Journal of Child & Adolescent Mental Health

ISSN: 1728-0583 (Print) 1728-0591 (Online) Journal homepage: http://www.tandfonline.com/loi/rcmh20

Characteristics of mentoring relationships Sarah Shelmerdine & Johann Louw To cite this article: Sarah Shelmerdine & Johann Louw (2008) Characteristics of mentoring relationships, Journal of Child & Adolescent Mental Health, 20:1, 21-32 To link to this article: http://dx.doi.org/10.2989/JCAMH.2008.20.1.5.490

Published online: 12 Nov 2009.

Submit your article to this journal

Article views: 216

View related articles

Citing articles: 7 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=rcmh20 Download by: [University of Bath]

Date: 05 June 2016, At: 18:52

Journal of Child and Adolescent Mental Health 2008, 20(1): 21–32 Printed in South Africa — All rights reserved

Copyright © NISC Pty Ltd

JOURNAL OF CHILD AND ADOLESCENT MENTAL HEALTH EISSN 1728–0591 DOI: 10.2989/JCAMH.2008.20.1.5.490

Characteristics of mentoring relationships Sarah Shelmerdine and Johann Louw*

Downloaded by [University of Bath] at 18:52 05 June 2016

Department of Psychology, University of Cape Town, Rondebosch 7701, South Africa * Corresponding author, e-mail: [email protected] Objective: Mentoring programmes offer an increasingly popular solution to the problem of youth ‘at-risk’. While positive outcomes of these programmes are widely documented, there has been little research into the operant processes. The aim of this study was to provide in-depth qualitative information about relationship processes in the Big Brothers Big Sisters of South Africa programme. Method: The sample consisted of mentors and youth in eight relationships. Semi-structured interviews were conducted with the participants and were analysed making use of the narrative method. Results: Relationships were classified into two groups, namely those constructed in terms of their purpose to help the mentees, and those constructed as friendship for their own sakes. These relationships displayed the qualities previously associated with negative and positive relationships, respectively. The results of the study further extend previous findings, revealing the interaction of these overt qualities with underlying constructions of relationships. Conclusion: The study highlights the influence of the programme and broader social context on the mentoring relationships. It indicates the importance of the partners’ ability to move beyond the constraints of contextually available narratives of their relationships, and to establish more personal connections, based on the value of the relationship itself, rather than on the programme’s ultimate goals.

Introduction Mentoring programmes are becoming an increasingly popular form of intervention for dealing with the problem of youth considered at risk of antisocial outcomes (Barron-McKeagney, Woody and D’Souza 2001; DuBois et al. 2002; Grossman and Tierney 1998; Tierney, Grossman and Resch 1995). These studies have found, amongst other improvements, that programme youth are less likely than controls to initiate drug or alcohol use, to hit someone or to play truant from school. School marks improve, youth feel more scholastically competent, and family and peer relationships improve in terms of trust, emotional support and intimacy of communication (Tierney et al. 1995). A number of studies have attempted to discern the characteristics of mentorships strongly associated with this positive change (DuBois and Neville 1997; Herrera 1999; Herrera, Sipe and McClanahan 2000; Rhodes et al. 2002). These studies have identified such qualities as closeness, emotional and instrumental support, empathy, trust, mutual respect and mentors’ sensitivity and responsiveness to youth’s needs and circumstances as key agents of change. Mentees’ positive perceptions of their relationships and of their mentors as friends rather than teachers have also been identified as among the strongest predictors of healthy development. Conversely, adult over-control and short duration of relationships have been found to be associated with negative outcomes (Rhodes 2002a). Quality of relationships with adults is considered to be of fundamental importance to any developmental setting (National Research Council and Institute of Medicine 2002). Research has therefore attempted to determine the characteristics of strong mentorships (Grossman and Rhodes 2000; Herrera et al. 2000; Phillip and Hendry 1996), identifying mentees’ agency in decision-making about the relationship, similarity in mentor and youth interests, engagement in social as well as academic

