Ragins Mentor Functions And Outcomes

Mentor Functions and Outcomes: А Comparison of Men and Women in Formal and Informal Mentoring Relationships The authors exarnined the effects of the type of mentoring relationship and the gender composition of the relationship on mentoring functions and career outcomes reported Ьу 352 female and 257 male proteges. Proteges of informal mentors viewed their mentors as more effective and received greater compensation than proteges of formal mentors. Proteges with informal mentors also received more career outcomes than nonmentored individuals, but no significant differences were found between nonmentored and formally mentored individuals. Тhе gender composition of the relationship affected mentoring functions and outcomes, and protege gender interacted with the type of relationship to affect mentoring functions. Mentoring relationships are а critical career resource for employees in organizations. Mentors are individuals with advanced experience and knowledge who are committed to providing upward support and mobility to their protege's careers (Hunt & Michael, 1983; Кram, 1985а). Mentors help their proteges Ьу providing two general types of be­ haviors or functions: career development functions, which facilitate the protege' s advancement in the organization, and psychosocial functions, which contribute to the protege's personal growth and professional development (Кram, 1985а). The presence of а mentor is associated with an array of positive career outcomes: Proteges receive more promo­ tions (Dreher & Ash, 1990; Scandura, 1992), have higher incomes (Chao, Walz, & Gardner, 1992; Dreher & Ash, 1990; Whitely, Dougherty, & Dreher, 1991), and report more moЬility (Scandura, 1992) and career satisfaction (Fa­ genson, 1989) than nonproteges. Mentoring has also been found to have а positive impact оп organizational socializa­ tion (Ostroff & Kozlowski, 1993), job satisfaction (Koberg, Boss, Chappell, & Ringer, 1994), and reduced tumover intentions (Viator & Scandura, 1991). Belle Rose Ragins, School of Business Administration, Univer­ sity of Wisconsin-Milwaukee; John L. Cotton, Department of Management, Marquette University. An earlier version of this article was presented at the meeting of the National Academy of Management, August 1998, San Diego, Califomia, and received the Best Applied Paper Аward in the Ca­ reers Division. This study was funded Ьу а 1991 grant from the Women's Bureau of the U.S. Department of Labor. Coпespondence conceming this article should Ье addressed to Belle Rose Ragins, School of Business Administration, Р.О. Вох 742, University ofWisconsin, Milwaukee,Wisconsin 53201. Elec­ tronic mail may Ье sent to Ragins@uwm.edu. Many organizations recognize the important benefits of mentoring and have attempted to replicate inforrnal men­ toring relationships Ьу creating forrnal mentoring programs (Burke & McKeen, 1989; Geiger-DuMond & Boyle, 1995; Zey, 1985). One key difference between forrnal and infor­ mal mentoring relationships is that inforrnal mentoring re­ lationships develop spontaneously, whereas forrnal mentor­ ing relationships develop with organizational assistance or intervention-usually in the foпn of voluntary assignment or matching of mentors and proteges. А second distinction is that forrnal relationships are usually of much shorter duration than infoпnal relationships (Douglas, 1997). Although it is clear that forrnal and informal mentoring relationships differ in how they are foпned and the length of the relationship, there is little research оп whether formal and infoпnal relationships differ in the functions mentors provide or the career outcomes proteges obtain during the mentoring relationship. Many organizations simply assume that formal relationships are as effective as informal rela­ tionships and implicitly offer their employees forrnal rela­ tionships as а substitute for informal mentoring relation­ ships (Keele, Buckner, & Bushnell, 1987; Кram & Bragar, 1992). Moreover, forrnal mentoring programs are being implemented across the nation: It is estimated that а third of the nation's major companies have а formal mentoring program (Bragg, 1989), and this figure is expected to con­ tinue to increase (Мuпау, 1991). These forrnal mentoring programs are being developed without the benefit or guid­ ance of empirical research. This situation has particular relevance for women, who face greater barriers to developing informal mentoring re­ lationships than men (Ragins & Cotton, 1991) and тау therefore Ье more likely to seek forrnal relationships as а substitute for infoпnal mentoring relationships. Addition­ ally, many organizations target women for foпnal mentor­ ing programs in the attempt to help them advance in the organization and break: through the "glass ceiling" (Cata­ lyst, 1993; Herry, 1994; Кram & Hall, 1996; Scott, 1992). However, these organizations may not Ье helping their female employees if foпnal mentors are less effective than infoпnal mentors. Accordingly, the first and primary purpose of this study is to compare the mentoring functions and career outcomes associated with foпnal and infoпnal mentoring relation­ ships. А second purpose of this study is to understand whether the gender composition of the