Targeted Mentoring: Evaluation of a Program

Abstract

Targeted mentoring refers to mentoring aimed at a particular population. This article presents the evaluation of a mentoring program for lesbian, gay, bisexual, and transgender (LGBT) persons in social work education. Forty-three mentors and protégés responded to a survey regarding their program experiences. The results highlight the need for targeted mentoring, although some disparities of experience for mentors and protégés in this program are apparent. In general, mentors felt positive about participating, giving back to the LGBT community, and were more satisfied with their experiences than were the protégés, who were looking for more specific types of instrumental and psychosocial support.

MENTORING, OR THE PROCESS of matching persons who have particular experience with persons with less experience, has been a topic of study since the late 1970s with the publication of texts such as Levinson′s The Seasons of a Man′s Life (1978) and Kanter′s Men and Women of the Organization (1977). Although the study of mentoring began largely in business, it has moved to encompass many other environments. In academia, research has often focused on mentor relationships of graduate students with student advisors and new faculty with more experienced faculty (Brown, Davis, & McClendon, 1999; Campbell & Campbell, 1997; Davidson & Foster-Johnson, 2001; Dixon-Reeves, 2003; Frestedt, 1995; Hollingsworth & Fassinger, 2002; Kurtz-Costes, Helmke, & Ulku-Steiner, 2006; Lark & Croteau, 1998; Ortiz-Walters & Gilson, 2005; Tenenbaum, Crosby, & Gliner, 2001; Wilson, Valentine, & Periera, 2002). Issues of criteria used in matching mentors and protégés have also been discussed in the literature (Armstrong, Allinson, & Hayes, 2002; Eby & Allen, 2002; Eby & Lockwood, 2005; Scandura, 1998; Waters, 2004; Wilson et al., 2002), as well as the need for mentors to be knowledgeable about and sensitive to “population-diversity” issues (e.g., Brown et al., 1999; Davidson & Foster-Johnson, 2001; Lark & Croteau, 1998 Simon, Bowles, King & Roff., 2004). For example, Ensher and Murphy (1997) found that protégés were more satisfied with mentors whom they perceived as similar to them. We call this concept of mentoring that is aimed at a particular population “targeted mentoring.”

In this article we present an evaluation of a targeted mentoring program that was developed for lesbian, gay, bisexual, and transgender (LGBT) persons involved in social work education. This program was geared specifically toward LGBT doctoral students and faculty and was sponsored and implemented by the Council on Social Work Education′s (CSWE) Council on Sexual Orientation and Gender Expression (CSOGE). It is important to note that similar mentoring programs, focusing on LGBT graduate students, exist in other discipline-specific organizations such as the American Psychological Association, the American College Personnel Association, and the American Sociological Association(ACPA SCLGBTA, n.d.; APAGS, n.d.; Barrett et al., 2002).

Literature Review

Within the research base regarding mentoring, much attention is paid to the importance and impact of minority status and mentoring within workplace and academic environments, particularly for women (e.g., Blake-Beard, 2001; Campbell & Campbell, 1997; Ensher & Murphy, 1997; Frestedt, 1995; Kram, 1985; Kurtz-Costes et al., 2006; Ragins & Cotton, 1991; Ragins & Scandura, 1994; Simon et al, 2004) and persons of color (e.g., Brown et al., 1999; Davidson & Foster-Johnson, 2001; Dixon-Reeves, 2003; Ensher & Murphy, 1997; Ortiz-Walters & Gilson, 2005; Simon, et al., 2004). Earlier studies on mentoring focused first on the impact of gender on the mentor relationship to determine the impact of having a same or cross-gender match. Because typically there are fewer women than men in business environments (Kanter, 1977; Kram, 1985), most of these studies looked at female protégés with male or female mentors. Although some studies found little or no difference based on same-gender or cross-gender matching (e.g., Campbell & Campbell, 1997), the majority of studies found positive benefits for female protégés who had female mentors. These benefits included increased satisfaction in the relationship, more instrumental support, and, most overwhelming, more psychosocial support. These benefits may be based on the ability of the mentor to be a stronger role model for the protégé as a result of the mentor′s shared experience of being a woman in a largely male work environment (Frestedt, 1995; Kram, 1985; Kurtz-Costes et al., 2006; Ortiz-Walters & Gilson, 2005).