Downloaded by [University of Bath] at 18:52 05 June 2016

22

Shelmerdine and Louw

activities and pre-match and post-match training and support. Other studies (Liang et al. 2002; Phillip and Hendry 1996; 2000) have emphasised the reciprocal nature of successful mentorships and the importance of mutual exchange. While the above studies identify key characteristics of strong relationships, very few in-depth qualitative studies of relationship processes have been conducted (Beam, Chen and Greenberger 2002; Morrow and Styles 1995; Styles and Morrow 1996). Beam et al. (2002) conducted one such study, concluding that strong mentorships are parent-like in their ability to provide the kind of support that peers usually cannot provide, and peer-like in their non-judgmental nature and provision of non-punitive fun. Styles and Morrow noted that interaction styles, rather than particular relationship activities, distinguish strong mentorships. Morrow and Styles (1995) provide a detailed qualitative study of the interactive processes within mentorships in the Big Brothers Big Sisters of America programme. These authors classified the relationships they studied as either ‘developmental’ or ‘prescriptive’; ‘developmental’ relationships being the more positive type. Within these relationships, mentors provided youth with a safe space within which they could focus on a broad range of developmental tasks. These mentors were flexible in their responses to their mentees and adjusted their expectations to fit their realities, circumstances and needs, actively incorporating them into decision-making concerning the relationships. ‘Prescriptive’ relationships, on the other hand, were those in which mentors were changedriven, with narrowly defined goals, and engaged in decision-making about their relationships in a controlling manner. Racial or ethnic difference between mentors and mentees, and its possible effects on the quality and effectiveness of relationships, is an area of substantial concern within the mentoring literature. It has been suggested that white mentors are unable to empathise with minority youth and are vulnerable to guilt, attempting to ‘save’ their mentees rather than focusing on the development of a trusting and supportive friendship. Research has however shown (Herrrera et al. 2000; Jucovy 2002; Morrow and Styles 1995) that there are no significant differences in either the qualities or outcomes of cross-race and same-race matches. The evidence suggests instead that it is the personal characteristics of mentors in these relationships and their sensitivity to and awareness of the nuances of cultural meanings that are important. In fact, it is possible that socio-economic differences cause greater problems for relationships than racial ones (Rhodes 2002b). Rhodes, however, cautions that these findings are based on small samples in programmes with a high degree of match supervision and support. The present study The present study explored the relationships between mentors and mentees within the Big Brothers Big Sisters (BBBS) of South Africa programme. The programme was established in South Africa in September 2000 and is modelled on Big Brothers Big Sisters of America, commonly considered the model of best practice in the mentoring field. The programme is preventative in nature and aimed at children between the ages of six and 18 years considered at risk of a range of antisocial outcomes. Children are identified as being at risk by teachers, parents, social service providers and court or diversion officers, and are referred to the programme. Eligibility criteria include living in a single-parent home with no significant involvement of the absent parent; living in a sibling-headed household; living in a home characterised by violence, sexual abuse, substance abuse or other familial disturbances; and evidence of antisocial behaviour. Volunteer mentors are screened and required to complete a two-day training course before being matched with a child, all of which happens according to formal guidelines. Mentors meet with their mentees on a one-to-one basis for at least one hour per week, for a minimum of a year. Relationships receive intensive supervision from the programme and both parties are contacted regularly by caseworkers for the duration of the relationship (see Grossman and Tierney (1998) for a brief description of the American programme). The BBBS of South Africa programme is an ideal one in which to explore the quality of mentoring relationships, since its primary purpose is to provide adult friendships to youth considered to be at risk of antisocial outcomes. Furthermore, since it is in its early phases of implementation in South Africa, it presents an ideal opportunity to reflect on the nature of the relationships formed