relationship affects the relationship's functions and outcomes and whether this effect varies Ьу the type of mentoring relationship. As discussed earlier, many organizations develop foпnal men­ toring programs that focus either exclusively or primarily оп their female employees. The assignment of а male or female mentor may produce different outcomes from the relation­ ship (Ragins, 1989) and may therefore Ье an important factor to consider in the development of а foпnal mentoring program. This study's objective, therefore, is to provide timely infoпnation for organizations seeking to develop mentoring programs and to contribute to the knowledge base necessary for emerging mentorship theory. Literature Review and Hypotheses Mentor Functions According to Кram's mentor role theory (1985а), men­ tors сап provide two broad categories of mentor functions. First, they provide career development functions, which help proteges learn the ropes and facilitate the protege's advancement in the organization. Кram (1985а) theorized that mentors сап provide five specific career development functions: sponsoring promotions and lateral moves (spon­ sorship); coaching the protege (coaching); protecting the protege from adverse forces (protection); providing chal­ lenging assignments (challenging assignments); and in­ creasing the protege's exposure and visiЬility (exposure). Psychosocial functions compose the second broad cate­ gory of mentor functions. These behaviors address interper­ sonal aspects of the mentoring relationship and enhance the protege's sense of competence, self-efficacy, and profes­ sional and personal development. Career development func­ tions depend оп the mentor's power and position in the organization, whereas psychosocial functions depend on the quality of the interpersonal relationship and the emotional bond that underlies the relationship. Career development functions focus оп the organization and the protege's career, whereas psychosocial functions affect the protege оп а more personal Ievel and extend to other spheres of life, such as the protege's personal development. Кram (1985а) theorized that пientors may provide four psychosocial functions: help­ ing the protege develop а sense of professional self (accep­ tance and confiпnation), providing proЫem-solving and а sounding board (counseling), giving respect and support (friendship), and providing identification and role modeling (role modeling). It is important to note that mentoring is not an all or none phenomenon; а given mentor may provide all of just some of these functions. Formal and Informal Mentoring Relationships There are distinct differences between foпnal and infor­ mal mentoring relationships that may impact the mentor's functions and the career outcomes of the relationship. These differences involve the way the relationship is initiated, the structure of the relationship, and the processes involved in the relationship. Initiation of relationship. Infoпnal mentoring relation­ ships develop оп the basis of mutual identification and the fulfillment of career needs. Mentors select proteges who are viewed as younger versions of themselves, and the relation­ ship provides mentors with а sense of generativity, or con­ tribution to future generations (Erikson, 1963). Mentors are usually in rnid-career stages that involve reassessment of life accomplishments (Erikson, 1963; Кram, 1985а; Levin­ son, Darrow, Кlein, Levinson, & МсКее, 1978), and gen­ erativity helps mentors avoid stagnation and allows them to progress to the next life stage. Proteges select mentors who are viewed as role models. Proteges are in early career stages that involve developing а sense of professional iden­ tity, and role modeling helps proteges advance through this stage. This mutual identification leads to the often-cited intensity of the infoпnal relationship and the parallels drawn between mentoring and parent-child relationships. Infoпnal mentoring relationships also develop on the basis of perceived competence and interpersonal comfort (Allen, Poteet, & Burroughs, 1997; Kalbfieisch & Davies, 1993; Кram, 1983, 1985а; Olian, Carroll, & Giannantonio, 1993; Olian, Carroll, Giannontonio, & Feren, 1988). Men­ tors tend to select high-perforrning proteges who are con­ sidered rising stars or even diamonds in the rough. Sirni­ larly, proteges select mentors with desired expertise. Members of infoпnal mentoring relationships select part­ ners they enjoy working with and often report а mutual attraction or chernistry that sparks the development of the relationship (Кram, 1983, 1985а). In contrast, members of foпnal mentoring relationships are typically assigned to one another Ьу а program coordi­ nator on the basis of application forms submitted Ьу the potential mentor and protege (Douglas, 1997; Gaskill, 1993; Murray, 1991). In many cases, the mentor and protege do not even meet until after the match has been made. Thus, in contrast to infoпnal relationships, identification, role mod­ eling, and interpersonal comfort do not play а role in the development of foпnal relationships. It is therefore reason­ aЫe to expect that the psychosocial functions of role mod­ eling, friendship, and counseling may Ье less in foпnal than infoпnal mentoring relationships. Foпnal