Similar findings have been reported regarding same race and cross race relationships. Because persons of color are typically underrepresented in many areas of business and academic environments, studies typically looked at same race pairs or relationships in which the protégé was a person of color and the mentor was White. Similar to matching based on gender, the majority of studies found a variety of instrumental and psychosocial benefits for protégés of color in same race matches based on the mentors′ and protégés′ shared experience of being a racial or ethnic minority, as well as possible shared experiences of past discrimination based on race or ethnicity (Brown et al., 1999; Dixon-Reeves, 2003; Ensher & Murphy, 1997; Ortiz-Walters & Gilson, 2005; Simon, et al., 2004). Several studies, recognizing that graduate students and young faculty of color may have a cross race match, point out the need for White mentors to gain a better understanding of the particular issues a graduate student or new faculty of color may face (Brown et al., 1999; Davidson & Foster-Johnson, 2001; Simon, et al., 2004).

These studies point out the potential importance and benefits to protégés of having a mentor with similar characteristics. Although mentors with different backgrounds can provide significant support, protégés who are in underrepresented population groups based on gender and/or race may find it important to have mentors matched on similar characteristics, or targeted mentors. These same race and/or gender pairings may or may not provide additional instrumental support, but almost all studies have found that a matched mentor can provide important psychosocial support and role modeling of how to succeed as a minority group member in work or academic environments. The question is, then, whether these findings are relevant when looking at the issues of sexual orientation and gender expression in work and academic environments. Is it important for protégés who identify as LGBT to have targeted mentoring?

The issues faced by members of the LGBT community in their places of employment or education suggest the potential usefulness of matched or targeted mentoring with LGBT or LGBT-affirming mentors. A number of studies have focused on workplace issues for LGBT individuals (e.g., Croteau, 1996; Lyons, Brenner, & Fassinger, 2005; Morrow, Gore, & Campbell, 1996; Ragins, Cornwell, & Miller, 2003). These results raise concerns about the presence and impact of formal and informal heterosexism and homophobia in the workplace (Barrett, 2001 Croteau, 1996; Lark & Croteau, 1998; Lyons et al., 2005; Morrow et al., 1996; Nauta, Saucier, & Woodard, 2001; Ragins et al., 2003); the impact of possible discrimination and homophobia on career choices (Morrow et al., 1996); the impact of homophobia and heterosexism on job satisfaction (Lyons et al., 2005); and the impacts of the gender, race, and other characteristics of the supervisor and workgroup on perceived heterosexism by the employee (Ragins et al., 2003). In addition, several studies highlight the experiences of students in graduate programs, finding that LGBT students actively seek support for LGBT issues when looking for graduate programs, may experience concerns about their decisions about coming out in the academic environment, and may experience a lack of support for LGBT-oriented research (Barrett, 2001 Greer, 2005; Lark & Croteau, 1998).

These studies point out the many ways in which LGBT individuals may struggle in the work environment. These potential struggles include making the decision to come out to coworkers and employers, and feeling supported and well-adjusted at work and in determining career plans. In some academic environments graduate students face concerns about lack of support, inability to get work or to get funding for research (Barrett, 2001; Greer, 2005; Lark & Croteau, 1998). These issues are compounded, subtly or blatantly, by the ability of many employers to legally discriminate against LGBT employees, whether by denying domestic partner benefits or actually firing employees because of their sexual orientation or gender expression. All of these issues make the need for identity related support apparent.

Very few studies have looked at mentoring relationships or the perceived mentoring needs of LGBT employees or graduate students. Those studies that do exist point to the importance of mentoring relationships and specifically to the importance of having LGBT or LGBT-affirming mentor relationships. Nauta, Saucier, and Woodard (2001) found that LGBT students were more likely to look for mentors with similar sexual orientation and gender than heterosexual students. Lark and Croteau (1998) found many students were looking for LGBT specific mentoring and were seeking LGB mentors or mentors who were supportive and knowledgeable regarding LGBT issues (LGBT-affirming). The graduate students in this study stated that LGBT or LGBT-affirming mentors were helpful for psychosocial and instrumental support in ways that non-LGBT affirming mentors were unable to be (Lark & Croteau, 1998). Employees are also impacted by having perceived support for their sexual orientation. For example, Ragins, Cornwell, and Miller (2003) found that diversity of workgroups for both gender and ethnic background, and an ethnic match with the supervisor, led to reduced feelings of heterosexism and surmised that the presence of diversity within a worksite led overall to less perceived heterosexism. Finally, Lin (2001) found that gay and lesbian protégés found more perceived support from gay or lesbian mentors than heterosexual mentors.