Downloaded by [University of Bath] at 18:52 05 June 2016

Journal of Child and Adolescent Mental Health 2008, 20(1): 21–32

23

in a context quite different from the American one. Since Morrow and Styles (1995) also used this programme to explore mentoring relationships, we believe useful comparisons could be made between what is essentially the same programme delivered in two different contexts. Our study therefore utilised a small number of in-depth interviews to move toward an understanding of the qualitative processes within the relationships, with a particular focus on the meanings attributed to them. The following questions were addressed: • How do the mentoring partners conceptualise themselves and each other within the context of their relationships? What positions do they assume relative to each other? What meanings do they attribute to their relationships? • How do the above constructions interact with more concrete relationship qualities? For instance, what is the relationship between these constructions and the nature and extent of communication and empathy between the partners? Are mentors sensitive and responsive to their mentees’ needs and circumstances? Are mentees able to enlist their mentors’ help, and is the help provided acceptable to them? To what extent are reciprocity and equality elements of the relationships? • What is the contribution of socio-economic factors, including racial differences and similarities, and the partners’ perceptions of these, to the relationships? Method Participants The sample for this study consisted of eight mentorships in the BBBS of South Africa programme. These eight relationships constituted the total number of mentorships in the programme that had been established for approximately one year by the time the study was conducted. All the mentors in this study were female, relatively well off, English-speaking university students. Seven of the eight were in their early twenties and one in her forties; three were white and five black. Seven of the eight mentees were female and one male. All were black, Afrikaans-speaking and living in an impoverished urban area in Cape Town. Mentees ranged in age between nine and 16 years. While the sample was small, the usefulness of evaluation findings during the programme’s formative stages created the need for this information to be produced sooner rather than later, when more relationships might have been available for study. Procedure All participants were interviewed about their relationships. Mentees were approached with the approval of the BBBS of South Africa programme staff, and participated on a voluntary basis. Once mentees agreed to participate in the interviews, the first author met with them in small groups, in which participants and the researcher engaged in creative activities (such as drawing, painting and clay-modelling). This was done to establish familiarity with the interviewer, and to establish the basic parameters for the interviews. Mentees were able to withdraw from the study at any stage, but none did so. Individual interviews with the youth were arranged, and were conducted in a variety of settings, at their request. All the mentors in the study were students at the University of Cape Town. Interviews with them were thus conducted in informal environments on the university campus, with the exception of two interviews that were conducted at the mentors’ homes. Interviews were semi-structured and qualitative, drawing on Morrow and Styles’s prior research (1995). In the interviews we explored the following: activities during shared time; content and nature of communication; participants’ perceptions of each other and of their relationships; processes in the relationships such as negotiating roles; mentors’ goals with regard to the relationship and mentee; the meaning and value of the relationships for participants; and contributions of socio-economic and racial differences to the relationships. It was considered important, however, in light of the aim to produce information sensitive to the local context, that participants also be allowed the opportunity to name their own experience (Scheurich 1997). Participants were thus initially encouraged to speak about their relationships as they saw them, and additional lines of inquiry were developed in light of their responses (Hutchby and Wooffitt 1998). Wherever possible, stories about actual events and

24

Shelmerdine and Louw

Downloaded by [University of Bath] at 18:52 05 June 2016

recounted conversations were requested, as we felt that these were most likely to elicit rich contextual information and provide insight into the underlying values and constructions from which the participants composed their experiences of the relationships (PANOS 1999). Analysis Complete interviews were audio-recorded but not transcribed. It was possible to listen to the tape of each interview several times in order to discover what sections were important enough to transcribe, and which needed only to be summarised. We followed the generic qualitative procedure (Patton 1990) of identifying salient themes and recurring ideas, organising and categorising them, and drawing connections between them. It was thus possible to engage with the material in an organised and systematic fashion, without transcribing the full interviews. Each relationship was analysed as a unit and the partners’ accounts interpreted with reference to each other. Once the relationships had been analysed individually in this manner, themes relevant to the broader emerging picture were drawn out and, where necessary, pursued through further similar inspection of the data. The analysis therefore attempted to unearth recurrent patterns across the narratives, while maintaining a sense of the coherence of individual relationships. The accounts of the relationships were regarded as narratives providing insight into the lived experiences of the relationships and the meaning made of them. According to Dyan and Katz (1992), story forms carry latent messages that echo the manifest messages of an event. Narrative forms that emerged in the data were thus examined to provide insight into participants’ experiences of their relationships. In this regard, participants’ accounts were analysed with particular attention to the use of metaphor, plot sequences, function words, choice of words, cultural associations carried by words, and so on. An approach that looks beyond the overt content of what people say, through analysis of the qualities of narratives and conversational structures, can betray the ‘deeper’ meanings in people’s accounts, beyond the consciously intended ones (Kress 1990). These narrative and grammatical structures, and the ‘little words’, as Billig (1999) would call them, show how the content words, or pieces of information in a narrative, relate to each other in the overall communication (Gee 1999). The analysis of narratives entails the researcher’s construction of the experiences of others and the imposition of her own narrative on the data. Typologies used were created by the researcher, and were grounded in the spoken data, but they were not necessarily used explicitly by the participants (Patton 1990). In addition, participants’ own accounts of their relationships cannot be seen as synchronous with the actual events described or as exact reflections of reality. Rather, they are particular versions presented for a particular audience at particular times. Results and discussion This study has produced findings consistent with those of previous research into mentorships and, through use of the narrative method and the very small sample, extended these in a number of ways. Analysis of the accounts of the relationships in this study revealed threads of commonality, allowing broad classification of the relationships into two categories. The first group consists of five relationships, classified as those dominated by the narrative of ‘help’. The second group is comprised of three relationships, constructed primarily in terms of a narrative of ‘friendship for its own sake’. The relationships in these two categories display overall the qualities associated by previous research with negative and positive relationships, respectively. (The authors need to add a note of caution here in that they use the term ‘friendship for its own sake’ merely as an identifier of the sort of narrative that emerged from these interviews; they are not implying that these are friendships in the sense of equal relationships. The term tries to capture a quality that was expressed in the interviews, rather than stating that mentors and mentees were ‘friends’.) Mentors in those relationships dominated by a narrative of help attempt to control both their relationships and their mentees, and to ‘pass down’ advice, models of behavior, and so on, instead of affirming their mentees’ own knowledge. Mentors are agents of the relationships, and mentees passive recipients of their help. They are further ‘change driven’, with narrowly defined goals,