mentoring rela­ tionships are also less likely to Ье founded on mutual perceptions of competency and respect. Foпnal mentors are selected on the basis of their competency, but this judgment is made Ьу the program coordinator rather than the protege (Gaskill, 1993; Murray, 1991; Phillips-Jones, 1983). Addi­ tionally, foпnal mentors may view their proteges as at-risk perfoпners who enter the program because they need reme­ dial attention (Ragins, 1997а). It is reasonaЫe to expect that the acceptance and confiпnation mentor functions, which are founded on respect and perceived competency, will Ье less in foпnal than infoпnal mentoring relationships. Structure of relationship. Foпnal and infoпnal mentor­ ing relationships differ in the length and foпnality in the relationship. Infoпnal relationships last between 3 and 6 years (Кram, 1985а), whereas foпnal relationships are usu­ ally contracted to Iast between 6 months and 1 year (Mur­ ray, 1991; Zey, 1985). Members of infoпnal relationships meet when desired, but the mode, frequency, and location of contact for foпnal relationships are often specified in а contract signed Ьу both parties (Murray, 1991; Zey, 1985). The goals of foпnal relationships are specified at the start of the relationship and are screened Ьу the program coordina­ tor. In contrast, the goals of infoпnal relationships evolve over time and adapt to the career needs of the individuals. There are three potential outcomes of these different structures. First, infoпnal mentoring relationships have more time to build psychosocial and career development functions. Mentoring may not have an immediate effect on career outcomes; it may take time for the benefits of men­ toring to materialize. Кram (1985а) theorized that the ben­ efits of mentoring extend beyond the duration of the rela­ tionship, and Chao (1997) found that the advantages of being mentored continue over time. This time-lag effect is particularly relevant when considering promotions and compensation, which usually change only once а year. Therefore, infoпnal mentors' career interventions may have more time to reach fruition than foпnal mentors' career interventions. Second, proteges in foпnal relationships may perceive that their mentors spend time with them because of а com­ mitment to the mentoring program and the organization, rather than because of personal commitment to the protege, or because the mentor believes in the protege's potential. Both parties recognize that the relationship is short-teпn and that the mentor may Ье assigned to another protege after the relationship is over. These factors may restrict the develop­ ment of trust and emotional closeness in the relationship and the provision of psychosocial functions. Third, foпnal relationships are contracted to focus on career goals that are short-teпn and relate to the protege's current pos1tюn (Geiger-DuMond & Boyle, 1995; Gray, 1988; Murray, 1991). In contrast, infoпnal mentors are concemed with the long-teпn career needs of their proteges; in some cases the protege's needs may take precedence over the needs of the organization, and the mentor may recom­ mend that the protege move to another organization. Infor­ mal mentoring relationships are therefore more aligned with the view that careers no longer unfold within а single organization but instead become "boundaryless" in span­ ning many different organizational settings (cf. Arthur & Rousseau, 1996; Hall & Mirvis, 1996). Processes in relationship. There are four additional processes that may lead to different functions in foпnal and infoпnal mentoring relationships. First, foпnal mentors may Ье less motivated to Ье in the relationship than infoпnal mentors (Ragins & Cotton, 1991). As discussed earlier, foпnal mentors may not identify with their proteges. In addition, foпnal mentors may enter the relationship to Ье good organizational citizens rather than because of their own developmental needs. Although foпnal mentors may receive more organizational recognition than infoпnal men­ tors, they may not Ье ready to Ье mentors, and they may Ье less likely to receive the intemal rewards associated with mentoring (cf. Ragins & Scandura, in press). It is therefore reasonaЫe to expect that foпnal mentors may not Ье as motivated as infoпnal mentors to provide career develop­ ment and psychosocial functions. Sвeond, foпnal mentors Пiау have less effective commu­ nication and coaching skills than infoпnal mentors (Кram, 1985Ь, 1986). Proteges select infoпnal mentors with strong communication and coaching skills (Kalbfleisch & Davies, 1993; Olian et al., 1988). Foпnal mentors may Ье viewed as having good communication skills Ьу program coordina­ tors, but if the protege does not share these perceptions, the relationship may become strained and less effective. А third factor that may limit the effectiveness of foпnal mentoring relationships is that many programs match mem­ bers from different departments or functional units in the attempt to avoid charges of favoritism (Douglas, 1997; Murray, 1991). This practice may impede the foпnal men­ tor's aЬility to intervene on the protege' s behalf and provide exposure, protection, sponsorship, and challenging assign­ ment functions. Moreover, infoпnal relationships often