Overall, studies on mentoring for LGBT employees and graduate students across disciplines found that these individuals are seeking support and mentoring from and feel more satisfied by support and mentoring from LGBT or LGBT-affirming supervisors and mentors. For a variety of factors, including the results of studies such as these, a number of mentoring programs that focus on providing support to LGBT students have been created over the past 10 years at undergraduate, graduate, and young professional levels (e.g., ACPA SCLGBTA, n.d; APAGS, n.d.; Barrett et al., 2002; University of Michigan Rackham Graduate Studies, n.d.; University of California, Los Angeles Student Psychological Services, n.d.. To date, however, there have been no published studies investigating the utility of programs that work to match mentors and protégés based on their shared sexual orientation or gender expression. The current study was completed to evaluate a particular mentoring project and gain a better understanding about mentoring programs of this type. In this particular program formal pairing of mentors and protégés is coupled with informal structure, in which the matched pair determines the nature and extent of the relationship. We focused on the following questions in designing the evaluation:

• Are mentors and protégés satisfied with the program?

• What is working well with the program, and what may need to be changed?

• What goals did protégés have for the relationship? Were these goals met?

• What goals did mentors have for the relationship? Were these goals met?

Method

Participants

In 1999 CSWE′s CSOGE began a mentoring program to provide a support network for doctoral students and new faculty who either identified as LGBT or wanted support on LGBT-related research with LGBT or LGBT-affirming faculty. The goal of this program was to meet a perceived need within the academic environments of social work to provide support for LGBT doctoral students and new faculty. Participants were recruited through print and online list server advertisements distributed by CSOGE and advertisements at conferences such as the CSWE Annual Program Meeting, the Society for Social Work Research, the National Association of Deans and Directors, and the Group for the Advancement of Doctoral Education. There were also periodic mailings regarding the project to social work deans and directors as well as doctoral program directors. Potential mentors and protégés were continually recruited into the program over the 6 years since the program began until the time of this study.

Anyone interested in being a mentor or protégé received a letter from the mentoring project coordinator that outlined the purpose of the project and requested that he or she fill out a questionnaire, which asked about demographics, primary substantive area, LGBT practice and research interests, and areas in which the individual would like to either provide or receive mentoring. Then mentors and protégés were matched on a variety of characteristics such as expectations of the mentor relationship, similarity of substantive area, gender, and research interests. There were no formal guidelines to determine the frequency, length, or nature of the mentor–protégé relationship. Each party, however, received a letter from the project coordinator requesting that the mentor initiate contact, and, with mutual permission, each party received a copy of the other person′s questionnaire. The project coordinator was available for any questions from mentors and protégés about the program and made changes in pairings as requested by either the mentor or protégé, but no formal follow-up with mentor and protégé pairs was completed. At the time of this study, 6 years after the program began, 63 potential mentors and 52 potential protégés had completed questionnaires to participate in the program.

Although there had been some informal, largely positive feedback given to CSOGE on the program, there was no formal evaluation of the program. To get a better understanding of the strengths and weaknesses of the program, and to decide on future directions, a more formal evaluation was completed.