Downloaded by [University of Bath] at 18:52 05 June 2016

Journal of Child and Adolescent Mental Health 2008, 20(1): 21–32

25

imposing their own values on their mentees and thereby undermining their cultural and social identities. They consequently pressure their mentees to disclose sensitive information and are prescriptive about the areas in which they offer help. Mentees in turn appear to dodge conversations about problem areas, with the result that they are unable to ask their mentors for help when they need it. An unidirectional pattern of interaction, which permeates all aspects of the relationships and is detrimental to communication between the relationship partners, is thereby established, Those mentors in relationships dominated by narratives of friendship for its own sake are, in contrast, sensitive and responsive to their mentees’ needs and circumstances, displaying an ability to empathise with them, and to listen, understand and affirm their mentees’ own knowledge. The focus is on the relationships as mutually valuable and enjoyable interactions in themselves, rather than on their overall purpose to help the mentees. The partners in this category construct their relationships in terms of reciprocal processes in which mentees become more equal participants and both partners are attributed with agency. A trusting, respectful and non-judgmental atmosphere is thereby created, in which mentees are free to discuss sensitive topics, as well as to enlist their mentors’ help, without pressure from their mentors in this regard. Thus far, these findings support those of previous research, identifying similar characteristics of relationships experienced by the mentoring partners as positive or negative. However, this paper makes a further claim; namely that the simple enactment of those relationship qualities identified as characteristic of positive relationships is, on its own, insufficient, and that the underlying meanings that mentors and mentees attribute to their relationships, and their constructions of themselves and each other within them, are at least as important and need to be congruent with the overt characteristics of ‘developmental’ relationships if such relationships are to be successfully established. While some of the relationships in this study were clearly more positively perceived than others, as described above, all the mentors seemed, at least superficially, to be enacting the practices identified by past research as characteristic of positive relationships, and encouraged through the training supplied by the programme. Closer inspection of the narrative structures in the accounts of mentors and mentees, however, reveals a tension in the accounts of those mentors involved in more negatively perceived relationships. One mentor, who constructs her relationship in terms of this narrative, and in which the negative characteristics identified by past research are clearly evident, states, for example: I was aware that ‘Don’t lecture, don’t do this, it’s gonna just put them off’ … I was very aware of it. And we did the workshop, so we went through all that sort of thing. You know, don’t impose what you think is right on them. ‘Do you think it’s right?’ and I ask them, ‘Do you think its right?’ … And they would tell me, ‘No, it’s bad, we wouldn’t do that’. While the above extract describes this mentor’s efforts to comply with advice received in the training course not to ‘impose what (she) think(s) is right’ on her mentee, closer inspection reveals a tension between her endeavors in this regard and a conflicting intention to influence her mentee in keeping with the programme’s overall objective to improve behavioural outcomes. The result is a change in strategy but not intent, and the overriding aim of her relationship practices remains the implementation of change. The emergent patterns in the two basic narrative forms allowed analysis of the underlying construction of the relationships in terms of four primary dimensions within the narratives, namely agency, social distance, knowledge and causal sequencing. The remainder of this paper will discuss the relationships with respect to these dimensions. We will show how construction of the relationships in terms of the two narratives is associated with distinct patterns of interaction between the relationship partners. Narrative form and agency A primary difference between the two narratives concerns the potential for action attributed to the mentors and mentees within the two relationship categories. Differential agency in the narrative of ‘help’ The unidirectionality of the narrative of ‘help’ inhibits the agency of mentees, relegating them to passive positions and limiting their potential for action. The accounts of these relationships are fraught with contradictions, explainable in terms of the unidirectional pattern of communication,