evolve on the basis of mutual interests, job functions, and career paths. Foпnal mentors who are in different depart­ ments or functional areas than their proteges may also have different career paths and may therefore provide less effec­ tive career counseling and role modeling for their proteges. Finally, because foпnal mentors are more visiЫe than infoпnal mentors, they may Ье more self-conscious about engaging in career development behaviors that may Ье construed as favoritism Ьу others in the organization. Infor­ mal mentors generally engage in such behaviors with im­ punity; infoпnal mentors have been found to sponsor their proteges into upwardly mobile positions, give them chal­ lenging "stretch" assignments, and buffer them from ad­ verse forces in the organization (Кram, 1983; Scandura, 1992). Because formal mentoring relationships are puЫic relationships that are monitored Ьу program coordinators, formal mentors may Ье less likely than informal mentors to intervene оп their protege's behalf. Review of Research As the above theoretical review indicates, formal mentors сап Ье expected to provide less of each of the nine career development and psychosocial functions than informal mentors. Only two studies directly investigated the relation­ ship between type of mentor and mentor functions. Оп the basis of existing theory, both studies proposed greater ben­ efits for informal than formal mentoring relationships. Al­ though the results of these studies were complementary with existing theory, only partial support for theoretical predic­ tions was found. In а study of 212 informal and 53 formal proteges, Chao, Walz, and Gardner (1992) found that pro­ teges in informal mentoring relationships reported more career development functions and higher salaries than pro­ teges in formal relationships. However, по support was found for their hypothesis that proteges in informal rela­ tionships would report more psychosocial functions than proteges in formal relationships. Fagenson-Eland, Marks, and Amendola (1997) found that informal proteges reported more psychosocial benefits than formal proteges in their study of 16 informal and 30 formal proteges employed at two merging organizations. However, informal and formal proteges did not significantly differ in reports of career development and role modeling functions. These studies, although groundbreaking, used only the broad categories of mentoring functions and did not provide an in-depth investigation of the nine specific mentoring functions theorized Ьу Кram (1985а). Chao et al., (1992) used Noe's (1988) 21-item mentor functions instrument, which measures Кram's (1985а) two broad categories of career development and psychosocial functions. The career development scale consisted of 7 items that measure the protection, challenging assignment, and exposure functions, but do not include Кram's coaching and sponsorship func­ tions. Psychosocial functions were measured using а 14item scale that omitted the friendship function and included the coaching function. The coaching function is а career development function (Кram, 1985а) but is loaded оп the psychosocial factor in Noe's scale (Noe, 1988). The instru­ ment is conceptually limited in that it does not allow for an analysis of the nine individual mentor functions because single items are used to measure many functions, and some functions are not represented in the instrument. Fagenson­ Eland and her colleagues (1997) used the Scandura and Katerburg (1988) 18-item mentoring functions question- naire, ·which collapsed Кram's (1985а) nine mentor func­ tions into three broad categories: career development, psy­ chosocial, and role modeling. However, like Noe's (1988) instrument, this instrument assesses only the broad catego­ ries of mentor functions and therefore does not allow for а full assessment of Кram's mentor role theory Ьу investigat­ ing the effects of each of the nine individual mentor functions. One objective of the present study, therefore, was to extend this prior research Ьу providing а more fine-grained analysis of the relationship between type of mentor and each of Кram's (1985а) mentor roles. Toward this aim, we used the Ragins and McFarlin (1990) 33-item mentor role instru­ ment, which allows for а separate analysis of the effects of each of Кram's (1985а) nine mentor functions, as well as two additional functions. Another objective of our study was to assess the relation­ ship between formal and informal mentoring and career outcomes, such as promotion rate and compensation. There has been а lack of research investigating these relationships. As discussed earlier, Chao et al. (1992) explored the rela� tionship between the type of mentoring relationship and compensation. She reported that individuals with informal mentors had greater compensation than individuals with formal mentors, but she did not investigate the relationship between type of mentor and promotion rate. Wе would like to replicate and extend her study Ьу investigating the rela­ tionshi,p between formal and informal mentoring, compen­ sation, and promotion. Our study also attempts to follow Chao et al.'s (1992) lead in recognizing that the effects of mentoring оп career outcomes may not Ье immediate, but тау take