Together these reasons brought us to focus on the following objectives in conducting the research for this book:

In the 1990s, many organizations adopted formal mentoring programs to increase the success of women and minorities. While some of these programs did much to advance careers, these formal mentoring programs did not fully replicate the spontaneous relationships that were so beneficial to protégés. Georgia Chao and her colleagues found, for example, that mentors in formal mentoring relationships received fewer benefits than did those in informal or spontaneous relationships.³ Issues of mentor and protégé selection, training, and follow-up activities became very important in making sure formal relationships worked. Therefore, further research investigated what types of protégés might be more likely to have mentors, and whether issues of race and gender, or the similarity of race and gender for protégés and mentors, made a difference in relationship quality and the ensuing benefits.⁴ Belle Rose Ragins, with her very comprehensive body of research, has made significant contributions to the field of mentoring by examining the impact of diversity and power; she inspired much of our thinking about issues of race, gender, and mentor functions in relation to mentoring.⁵

Scholars Lillian Eby, Stacey Blake-Beard, and Herminia Ibarra, as well as our own work published in 1999, highlight the proliferation of different types of mentoring relationships in organizations. ⁶ Other researchers investigating these different forms of mentoring relationships include Gayle Baugh and Terri Scandura, in 1999; Monica Higgins, 2000; David Thomas and Monica Higgins, 1996; Higgins and Kram, in an Academy of Management Review article from 2001, show the importance of shared networks of relationships in career success.⁷ With the advent of increased communication technology in organizations, some mentoring relationships are taking new forms, such as those outlined by Ensher and colleagues in a recent article on e-mentoring.⁸ Furthermore, we see new research on the link between managing and mentoring employees as many firms delineate coaching, mentoring, or both as a managerial responsibility.⁹ Taken together, these ideas informed our inquiry and analysis of our interviews.

Although it has been comprehensive, mentoring research has three main shortcomings. First, HR practitioners or managers have not been privy to many of the academic research findings. For example, Joyce Russell and colleagues, in a special mentoring issue of the Journal of Vocational Behavior, counted more than 500 articles on mentoring, most of which came from academic journals.¹⁰ To date, many of the empirical studies on mentoring have been gathered in books, but only a few new books bridge academic and practitioner material. Second, even though it is informative, much of the academic literature from the past 10 years relies on questionnaires and surveys to describe the nature and benefits of mentoring relationships. However, it is difficult for quantitative survey research to capture the richness of mentoring relationships. The third shortcoming of academic research is that it tends to focus almost exclusively on one side of the relationship at a time. The focus is on the protégés’ perspective most often, the mentor’s perspective is rarely gathered (with the exception of Allen, Poteet, and Burroughs in 1997), and never are both perspectives considered simultaneously.¹¹ Thus, the purpose of this current project was to address some of the gaps in the research knowledge by offering an integration of academic findings and real-life examples from datarich, in-depth interview data from highly successful and well-known leaders, and give voice to both the mentors and the protégés, thus illuminating both perspectives.

In developing our interview protocol, we kept in mind a number of theoretical lenses that would come into play during data analysis. For example, using Kram’s stage theory of attraction, cultivation, redefinition/separation, we wanted to explore in depth what predictors and processes were critical in each phase and how these led to the next.¹² We used a modified version of that model we published previously.¹³ We were interested in ideas of how individuals came to their roles of mentor and protégés through the lens of role-making theory.¹⁴ We were interested to what extent some of the mentoring relationships followed the precepts of social exchange theory, or what additional insight into the role of perceived similarity in values and attitudes in the mentoring exchange could be gained through knowledge of leader-member exchange theory of leadership.¹⁵ Warren Bennis’s work on leadership, particularly the emphasis he placed on crucible moments, also informed our thinking about defining moments for mentors and protégés.¹⁶ We were also eager to consider how network constellations and alternative forms of mentoring suggested by Monica Higgins and Kathy Kram might play a role in today’s mentoring, and how the mentor’s perspective of the benefits of a relationship affected the types of mentoring that occurred.¹⁷ And finally, we reviewed literature on important aspects of close relationship development as a method for understanding developing mentoring relationships.

For our 50 interviewees, 17 were men (34 percent), 14 were people of color (29 percent), and the average age was around 48 years, ranging from approximately age 30 to 70. Individuals from three industries were primarily targeted for inclusion in the study. Many people would agree that the U.S. economic impact on the rest of the world, for better or worse, is strongly determined by our government, our television and movies, and our technological products. These industries impact how people govern, live their lives, work, and play, and they have an undeniable American imprimatur. We interviewed approximately equal numbers of individuals from these three broad industries. We felt that exploring how mentoring has been important to the prominent individuals in this study would be useful to workers in a variety of fields, as careers in nearly all industries are becoming increasingly more boundaryless. These three industries also all support virtual organizational structures or boundaryless careers. Virtual organizational structures are those in which individuals are linked together through technology by their product and service rather than by their organizational membership or geographical location. An example of this might be an independent political consultant who acts as a broker of services and relies on a cadre of colleagues with varying areas of expertise to respond to client needs. As authors such as Philip Mirvis, Tim Hall, and Michael Arthur have found, the boundaryless career is characterized by portable skills, knowledge, and abilities across multiple firms, personal identification with meaningful work, on-the-job action learning, the development of multiple peer-learning relationships, and individual responsibility for career management.¹⁹ Career pundits predict that the wave of the future for many other industries and their workers is in boundaryless careers and organizational structures; therefore much can be learned from these three benchmark, boundaryless arenas. If new forms of mentoring can thrive in industries such as these, where there is enormous competition and amorphous career paths, then these new forms of mentoring should work well in other industries facing similar challenges.

An initial set of questions was developed to ask generally about various aspects of interviewees’ current mentoring relationships. Once initial themes about tests and challenges and specific mentor functions were mentioned in these preliminary interviews, we modified the interview protocol to further investigate these aspects. The final interview protocol included open-ended questions focusing on career history, mentoring relationship development, defining moments, and ideas about protégés and mentor characteristics; it is shown in Exhibit B.1. When the interview focused more on a protégé’s relationships with his or her mentor, the protocol was changed slightly. Additional information about the participants was collected from their biographies, obtained by their staffs, and through information from newspapers, trade journals, and Internet resources.

Our data analysis followed the following stages:

Stage 1: Once all the interviews were completed, the preliminary data analysis efforts involved reading the transcript of each interview at least two times in its entirety.

Stage 2: Transcripts were reviewed to generate underlying themes. An initial scan for themes used a grounded theory approach, where interesting themes emerged from the data. Many of these themes emerged not from a direct question but from an expanded answer to another question or to a seemingly unrelated question. For example, in our initial stages of analysis, several initial overall themes emerged such as:

Stage 3: In the next stage of analysis, we went beyond answering “what is interesting, surprising, or similar” in looking for patterns, and we used a more stringent approach. Here, we applied various theoretical lenses as described previously from the fields of mentoring, leadership, and social psychology to further analyze the data.

Stage 4: We used NVivo to sort data into thematic categories and then began an iterative process of refining the major themes. The models we used in guiding our research followed some of the existing models of mentoring relationship development we described in the theoretical background section, and specifically focused on:

Once we had the data sorted into major theme documents, we were able to finalize our major categories of frequently occurring ideas, which then became the basis for the book chapters.