Materials

Program participants had all shown a willingness to complete a brief questionnaire in order to be a part of the mentorship program. For this reason, a short survey/questionnaire was designed to gather information from the past and current participants in the program to evaluate its effectiveness. To encourage participation, the survey was brief, and the content of consisted of questions that addressed the particular role each participant played in the program. Questions were a mixture of openended and 5-item Likert scale questions. We focused on the following content areas of the survey in this analysis:

• Background information: academic status, level of LGBT discrimination in work or school environment

• Status/length of mentor relationship

• Amounts/types of contacts and satisfaction with each

• Goals for the relationship, whether or not goals were met

• Role of this program versus other mentor relationships

• Strengths and weaknesses of the program

Design and Procedures

The sample for this study consisted of all 115 past and present members of the program. The survey was first sent by mail to all participants in the mentoring project. E-mail versions were sent to those participants who had not returned the survey 3 weeks after the initial mailing. If the mentor or protégé had been in more than one relationship, he or she was asked to fill out a survey for each relationship. A total of 43 responded, consented to participate, and completed the evaluation; five surveys were undeliverable by mail and e-mail. This led to a response rate of 39%. Of the surveys that were returned, 21 respondents were mentors, 13 respondents were protégés, and 9 respondents had applied to be mentors but reported they had never been matched. Because there is a larger mentor than protégé pool, this is an expected result.

A number of methods were used to analyze and describe the data. We treated the Likert scale questions as ordinal data and used tests for ordinal data to analyze the responses (Clason & Dormody, 1994; Jakobsson, 2004). Somer′s d tests were used to look at measures of association (Clason & Dormody, 1994; Garson, n.d.; Yaffee, 2003). With the open-ended questions, content analysis was used to find categories or themes in the data, and then the frequency of each theme or category was tabulated (Neuendorf, 2002).

Results

Background Information

All protégés and all but one mentor were currently employed in universities. As is shown in Table 1, both mentors and protégés rated their work environments as LGBT friendly, with mentors indicating that their workplace was very supportive and protégés indicating their workplace as supportive. Only one person, a protégé, described the work environment as being unsupportive.

TABLE 1.

Summary of the Median Values for Likert-Type Survey Questions

Question	Mentor Median	Label for Mentor Median Response	Protégé Median	Label for Protégé Median Response
Workplace LGBT supportive	5	Very supportive	4	Supportive
Amount of contact	4	Every 2–3 months	2	Once per year
Satisfaction with amount of contact	4	Satisfied	2	Dissatisfied
Satisfaction with type of contact	4	Satisfied	3	Neutral
Goals met for relationship	3	Goals met	2	Goals met below expectations

Note. LGBT=lesbian, gay, bisexual, or transgender.

Status/Length of the Mentor Relationship

Of the mentors and protégés who returned the surveys, only five stated they were still in a mentor relationship. The most typical response for length of the relationship was very short, consisting of one to a few e-mail contacts. There were several exceptions to this, however. Four mentors and one protégé report having continuing mentoring relationships.

Amount of Contact With Mentor or Protégé

amount of contact a protégé had with his or her mentor significantly affected his or her satisfaction with the mentoring relationship. In this case, as is reported in Table 1, protégés stated that they were generally dissatisfied with the amount of contact (between none at all and once per year) that they received. The Somer′s d test, shown in Table 2, indicates a strong connection between little to no contact with mentors and dissatisfaction with the mentoring relationship. The majority of protégés stated that their only contact with their mentors was the initial e-mail.

TABLE 2.

Summary of Somer’s d Tests

Test	Mentors	p-Value Mentors	Protégé	p-Value Protégés
Amount of contact associated with satisfaction with amount of contact	.531	.06	.893	<.0005
E-mail contact associated with satisfaction with type of contact	.290	.414	.231	.631
Phone contact associated with satisfaction with contact	.481	.193	.917	.011
In-person contact associated with satisfaction with contact	.742	.006	1.00	.003

On the other hand, mentors reported feeling satisfied with the amount of contact they had with their protégés, and the reported median level of contact with their protégés was once every 2 to 3 months. In this case, the Somer′s d test, as reported in Table 2, approached significance, but was not significant. Although it seemed as though both mentors and protégés were more satisfied with the relationship if there was more frequent contact, or were dissatisfied by little contact, mentors appeared more likely to be content with any amount of contact between their protégés and themselves.

Type of Contact With Mentor or Protégé

The results of this set of questions were similar to the results regarding the amount of contact between mentors and protégés. Protégés generally felt neutral (see Table 1) about the type of contact they received, with all but 3 of the 13 protégés having only e-mail contacts. Mentors generally felt satisfied (see Table 1) with the type of contact they had with their protégés, but in this case 12 of the 19 mentors contacted their protégés by phone, and 9 mentors saw their protégés in person. For protégés, as is reported in Table 2, the Somers d tests of the relationship between each type of contact and level of satisfaction showed a significant relationship between phone or in-person contact and level of satisfaction. The only significant relationship identified by the Somer′s d tests (see Table 2) for mentors was in-person contact. Therefore, for both mentors and protégés it appears that in-person contact produced higher levels of satisfaction with the relationship than using e-mail alone to communicate.