place over time. This is particularly relevant when investigating formal mentoring relationships. As discussed earlier, because for­ mal mentoring relationships usually last а year or less (Murray, 1991; Zey, 1985), static measures, although suit­ aЫe for capturing current perceptions of the mentor's be­ haviors or functions, fall short of capturing the career out­ comes of the relationship; the effects of а formal mentoring relationship тау not Ье realized for 1 or 2 years following the relationship's termination. Chao and her colleagues (1992) recognized this issue and wisely included duration of mentoring relationship as а control variaЫe in their study. They also tested for differences between proteges with current or recent mentors and proteges who had relation­ ships that ended 2 years prior to being surveyed. They found по differences between these groups, but they made the valuaЫe point that it is important to examine historical effects and control for the duration of the relationship when investigating the relationship between type of mentoring and career outcomes. Toward that end, we controlled for the duration of mentoring relationship and included retrospec­ tive measures of the history of mentoring relationships in our analyses. Hypotheses of Fonnal and Informal Mentoring Relationships On the basis of the theory and research reviewed above, we hypothesized that proteges in inforrnal mentoring rela­ tionships would report that their mentors provide more career development and psychosocial functions than pro­ teges in forrnal relationships. Hypothesis 1: Proteges in infoпnal mentoring relationships will report that their mentors provide more career develop­ ment functions (sponsorship, coaching, protection, challeng­ ing assignments, exposure) than proteges in foпnal mentoring relationships. Hypothesis 2: Proteges in infoпnal mentoring relationships will report that their mentors provide more psychosocial functions (acceptance and confiпnation, counseling, friend­ ship, and role modeling) than proteges in foпnal mentoring relationships. Wе also expected that inforrnal proteges would report more overaП satisfaction with the perforrnance of their mentor than proteges in forrnal relationships. Hypothesis 3: Proteges in infoпnal mentoring relationships will report greater satisfaction with their mentors than pro­ teges in foпnal mentoring relationships. We also expected а positive relationship between the history of prior mentoring relationships and career out­ comes. In particular, proteges who had primarily inforrnal relationships in the past should report more compensation and а higher promotion rate than proteges with а history of primarily forrnal relationships. Hypothesis 4: Proteges with а history of infoпnal mentoring relationships will receive more compensation and promotions than proteges with а history of foпnal relationships. The next section explores the conjoint effects of the gender composition of the relationship and the type of mentor on mentor functions and career outcomes. Gender Composition of Relationship Existing mentorship theory holds that the gender compo­ sition of the mentoring relationship is а critical factor af­ fecting mentoring functions and outcomes (cf. Ragins, 1997а). Ragins (1997а, 1997Ь) observed that social identity theory (Tajfel, 1978), the similarity-attraction paradigm (Byrne, 1971), and the relational demography perspective (Tsui, Egan, & O'Reilly, 1992; Tsui & O'Reilly, 1989) аП predict more perceived similarity, identification and role­ modeling in same-gender as compared with cross-gender relationships. Carrying this logic to the mentoring arena, Ragins (1997а) predicted that psychosocial functions, which incorporate similarity, identification, and role mod- eling, should also Ье stronger in same-gender as compared with cross-gender mentoring relationships. Hypothesis 5: Proteges in same-gender relationships will re­ port тоге psychosocial functions (acceptance and confiпna­ tion, counseling, friendship, and role modeling) than proteges in cross-gender mentoring relationships. Ragins (1997а) also proposed that mentors' power in the organization influences their aЬility to provide their pro­ teges with such career development functions as sponsoring their proteges to high-ranking positions, protecting them from adverse forces, and giving them needed exposure. She proposed that because majority mentors (i.e., male mentors) generally have more power in organizations than minority mentors (i.e., female mentors), they should Ье better аЫе to provide career development functions and organizational outcomes. This proposition was partially supported in а recent study Ьу Dreher and Сох (1996), which found that proteges with male mentors received greater compensation than proteges with female mentors. However, their study did not investigate career development functions or promo­ tion rates, or the impact of the history of mentoring rela­ tionships on compensation and promotion. Wе therefore wanted to build on their study and test existing theory Ьу proposing the following hypotheses: Hypothesis ба: Proteges with male mentors will report more �eareer development functions (sponsorship, coaching, protec­ tion, challenging assignments, exposure) than