Downloaded by [University of Bath] at 18:52 05 June 2016

26

Shelmerdine and Louw

and indicative of a failure to approach mutual understanding or shared meaning. In the following example, Feroza and her mentee, Adam, refer to the same conversations but detail very divergent perceptions of them. The conversations of interest centre around the topic of antisocial behaviour, in particular gangsterism and drugs. While Feroza claims that these conversations arise spontaneously between them, due to the centrality of these issues in Adam’s reality, Adam reports a lack of interest in such topics, describing them as ‘boring’ and irrelevant to his experience, and expressing resistance to engaging in them. Contradictory on the surface, similar features of their accounts resolve their discrepancies. In both accounts the mentor is portrayed as the initiator of conversation and the mentee the respondent. Adam, in fact, portrays his own contributions as little more than reluctant responses to Feroza’s questions, which are of little relevance to him: F: ‘Are you aware of the consequences of drugs?’ … ‘Did your mother speak to you about these sort of things?’ You know, ‘Are you gonna do it, do you want to do it?’ A: ‘I say, ‘Yes, but I don’t like it,’ and I tell her, ‘I don’t like it.’’ In a similar vein, mentees in these relationships are unable to identify areas in which they require help, another relationship feature identified by previous research (Morrow and Styles 1995) as important to successful mentorships. Instead, mentors in these relationships assume knowledge of their mentees’ needs and of the most appropriate form of help. Not only is the help offered in this way potentially irrelevant and in a sense unacceptable to the mentees, but it results in conflict between the partners in some of these relationships. Adam, for example, talks of the ‘helping’ conversations that occur between him and Feroza: Some make me a little bit angry and stuff … not like angry, angry, but just mad like … I must help my mother to make up the bed, and, and I must … It makes me angry. Mentees’ inability to communicate adequately with regard to their feelings about these relationships further contributes to the lack of shared meaning and to feelings of inadequacy and frustration among mentors. While all the mentees in this study described their relationships as highly valuable and meaningful, mentors in this category were largely unaware of their significance in their mentees’ lives. The feelings this produced were exacerbated by their belief in the necessity of the strength of their relationships to achieve their ultimate goal, namely positive behavioural change. Again this finding is supported by previous research (Morrow and Styles 1995) in which mentees’ failure to provide reassurance and feedback on the relationships’ meaning is identified as the primary source of mentors’ frustrations. Its importance is further emphasised by the fact that this frustration is the chief reason provided by mentors for the premature termination of many relationships (Rhodes 2002a). Agency in the narratives of ‘friendship for its own sake’ The more equal distribution of agency between the relationship partners in these narratives enables the mentees, in contrast to those above, to actively contribute to their relationships. The children in these mentorships initiate conversations, actively participate in defining areas of need and enlisting their mentors’ help, and openly express their appreciation of their relationships. Mentors, in turn, offer help that is responsive and relevant to their mentees’ experiences and derive considerable reward and satisfaction from the relationships. Furthermore, the accounts of these relationships are notably consistent with one another, apparently due to more effective communication and suggestive of shared meaning between the partners. Constructions of these relationships in terms of reciprocal exchange, in which both partners are active, is clearly expressed by one of the mentors in this category: And I don’t feel as if it’s a one-way relationship, like I’m just giving, and, you know, it’s, she’s, you know, reciprocating as well on her own time, at her own pace. Even though it’s not the same way that I am, it’s still in her way. And that, for me, is meaningful. Neither are the mentees in this category restricted in conversation to the role of respondent. In the words of one mentee, Melanie: ‘I first talk, then she talks … We talk to each other … When I come into the car then I talk.’ Not only are these mentees able to communicate about their realities more generally, but, in direct contrast to those who construct their relationships in terms of a narrative of ‘help’, they are able also to share sensitive information — ‘A Big Sister is like a friend …then you can tell her secrets.’