Responses to Open-Ended Questions

The open-ended questions focused on in this article included those asking respondents to identify the goals they hoped to achieve by participating in the mentoring relationship and the strengths and weaknesses of the project. The results of these questions are reported in Table 3.

TABLE 3.

Summary of Themes From Open-Ended Questions Asked on the Survey

Question	Mentors	n	Protégés	n
Goals	General support	11	Career development	14
	Give back/share experiences	8	LGBT mentoring	6
	Assist LGBT community	6	New relationship	4
	No goal in mind	1	General mentoring	2
Successes of program	Help protégés meet goals	4	Available mentor	7
	Share own experience	4	No success	3
	Service to LGBT community	3
	No success/too soon	3
	Participation in program	2
	Meeting protégé in person	1
Suggested changes	No changes	7	Ensure mentor participation	4
	Increased contact with program	4	More mentors	3
	Increased structure	3	Increased contact with program	1
	More protégés	2	No changes	1
	Increased matching	1

Note. LGBT=lesbian, gay, bisexual, or transgender.

Goals of the relationship; whether goals were met

Mentors and protégés reported different sets of goals and different levels of satisfaction regarding whether goals were met in the mentoring relationship. For protégés there were 26 responses in total. As Table 3 shows, the results included, in decreasing frequency, wanting career development support, wanting LGBT-specific mentoring, wanting to gain new relationships, and looking for general mentoring. A typical response to this question was “To get input on career development issues, some having to do with LGBTQ (lesbian, gay, bisexual, transgender, and questioning) issues.” As indicated in Table 1, protégés felt their goals for the relationships were met below expectations. The majority of protégés were seeking instrumental types of support and, in general, felt they did not get enough support from their mentors.

The 26 mentor goals included, in decreasing frequency (see Table 3), wanting to give general support with nothing more specific, wanting to give back and share their experiences, wanting to assist LGBT individuals and the LGBT community, and having no goals in mind. An example of a typical response to this question was “To ‘pass it on’ and be supportive of young LGBT faculty career development.” As indicated in Table 1, mentors felt their goals for the relationships were met. In this instance, there was a mixture of instrumental and psychosocial support, with the most typical response being one of “general” support with no particular goal.

Successes and suggested changes to program

Mentors and protégés were asked to list the most important or successful parts of the program and recommend changes to the program. For protégés, as summarized in Table 3, the majority of the 10 responses reported that the greatest success was the availability of support from a mentor, whereas several other responses reported that there were no successes in the program. A typical response to this question was “the sense that support was available if I needed it.” The 9 responses to suggested changes included requests for having the program ensure mentor participation, increasing the number of mentors, increasing contact with the program coordinator, and finally that no changes were necessary.

As summarized in Table 3, mentors gave 17 responses regarding the success of the program. These included helping protégés meet goals, sharing their own experiences, service to the LGBT community, no success or too soon to tell, overall participation in program, and in-person meeting with the protégé. In terms of suggested changes, the 17 responses included (see Table 3) making no changes to the program, increasing the contact with the program coordinator, increasing the structure of the mentor relationship, having more protégés, and having more focus on matching characteristics.

Exceptional Cases

Three cases did not fit into the typical responses given on the survey. One of these cases was a mentor of two protégés. The other two cases were protégés. Because this is not a matched sample, we cannot tell whether there was any relationship among these three individual cases. In all three cases, the individuals reported high levels of satisfaction with the mentor relationships. In the first case, the mentor maintained relationships with both assigned protégés, provided a wide array of instrumental and psychosocial supports over the years, and had contacts with the protégés by e-mail, phone, and in person. The two protégés expressing satisfaction with the relationship also reported having contact by e-mail, phone, and in person and reported receiving a variety of supports over time. For those three matches, then, this project was a successful way to make a lasting and beneficial connection that they reported went above and beyond the expectations of the mentor and protégés.

