When Kathy Kram carried out her seminal research into mentoring at the beginning of the 1980s, she focused on a small number (22) of informal mentoring relationships, taking place within a specific cultural context – educated North Americans. She identified a number of themes that recurred in these relationships and, as she herself has readily admitted, struggled to find a word that categorised them all. Eventually, she settled on the term “function” as a term that encompassed a mixture of behaviours, personal attributes or qualities, and roles. That excellent piece of research has spawned a vast array of mentoring research – but much of this research is of marginal value, because it is based on a fatal misunderstanding of the nature of “functions”.
Various scales and instruments have been designed to measure how many functions a relationship exhibits, on the assumption that the more functions that occur, the more effective the relationship will be. But where is the evidence for this assumption? A mentor may exhibit all the functions at a low level, or one or two at a very high level. One of my mentors, to whom I am deeply grateful, exhibited very few of the functions. We never became friends, for example. But he spent an hour with me explaining the political realities of being a non-American working for McGraw-Hill in the early 1980s – and was directly responsible for launching my career (outside the company).
A parallel assumption is that successful mentoring is mostly about what the mentor does. Again, this is manifestly untrue. The mentoring relationship is a two way dynamic, so the “functions” of a mentee should be equally relevant. The efficacy of the relationship is in large part about the interaction between mentor and mentee. When I carried out my own doctoral studies into developmental mentoring a decade and more ago, there was a paucity of studies that explored this dynamic – and there still is.
Moreover, if we look at the functions closely, we see that they are all input factors. Research on efficacy typically examines outputs. Efficacy in mentoring relates to the defined purpose of a relationship (although this may be emergent rather than present from the beginning) and in a programme context, the degree of achievement of organizational or societal goals, such as improving graduate retention, creating greater sexual and racial equality at higher levels of management, or reducing the rate of recidivism of ex-offenders. There are studies that link mentor functions with career outcomes (for example, Ragins and Cotton, 1999), but these are quite limited in scope.
The original Kram functions were career focused (sponsorship, exposure and visibility, coaching, challenging assignments, and protection to the protégé) and psychosocial (role modeling, acceptance and confirmation, counselling, and friendship). Various minor additions have been proposed and incorporated into instruments over the years.
In Northern Europe, a very different model of mentoring emerged to that described by Kram, much more consistent with values of encouraging people to take responsibility for their own personal growth. The concept of sponsorship, in particular, is reflected in this model. These two approaches, which have been clearly established as two different constructs, are related to different aspects of the story of Athena in her dialogues with Odysseus and his son Telemachus. In one aspect, Athena is the protectress, intervening directly to assist – for example, by covering Odysseus with her aegis, or cloak of invisibility (shades of Harry Potter!) – hence the term protégé (one, who is protected). In the other aspect, Athena is the goddess of wisdom, using her wisdom to help Odysseus and Telemachus develop their own wisdom. From this tradition comes the term mentee (someone, who learns by being caused to think).
If we think about the function of mentoring, rather than about mentor functions, the literature suggests that it is primarily about career assistance (in sponsorship mentoring) or about self-learning and self-growth or wisdom (in developmental mentoring). Of course, a sponsorship mentoring approach may in some circumstances result in deep personal learning (for example, through cycles of role modelling and rejection) and developmental mentoring typically has career impacts as the mentee gains greater clarity about both their inner context (who they are and what they want) and their outer context (the opportunities in the world around them).
So what’s the bottom line for future research in mentoring? Should we consign mentoring functions to an interesting footnote in the development of our understanding of the mentoring phenomenon? From a global perspective this may be easier outside the US than within, not least because in Europe and other continents the debate has moved on to focus more on the competencies of mentors and mentees. There is potentially some overlap between functions and competencies, but competencies take us beyond observed behaviours to what behaviours have greatest impact works and how. So some of the important questions for my current studies are:
- What is about how effective mentors and mentees listen, craft questions, and reflect together and apart, that leads to good outcomes?
- What is different about the mentoring relationship and the mentoring dialogue compared to other conversations in the workplace?
Perhaps it’s not time yet to bury mentoring functions, but it probably is time to move on.
 Ragins, BR and Cotton, JL, 1999, Mentor Functions and Outcomes: A Comparison of Men and Women in Formal and Informal Mentoring Relationships, Journal of Applied Psychology Vol. 84, No. 4, 529-550)