proteges with female mentors. Hypothesis бЬ: Proteges with а history of male mentors will report more compensation and promotions than proteges with а history of female mentors. Expanding on this framework, Ragins (1997а) proposed that compared with other gender comЬinations, male pro­ teges with male mentors should receive the most benefits from the mentoring relationship; these proteges should Ье the most satisfied with their relationships and should receive more psychosocial and career development benefits from their relationships than any other gender comЬination. Cor­ respondingly, she proposed that relationships involving mi­ nority mentors (i.e., female mentors) and majority proteges (i.e., male proteges) should Ье the most limited in providing mentoring functions; they are limited in providing career development functions because of the female mentor's rel­ ative lack of power, and they are limited in providing psychosocial functions because the relationship is cross­ gender. Existing theory (Кram, l 985a) and research (Chao et al., 1992) indicates that psychosocial and career devel­ opment mentor functions are inter-related and тау syner­ gistically build on one another, thereby making the differ­ ences between these dyads even more salient (Ragins, 1997а). This theoretical perspective was tested Ьу the following hypotheses: Hypothesis 7а: Male proteges with male mentors will report more career development and psychosocial functions and more satisfaction with their mentors than any other gender combination of mentoring relationship. Hypothesis 7Ь: Male proteges with female mentors will report less career development and psychosocial functions and less satisfaction with their mentors than any other gender comЬi­ nation of mentoring relationship. We also expected that male proteges who had primarily male mentors in the past would receive greater career out­ comes than any other gender comЬination. Similarly, male proteges with а history of primarily female mentors should report less career outcomes than any other gender comЬi­ nation. These expectations are based оп the theory discussed above (Ragins, 1997а), as well as on the finding of positive relationships between compensation and psychosocial and career development mentoring functions (Chao et al., 1992; Scandura, 1992). Hypothesis 8а: Male proteges with а history of male mentors will report greater compensation and promotion than any other gender combination. Hypothesis 8Ь: Male proteges with а history of female men­ tors will report less compensation and promotion than any other gender comЬination. Finally, we wanted to explore whether the gender com­ position of the relationship interacts with the type of rela­ tionship in influencing the mentoring relationship. Because there was no theory or research to provide direction, we tested the following research question: Does the gender composition of the relationship moderate the relationship between type of mentoring (formal vs. informal) and reports of mentor functions and satisfaction with the relationship? То date, only one study investigated the impact of gender composition of mentoring relationships on mentor func­ tions, and there has been no research investigating the relationship between the historical composition of the rela­ tionship and career outcomes. In а survey of 181 proteges in informal mentoring relationships, Ragins and McFarlin (1990) found that same-gender proteges reported engaging in more social activities with their mentors than cross­ gender proteges and that female proteges with female men­ tors reported more role modeling than any other gender comЬination. However, one limitation of this study was that а restricted sample of female mentors prevented а compar­ ative analysis of all four gender comЬinations. Because this situation is relatively common in male-dominated organi­ zations and occupations (Noe, 1988; Ragins, 1989), we obtained а sample of individuals from male-typed, female­ typed, and gender-integrated occupations to obtain adequate sample sizes for each of the four gender comЬinations of mentoring relationships. In а related study, Ensher and Murphy (1997) examined the effects of race similarity among 104 minority students who were assigned to formal, same-gender mentoring rela­ tionships that lasted for 8 weeks. The researchers found that minority proteges assigned to same-race mentors reported more career development functions than minority proteges assigned to different race mentors but found no support for their prediction that psychosocial support would Ье greater among same-race mentors or that female mentors would provide more psychosocial support than male mentors. However, because proteges were assigned to same-gender mentors, the researchers were unaЫe to examine the effect of the gender composition of the relationship on mentoring functions. In this study, we examined the effects of gender compo­ sition and the type of mentoring relationship оп mentoring functions and outcomes. We extended prior research and tested new theory Ьу measuring specific mentor roles and Ьу investigating the impact of history of mentoring relation­ ships оп career outcomes. Method Procedure and Respondents Sampling procedure. One goal of our study was to