Journal of Child and Adolescent Mental Health 2008, 20(1): 21–32

27

The narrative of ‘help’ is present as a subsidiary thread in these accounts of the relationships. However, the help that the mentors in these relationships are able to offer is a direct response to their understandings of their mentees’ subjective realities and perceived needs, rather than based on their own imposed constructions and interpretations. This help is in turn gratefully accepted by the mentees, as indicated by their expressions of appreciation in their accounts. Attributions of agency thus appear to have considerable import for more tangible relationship elements, such as communication, mentees’ ability to ask for assistance, the extent to which mentors are able to offer support that is responsive to the needs of their mentees and the level of frustration or reward experienced by mentors in these relationships.

Downloaded by [University of Bath] at 18:52 05 June 2016

Social distance in the relationship narratives Socio-economic, rather than racial, differences emerged as key features of these narratives. Despite the small sample size, the issues discussed in this section are relevant to an understanding of how narrative constructions of difference and sameness can enable and constrain relationships and communication. Social distance in the narrative of ‘help’ Opposing definitions of ‘mentor’ and ‘mentee’ within the narrative of ‘help’ engender constructions of the relationship partners in terms of difference. Mentors and mentees within these narratives are constructed as opposite to each other and similar amongst themselves, creating an ‘us–them’ dynamic. This dynamic is evident in the manner in which the partners in these relationships refer to themselves and each other throughout their accounts, frequently indexing individuals with plural pronouns, and attributing collective responsibility for individual actions. Vera, for example, complains of her mentee’s failure to say ‘thank you’, explaining this behaviour in terms of stereotypical assumptions about the mentees as a group — ‘She doesn’t say “thank you” … but obviously they, they haven’t been taught to say “thank you” for things.’ Constructions of ‘mentor’ and ‘mentee’ are thus inextricably linked in these accounts to stereotypical notions about broader socio-economic groups. As Layla, a mentor, put it: I thought … the influences of the relationship would affect her morals and her values. Um, compared to the environment which she’s in … the community in which they live is quite, you can’t say that any of them, they don’t grow up with any morals and ethics … for instance, bullying at school … They wouldn’t walk away from the situation and face the fear, they would just act upon their instincts. Someone’s hurt me, I need to hurt them back. The social groups to which mentor and mentee belong are in fact constructed as distinct from each other in terms of the very ‘morals and ethics’, ‘beliefs and attitudes’ and modes of behavior the mentors hope to influence. The mentees’ need for a mentoring relationship is thus directly related in these accounts to stigmatising constructions of their identities in terms of socio-economic grouping. Within the mentorships in this study, these binary constructions appear to diminish the potential for empathetic understanding. Mentors in these relationships frequently relate what they perceive to be their mentees’ experiences with shock, making use of such adjectives as ‘scary’ and ‘insane’, describing their mentees’ experiences as contrary to their expectations, and thus indicating a diminished potential for empathy between the partners. Distance and similarity within the narratives of ‘friendship for its own sake’ Mentors and mentees in these relationships are, as in the mentorships above, from different socio-economic backgrounds, and difference between the partners is undeniably evident in these accounts. However, the greater equality and reciprocity, and the diminished focus on the help which the mentees are intended to derive, allows for an emphasis on similarity rather than on difference between the relationship partners. Construction of these relationships primarily as friendships, as opposed to interventions, allows their narration in terms of more personal sub-narratives, rather than solely in terms of the programme’s overall goals. These more personal sub-narratives allow the emergence of experiences common to both partners. Naefa and her mentee, Michaela, for example, cite their mutually experienced difficulty and delight in their younger brothers as a major focus of the interactions between