Discussion

The results of this survey highlight the need and positive expectations for a targeted mentoring program, while pointing out the disparities of experience for mentors and protégés in this project. Overall, mentors reported satisfaction with the project as it was designed. Many mentors volunteered for this project based on their desire to give back to the LGBT community. Mentors, in general, did not have particular goals in mind when beginning the relationship and looked to their protégés to determine the nature of the relationship. Because the mentors did not report having any particular goals of their own in mind, they were generally satisfied with the outcome of the relationships, which, by their own report, typically consisted of one to several contacts, usually by e-mail.

Protégés, on the other hand, were looking for more specific types of instrumental and psychosocial support, and many signed up for this project specifically looking for LGBT–oriented mentoring. This result matches well with the previous research on LGBT graduate students, because the students in those studied also sought LGBT oriented mentoring (Barrett, 2001 Greer, 2005; Lark & Croteau, 1998). With these goals in mind, protégés typically reported dissatisfaction with the relationship and were unhappy with the one to several contacts by e-mail also reported by them. Protégés were seeking more mentor initiation of the relationship but may also have felt some amount of responsibility for not having a closer relationship with their assigned mentor. For example, one protégé stated, “I recognize that the success of the mentor relationship is partly the responsibility of the mentor and the mentee; I didn’t reach out to my mentor, and nor did she to me. Maybe it was that I didn’t know how to use her.”

Because of the nature and scope of this study, we cannot determine what particular characteristics made a few matches work well and the majority of matches not as well, particularly from the viewpoint of the protégés. The research and writing done on negative mentoring relationships, however, does appear to overlap well with some of this study′s findings. Researchers discussing characteristics of the mentor or protégé that may lead to negative mentoring relationships include Eby and Allen (2002), who reported that “Poor Dyadic Fit” (p. 467), or a mismatch between mentor and protégé characteristics, was a primary class of negative experiences for protégés in negative mentoring relationships. Eby and Lockwood (2005) found that protégés reported neglect from their mentors, unmet goals, and structural separation from their mentors to be the most common issues in a problematic mentoring relationship. The results of those two studies align with the results from this study, in which protégés reported mentor/protégé mismatch, unmet goals, and neglect as causes for dissatisfaction in the mentor relationship.

The overall structure of this mentoring program may also have affected the outcome of this project. Unlike many of the mentoring programs evaluated in the literature, this program matched people from different institutions. This built-in physical distance and the distance from the protégés′ work environments may have created different expectations for both mentors and protégés. Three of the 21 mentors explicitly stated this. One example of this was a mentor who stated, “In my university, I would be more pressured to help the colleague meet tenure/PhD requirements. There were no explicit expectations here.”

Relationship building can be a challenge even when both people work at or attend the same institution, but relationship building with the possibly of no previous introductions seemed to be a challenge for both mentors and protégés in this study. Without the foundation of a strong relationship, protégés may have felt unable to ask for what they wanted from the relationship, and mentors may not have been given enough feedback from protégés to make any changes to the relationship. Finally, it is important to note that all mentors were volunteers who did not receive compensation or release time to participate in this program. As such, it may have been difficult for mentors to actively participate in a meaningful way given other time commitments and constraints. Prior research has discussed the added stressor mentors from underrepresented populations often face when they are often asked to mentor several people, and because they may face their own workplace struggles based on their minority status (e.g., Dixon-Reeves, 2003; Kanter, 1977; Kram, 1985; Parker & Kram, 1993).

In sum, there were differences in the experiences of mentors and protégés based primarily on their perceived needs and reasons for participating, as well as in the methods and frequency with which they were in contact, the geographic separation of mentor and protégé, and finally the challenges with completing volunteer work in addition to typical time commitments. These experiences may be common in mentoring programs, particularly when the program is less formally structured, that is, although the participants were formally matched, expectations, goals, and processes were defined individually by those involved in the mentoring relationship.

