directions due to the mentor’s position rather than accepting information offered as mentor advice. Only Fagenson’s (1988) study found mentored individuals perceiving more resource-based power due to their relationship with the mentor than nonmentored employees. Accordingly, future research might examine how differential perceived power bases (e.g., referent, expert, positional) may interact with gender composition of the dyad to influence mentoring processes and outcomes.mentoring processes and outcomes. Several sample-related study limitations are suggestive of future research paths. First, study participants were graduate students, who collectively represented working professionals from a variety of ages, backgrounds, occupations, and industries. Such a sample was judged preferable to using employees within the same organization due to the potential for data reflecting shared participant pool, organizationally specific values, or mentoring relationships that may or may not be representative of the general population. Nevertheless, limitations of generalizations from “convenient” sample data are acknowledged. For example, we advise caution in making inappropriate generalizations of our findings to female-dominated occupations such as nursing or teaching, where many female mentors might be available. Second, our sample consisted predominately of male mentors and prote´ge´s and therefore our findings require replication in future research. For example, due to the relatively small number of dyads in the female mentor/male prote´ge´ cell, results warrant cautious interpretation because the present study may be examining cross-gender relationships for women but not for men. Subsequent investigations could employ samples from multiple industries consisting of an equal number of dyads in same-gender and cross-gender cells. Third, given that 91% of our sample were involved in informal mentoring relationships, the results of the present study are generalizable to informal mentoring relationships. Chao et al. (1992) noted distinctions in process between formal and informal mentoring relationships. Future research should replicate the present study using a sample composed of primarily formal mentoring relationships. Also, given that 85% of the sample included mentors who were the direct supervisors of the prote´ge´s, the results of this study should be generalizable to these types of dyads. Future research should attempt to investigate differences between process and outcomes of supervisory and nonsupervisory mentor– prote´ge´ dyads. For example, nonsupervisory mentors may provide lower levels of perceived career development functions than supervisory mentors due to a perceived lack of position power than can be used to advance a prote´ge´’s career. A set of methodological limitations also provides opportunities for future research. First, the present study examined the effects of the gender composition of the mentoring relationship on mentoring functions provided. Future research could focus on how variables such as tenure and education (Koberg et al., 1998), culture (Cox, 1993), stages of professional careers (Dalton, Thompson, & Price, 1977), and stages of adult life (Levinson et al., 1978) may interact with the gender composition of the mentoring relationship to influence mentoring function variables. Second, the present study used prote´ge´ ratings of mentor idealized influence behavior and attributes, measured by the MLQ-5X (Bass & Avolio, 1997), as a proxy measure for the role modeling function. In future studies, investigators may wish to use specific role modeling items from Noe (1988) or other measures of role modeling behavior (e.g., Scandura, 1992). In addition, researchers may wish to use multiple measures of role modeling behaviors and attributes to assess the convergent and discriminant validity of the role modeling construct in mentoring contexts. Third, the present cross-sectional study collected data at one point in time. Such a “snapshot” approach does not allow an examination of the changes in mentoring relationship processes and outcomes that evolve over time (e.g., Levinson et al., 1978; Ragins & Scandura, 1997). This approach also precludes causation from being accounted for by the ANCOVA designs employed in the present study. It is therefore important for future research to examine mentoring relationships using longitudinal study designs,collecting data at multiple points over time and analyzing the data using techniques which allow for examination of causal inference (e.g., LISREL, partial least squares). Results of the present study raise an interesting implication for practitioners. Prior research suggests that mentors shy away from cross-gender relationships with prote´ge´s due to potential problems such as sexual innuendoes (Ragins,1989) and lack of perceived similarity (Ensher & Murphy, 1997). However, study results indicate that female mentor/male prote´ge´ dyads were associated with the higher ratings of role modeling and psychosocial support functions than homogeneous male dyads. Moreover, male mentor/female prote´ge´ dyads were associated with the highest amounts of career development functions. Therefore, organizations wishing to enhance the career development of women and the psychosocial well-being of men may consider developing and delivering training modules addressing issues critical to the success of cross-gender mentoring relationships (e.g., gender differences, sexual harassment, transformational leadership). For example, modules on transformational leadership can help mentors understand how idealized influence may be applied to relationships with their prote´ge´s. Such training may focus on teaching mentors to learn how to talk about shared values and beliefs, build prote´ge´ trust, and respect for and identification with the mentor and to appreciate the prote´ge´’s potential to contribute in unique ways. Transformational leadership involves finding a common ground based on shared values of human development, cooperation, and need for positive change (Avolio, 1999). As such, mentors who display transformational leadership may encourage their prote´ge´s to rise above identification problems associated with cross-gender mentoring relationships. This implication is especially important given trends of increasing diversity in organizations (Cox, 1993), glass ceiling limitations to career advancement for women (Morrison et al., 1992), and the limited supply of female mentors in organizations (Ragins, 1997). Additionally, educating both genders on how to communicate in a manner that works effectively for both parties (Tannen, 1995) may encourage successful mentor–prote´ge´ dyads. In conclusion, this study has broadened our understanding of the effects of gender composition of mentoring relationships on prote´ge´ perceptions of role modeling and psychosocial and career development mentoring functions received. By providing an empirical test of Ragins’ (1997) model of diversified and homogenous mentoring relationships, this study has shed some light on the nature of mentoring processes in diversified and homogeneous mentoring relationships. By examining these relationships using a dyadic approach, it has also provided a more balanced focus of research to include both the mentor’s and prote´ge´’s characteristics and their interactions. It is hoped that this paper encourages practitioners to consider ways they might employ diversified mentoring relationships to develop individual potential. Moreover, it is hoped that this paper stimulates further research on diversified and homogeneous relationships beneficial to both men and women in organizations