Downloaded by [University of Bath] at 18:52 05 June 2016

28

Shelmerdine and Louw

them. For another pair it is their common enjoyment of Michael Jackson — ‘And we’ve got one commonality, she likes Michael Jackson, and I love Michael Jackson, so we’ve got like a huge thing going there.’ Within these more personal sub-narratives, the mentors further construct their mentees’ needs for their relationships with reference to specific aspects of their unique circumstances, rather than in terms of the stigmatising constructions of group identity discussed above. For example, mentors in this category refer to particular circumstances such as the mentees’ lack of a consistently dependable figure resulting from their unique family circumstances, the lack of ‘quality communication time’ they receive from the adults around them for the same reasons, an overly busy mother, and so on, as the reasons for which they require additional support. The mentors in these relationships very often construct their ability to help their mentees in terms of experiences in their own lives similar to those with regard to which the children require support. In the words of one of the mentors: I don’t want her to think that that I’m all wonderful, because I’m just an ordinary person and I have difficulties just like her, just in different forms … I think she needs encouragement because … I sense that she undermines herself. So I just want to show her, like, I also have a tendency of doing that, so I can relate to that. Where social difference is acknowledged in the accounts of these relationships, it is constructed in ways that either attempt to balance the power between partners or to further mutual understanding. One mentor, for example, made specific attempts to invert their respective positions within imaginary games, initiating games in which she assumed the role of disobedient pupil, for instance, and her mentee of teacher. Alternatively, these mentors used differences between themselves and their mentees as opportunities for furthering mutual understanding. In the words of Naefa, the Muslim mentor of a mentee from a Christian family, I’ll explain things to her and if she doesn’t understand she’ll ask me. And I’ll ask her: ‘What do you normally do on Christmas day, how d’you normally spend your Christmas?’…And it just makes me understand more. Thus, constructions of self and other, and of the purpose of these relationships in terms of difference or similarity, have considerable import for the interactions, and in particular the empathy and the potential for mutual understanding that is established between partners. The role of knowledge in the relationship accounts The centrality of ‘knowledge’ to the participants’ constructions of the mentorships is related to the above binary constructions of self and other, as well as to constructions of the relationships’ purpose and meaning in these terms. Knowledge in the narrative of ‘help’ The importance of knowledge to the workings of these relationships stems from its perceived centrality to the intended process of change. Mentors depict themselves and their ability and purpose within their relationships in terms of their possession and dissemination of knowledge, describing their intentions, for example, in terms of sharing the ‘huge amount I have learned’ or teaching their mentees ‘the correct way of doing things’. The mentees, on the other hand, ‘don’t know what’s best for their future … they just go about and do what they want’. In order to effect change in their mentees, these mentors further require information about the children’s lives and experiences. As Feroza put it when asked what she would have liked to have done with the information she unsuccessfully attempted to extract from her mentee: ‘If he was in the wrong maybe try and correctify, or make him see that he was in the wrong … Just guide him to what is, what I think is the right thing.’ This requirement for information about their mentees’ lives and experiences motivates mentors in this category to pressure their mentees to disclose information they perceive as pertinent to the intended change: I tried to get things out of him, but he wasn’t like willing to open up, you know. Like, ‘How are things going at school?’ you know? That sort of thing … I would basically have to draw