On the other hand, it is important to note the significance, from the perspectives of mentors and protégés, for LGBT-specific, (i.e., targeted) mentoring. This was highlighted in a number of ways. LGBT-specific mentoring goals were common for both mentors and protégés. Many protégés reported having other mentoring relationships, but were actively seeking additional, LGBT-specific mentoring. Finally, many mentors discussed their ability to “give back” to the LGBT community as a strength of the program, and many protégés mentioned that the presence of LGBT faculty was a main strength of the program. These responses make it clear that both mentors and protégés continue to see the need for LGBT-oriented support, and that this is support that they may not be able to receive or provide in their current working environment. These findings, on the part of protégés, replicate previous findings in this area that also highlight the importance of having LGBT-targeted mentoring (Barrett, 2001 Greer, 2005; Lark & Croteau, 1998). These results also align well with findings regarding the potential importance for persons of color to have support from other ethnic or racial minorities in work and academic environments (Brown et al., 1999; Dixon-Reeves, 2003; Ensher & Murphy, 1997; Ortiz-Walters & Gilson, 2005; Simon et al., 2004) and for women to have the support of other women in work environments (Frestedt, 1995; Kram, 1985; Kurtz-Costes et al., 2006; Ortiz-Walters & Gilson, 2005).

Implications

Because this is an initial study of a specific mentoring project, the results may not generalize well to other mentoring programs. In addition, our small sample size, particularly for protégés, may increase our sampling error and impact our study results. Although all 115 past and present members were asked to participate in the study, the response rate of 39% (n=43) leaves uncertainty as to the feelings of nonrespondents and may lead to possible response bias. For example, it may be unclear whether the respondents included a majority of dissatisfied protégés and satisfied mentors. However, because all members were asked to participate, drastically altering the sample size and response rate would be difficult without significant incentives for participation. Further, the lack of a comparison group of mentor pairs not matched on the basis of sexual orientation and/or gender expression also is a study limitation. A comparison group would have allowed a more specific exploration of targeted mentoring regarding the importance of the mentor being an LGBT faculty member.

However, this study, along with other studies of the LGBT population (e.g., Croteau, 1996; Lin, 2001; Lyons et al., 2005; Morrow et al., 1996; Ragins et al., 2003) show the continued need for support for graduate students, new faculty, and working professionals who identify as LGBT. One other indication of this need is the existence of similar mentoring programs in other disciplines as mentioned previously (ACPA SCLGBTA, n.d.; APAGS, n.d.; Barrett, et al., 2002).

However, it is apparent from this study that designing a successful mentoring program can be a challenge. Although the individuals participating in this project were motivated enough to voluntarily sign up to be mentors and protégés, it was not enough, in general, to lead to successful mentoring relationships. The additional commitment of mentors and protégés to provide and receive support as members of an underrepresented community also did not, for the most part, lead to strong mentoring relationships. The factors discussed here should be explored in greater detail in future studies. These issues may already be affecting other existing mentoring programs. For example, the American Psychological Association, as of early 2007, reports on their Web site that they are in the process of changing their mentoring program (APAGS, n.d..

A number of possible solutions have been discussed by these researchers as well as by the CSOGE. One solution may be to continue the program but set expectations about time commitments and goals explicit at the beginning of the mentoring relationship. It may also help to have potential mentors and protégés fill out a more extensive application to learn more about what each person expects to give or receive from these relationships. Making goals and expectations formal and gathering more information for matching purposes may reduce the amount of perceived mismatch between mentors and protégés.

Another possibility is to alter the program, making it more of a mentoring network rather than a mentoring relationship. Individuals seeking advice and support for LGBT-related issues could contact the network and connect in a more informal manner with others who have had similar experiences or have knowledge regarding the particular issue. Although this would not lead to the development of longer-term relationships, it might provide support for graduate students and new faculty around LGBT issues. Because many of the protégés noted that the biggest strength of the program was the knowledge that there were other LGBT or LGBT-affirming individuals from whom to get support, perhaps a mentoring network may be a good alternative.

The importance of continuing to offer support for LGBT doctoral students and new faculty remains clear. Although almost all of those surveyed in this study felt that they worked within LGBT supportive environments, many of the protégés and mentors stressed the need for LGBT targeted support. Many other professional organizations have also recognized the need for LGBT related support and mentoring for graduate students and new faculty. This support or mentoring program, however, needs to be tailored for the needs of the individuals participating, because mentor/protégé matching, as in this program, may not be the most effective mechanism for providing this support.