investigate the effects of the gender composition of the mentoring relationship оп mentor functions and outcomes. Because male-dominated oc­ cupations have а shortage of higher-ranking, female mentors (Ra­ gins, 1989), and this study called for а sample of male proteges with female mentors, we obtained а sample of men and women in male-typed, female-typed, and gender-integrated occupations. Тhree occupations were selected оп the basis of labor statistics and research оп occupational gender-typing (Beggs & Doolittle, 1993; U.S. Department of Commerce, 1996): engineering (male­ dominated), social work (female-dominated), and joumalism (gender-integrated). Forrnal mentoring programs are used in all three of these occupations (Paine, 1986; Smith, Chase, & Byrd, 1986; TaibЬi, 1983). Wе oЬtained а national random sample Ьу using mailing lists of national professional associations representing these three occupa­ tions. То obtain а gender-balanced sample, equal numbers of male and female names were randornly selected from each mailing list. А total of 3,000 surveys were mailed; 1,000 (500 to men, 500 to women) were sent to each of the three occupations. Follow-up surveys and reminder letters were sent according to а modified version of the Dillman mail survey method (Dillman, 1978). А total of 1,258 surveys were retumed, for а response rate of 42%. Self-employed and retired employees were excluded from analy­ ses. Relatively complete data for analyses were availaЫe for 1,162 respondents. Respondents. Respondents consisted of 654 women and 500 men; 8 respondents did not report their gender. Тhе occupational breakdown of the respondents consisted of 362 joumalists (31.2%), 414 social workers (35.6%), and 386 engineers (33.2%). We used an estaЫished definition of mentor (Ragins, 1989) in our survey: "А mentor is generally defined as а higher ranking, influential individual in your work environment who has advanced experience and knowledge and is committed to providing upward mobi1ity and support to your career. Уour mentor тау or тау not Ье in your organization and s/he тау or may not Ье your imme­ diate supervisor." Foпnal mentoring was defined as follows: "In order to assist individuals in their development and advancement, some organizations have estaЫished formal mentoring programs, where proteges and mentors are linked in some way. This тау Ье accomplished Ьу assigning mentors or Ьу just providing foпnal opportunities aimed at developing the relationship. То recap: For­ mal mentoring relationships are developed with organizational assistance. Informal mentoring relationships are developed spon­ taneously, without organizational assistance." То ensure that re­ spondents had а clear understanding of the distinction between formal and informal mentors, we asked respondents to describe their formal mentoring program immediately following the ques­ tion asking them to identify whether their mentor was formal or informal. Wе also asked the respondents to describe the history of their mentoring relationships over the last 10 years, starting with their most recent mentor. For each relationship, we asked the respondents to give information on the mentor's gender, the type of relationship (formal/informal), the mentor's position (supervi­ sory/nonsupervisory), and the duration of the relationship. On the basis of these definitions, 510 respondents (43.9%) reported having an informal mentor, 104 respondents (9%) re­ ported having а formal mentor, and 548 (47.2%) did not have а mentor. The final sample was composed of these 614 proteges, which consisted of 352 female proteges, 257 male proteges, and 5 who did not report their gender. Although men and women did not significantly differ in having а mentor, X"(l, N = 1,135) = 0.36, ns, men were nearly twice as likely to Ье in а formal mentoring relationship: 22% of the men reported cuпently being in а formal relationship, compared with 12% of the women in the sample, X'(l , N = 608) = 10.39, р < .001. Men were also significantly more likely than women to have а history of formal mentoring relationships: 18.5% of the men reported having primarily formal mentors in their past, compared with 9.7% of the women, X"(l, N = 579) = 9.37, р < .01. One implication of these gender differences is that if the type of mentoring relationship does affect mentoring functions and outcomes as proposed, the type of men­ toring relationship should Ье included as а control variaЫe when testing the gender composition hypotheses. The sample involved 348 (57.1%) individuals in same-gender relationships and 261 (42.9%) in cross-gender relationships. А more fine-grained breakdown revealed 233 male proteges with male mentors, 115 female proteges with female mentors, 24 male proteges with female mentors, and 237 female proteges with male mentors. The median length of the cuпent mentoring relationship was 4.0 years, and the average length was 6.7 years. The cuпent mentoring relationship for male proteges was significantly longer than for female proteges, t(426) = -2.77, р < .01, indicating that this variaЬle should also Ье used as а control variaЫe in the analysis. Fifty-three percent of the proteges reported that their mentors were also their supervisors, and there were no significant gender differences on this variaЫe, X"(l, N = 611) = 1.06, ns. The average age of the proteges was 46 years, and 92% were Cauca- sian. Most were married (70%) and had completed (63%) or pursued (12%) graduate degrees. Measures History of mentoring relationships. We asked our respondents to give information about each mentoring relationship they had over the last 10 years, as well as their cuпent mentoring relation­ ship. Тhеу were asked to report whether their prior relationships were formal or informal, the gender of their mentors, their men­ tors' positions, and the duration of the relationships. Тhе history of prior relationship type was measured Ьу dividing the number of prior relationships that involved formal mentors Ьу the total number of prior formal and informal mentoring relation­ ships. Higher values therefore reflect а greater proportion of for­ mal than informal mentoring relationships. The history of prior gender of mentors was measured Ьу dividing the number of prior relationships involving male mentors Ьу the total number of rela­ tionships involving both male mentors and female mentors. Higher values thus reflect а greater proportion of male than female men­ tors. The history of relationships involving specific gender com­ positions was computed in а sirnilar manner. Mentor functions. The Mentor Role Instrument (MRI) was used to measure mentor functions (Ragins & McFarlin, 1990). Тhе 33-item instrument was developed via confirmatory factor analy­ sis, and independently measures each of Кram's (1985а) nine mentor roles. Тhе instrument also assesses two additional psychosocial-related roles: parent and social interactions. Accord­ ing to Кram (1985а), these roles may emerge in response to gender issu�§ in mentoring relationships; proteges may seek to avoid sexual issues in cross-gender relationships Ьу viewing their mentor as а parent figure or Ьу avoiding informal, after-work social interactions. The MRI has proven reliaЬility and prelirninary evidence of validity (Ragins & McFarlin, 1990). The 33-item instrument has 3 items per mentor role and was measured on а 7-point Likert scale with responses ranging from 1 (strongly disagree) to 7 (strongly agree). The instrument may Ье found in the Appendix. Тhе coef­ ficient alphas for the eleven mentor roles ranged from .63 to .91 and are listed on the diagonal of the coпelation matrix in ТаЫе 1. Mentor satisfaction. The protege's satisfaction with the men­ tor was measured Ьу а 4-item scale. The items, which are listed in the Appendix, used the same 7-point Likert scale used in the MRI. The coefficient alpha for the satisfaction scale was .83. Career outcomes. Promotion rate and compensation were used to measure mentor outcomes. Using estaЫished definitions of promotions (cf. Whitely et al., 1991), promotions were defined as involving two or more of the following criteria: significant in­ creases in annual salary, significant increases in scope of respon­ siЬility, changes in job level or rank, or becorning eligiЫe for bonuses, incentives or stock plans. Given the above definition, respondents were asked how many promotions they received over the last 1О years. Respondents were also asked to report their cuпent annual compensation, which included salary, bonuses, commissions, stock options, and profit sharing. Control variaЫes. Six variaЫes that may Ье related to promo­ tion rate and compensation were considered as potential covari­ ates: organizational rank, organizational tenure, position tenure, number of career interruptions, education, and occupation (engi- 536 tfj 3 "i "^ ^ .0 -S %> £ y -SJ •5 "z* a „ a -c -Ss: ^ •2 a S Q "P I C i § NO CM CM ^ CM C-> CM CM CM O CM ON oo r- NO '"' S cr> ~~* \ O O O m m oJ o o o \ \ I ooo \ \ C •> —> <-' 1 c3 orj c3 \o — o —' -« -3- ON c CO —i — —• o 3 o o o c1 \j —> <-< ! p-.. -rj- —> o o O rn —-i \ \ \ m CM r RAGINS AND COTTON NO I r1 _i -^ f. O OC | 3r-cMmooocMOinooNOoo """' ^ O —i —' CM CO ^_^ ^ " ' Sooog-S^S^ftSSoSmS ol O -H O I CM Tj- - -C - H —^ i n c M m c M N O - —i —" O < n c M o o — O Cs o o o ^ o c M c n — m c M N o m c M r ^ 1 " ' 1 i ON NO C — r- - —[ ( M e *1 O CO •<*• —' m 1 ^ O ^^^S^SmS^^^^SS^ O w rru-iocNcNr-nm — m m ^ o o m c M '—' CS — 1 r*"> O 1 O O - O cO t— C O OC 1 CO O 1 S — •<*• O 1 — < Tt f _, o C 2 1 r- NO v 1 o 5 i3 o o —« co ^r r-^ I — O C^ o o o — _ H C 4 r - . o o r - , oio o N o o i n v ? O - H — (Mf-- nNo>nco -g3gooo^^oo§ ! 1 7 1 1 C 1 —— rn \ NO oo o — 00 T -H -H 1 H O O r^ oo fi oo \ \ ^- ^H r- o O 1 CM — CM 1 oo 1 —| O OC o r- c- ON. r- v o -^ r-H -H o u 1 5*".c:«> IS £ c S-° u IIj > ON O —< O _ w m t; C5 •ar —i O 1 1 c— ro o CM NO — — cM ^- O O t c2 r•1 c3 C3 "^ O 1 — \o —i o oo \ 1 \ ooo O CM O 1 1 c*"i oo in -HO— CM O ON 0 ex O .™ oo •""'>"» O -^•^ w -CMm^J- «DNO r- S 1 u 5 1 1i 1 dI oo ON o— 2222SE:22S?3SSSSS 1 3 1 11£5lll-<3lsl'll(£lll 51 a u 3 c o *- tT 0 C ^ " .t: T3 fa ^ -u J3 —^ *§) S o .1 "i" •§1 x. ^ ""* . ss o— =: ^ ^- o. c c 11 £ 5 ex g •a — ^0 t— i^- !! S = 'I 0 0 "^ C •I a. l | S S ^1 «n ^ c« 11 — "^ [I *""' > l.-f .S>-c ux "oa E5g ^ ^ s1 i s.,1 S" u c3 --^ m o S S ^ _u •" £ 1 r- 2 "«s O •— O c 6 So S1 O QJ "o S "•£ °" W § *S w» " £ O M S g | *- £ 5 "o-n J5 "2S §E. S o H 5 > 87 s 0 c = -i -^ fc ™ ° "«5 .1 "-S £ 1 jj 5 2 c t3 2 i^1^ i§ j^1 i1 £ £ 2 l"| A aj