Downloaded by [University of Bath] at 18:52 05 June 2016

Journal of Child and Adolescent Mental Health 2008, 20(1): 21–32

29

things out, ask specific questions. He wouldn’t like tell me, ‘This is what happened and …’, you know? ‘What happened, Why did she do this?’, you know, ‘Did he do anything wrong?’… And stuff like that. The application of pressure tends to encourage mentees to ‘clam up’ or resist the ‘helping’ conversations their mentors initiate. Morrow and Styles (1995) further suggest that, particularly when disclosure of sensitive information is associated with intended change, as in this case, mentees frequently resist pressure to divulge this information, and may even begin lying to avoid possible resultant lectures. At the same time that these mentors pressure their mentees for disclosure, they devalue what information they do provide, attributing greater validity to that attained from other sources, such as the school social worker or the child’s ‘social history’ provided by the programme. Rather than an understanding of their mentees’ subjective experiences, the mentors’ purpose in acquiring knowledge about their mentees is the attainment of ‘objective’ knowledge about ‘the real situation’ or ‘what actually happened’, which they perceive as more useful to effecting change. Again, the development of empathetic understanding is obstructed. In addition, only those exchanges that concern the divulgence of ‘personal, personal information’ on the part of the mentees, or the dissemination of superior moral knowledge on the part of the mentors, are considered valuable by the mentors. Mentees are clearly also aware of this requirement, as illustrated by Janet’s statement: ‘I feel more comfortable to talk to her about important stuff but (than) unimportant stuff.’ However, these children derive enjoyment and a sense of value primarily from other aspects of their relationships, such as outings and amusing events. This frequently adds to the discrepancy between the relationship partners’ interpretations of the same events, leaving mentors unaware of their significance for their mentees and compounding feelings of inadequacy and frustration. As will be further discussed with reference to the causal sequencing of events in the narrative of ‘help’, the mentors’ preconceived agendas for the relationships in this regard thus tend to be counter-productive to both shared meaning and to their intended aims. Knowledge in the narratives of ‘friendship for its own sake’ Those relationships constructed as ‘friendships for their own sakes’ are explicitly conceptualised as reciprocal processes, within which both partners share information about themselves and learn from each other. In the words of Naefa: It’s a sort of a two-way relationship, you have to give and impart information and share ... I mean I don’t only expect her to learn from me, but I have learned a lot from her as well. In addition, these mentors share information abut themselves with their mentees: ‘I’ve asked her, like, “Is there anything you’d like to know about me or about my family?’’’ They further recognise their own ability to learn from their mentees, alluding to the process of mutual exchange identified by Phillip and Hendry (2000) as important to successful relationships. Rather than constructing their mentees’ knowledge and coping resources as inferior to their own, or those supplied by the programme, the mentors in this category express, for example, admiration for their mentees’ ability to cope with their difficult life circumstances: ‘I admire some of those kids … I don’t, you know, I wouldn’t be able to cope if I would, if I was in her situation.’ The relative freedom of the partners in this category to construct their relationships in their own terms, rather than in those prescribed by the programme, allows for a much broader range of exchanges to become meaningful. In fact, reciprocal exchange of information is constructed as valuable in itself. In the words of one mentor: ‘We talk about everything … we talk about, um … that programme comes on Thursday nights … she likes to talk about that … And she likes to talk about ‘Generations’ and all the soaps and stuff like that.’ Mentors further overtly state their respect for their mentees’ choice in this regard, evidence of the ‘voice and choice’ Morrow and Styles (1995) identify as an important aspect of mentees’ contributions to positive relationships. While Naefa states: ‘I don’t want her to feel as if I’m pressing her for information,’ Daneel, another mentor, expresses disapproval of her mentee’s teacher’s assumption that she is entitled to knowledge about the child’s misbehavior at school: ‘Her teacher confronted me about Marlein being naughty. So I wasn’t impressed with that because she put Marlein on the spot.’

30

Shelmerdine and Louw

Thus, constructions within the narrative of ‘help’ of mentors’ knowledge of their mentees as the key instrument of change appears to engender poor communication, misunderstanding and, in some cases, conflict. Within those relationships recounted in terms of a narrative of ‘friendship for its own sake’, on the other hand, mentors’ ‘not knowing’ (Shotter 1993), and respect for their mentees’ own knowledge, as well as their choice in sharing it, appears to establish the basis for a far more equal relationship, within which reciprocal exchange is encouraged and mutual understanding enhanced.

Downloaded by [University of Bath] at 18:52 05 June 2016

The relationships’ meaning and the narrative plot According to Hermans and Kempen (1993) the plot of a narrative functions to transform an otherwise purely chronological listing of events into a coherent story, highlighting certain events as more relevant than others to the story’s development. Plot sequences are thus inseparable from constructions of the relationships’ meaning and of the events and actors within them. The cause-effect plot in the narrative of ‘help’ As described, the narrative of ‘help’ is structured in terms of a cause-effect plot, in terms of which the mentors effect change in their mentees and their lives. This plot sequence makes for a rigidly prescribed approach to the relationships in terms of their enactment, as well as in terms of expectations for their development and outcomes, especially on the part of the mentors. In terms of this narrative structure, the relationship stories can have one of only two fixed outcomes — success or failure. These mentors locate the value of their relationships exclusively in their ability to influence the behaviours and outcomes of their mentees; for example, in their ability to influence them to ‘grow up decent’ or to ‘finish high school’ and ‘get (themselves) a job’, or to ‘treat (their) friends in a better way’. Due to their arguably unrealistic expectations of their ability to influence their mentees in this way, these mentors tend to negatively evaluate their relationships and express considerable frustration, despondency and inadequacy when anticipated changes do not occur. Layla describes her feelings in this regard: I mean all that effort that I’ve put into this whole thing, and it just ends, you know? … I feel like, you know, I’ve made a difference, but I haven’t made that much of a difference. That’s how I’m feeling … I can’t see changes, that’s the problem, I suppose … I can’t see the actual changes. Ja, I can’t see the effects. This frustration and dissatisfaction has been found by previous research to be the primary reason for premature termination of relationships (Rhodes 2002a). The focus on change thus appears to contribute not only to poor relationship development, but potentially to short duration of relationships, which has further been found to be associated with negative outcomes (Rhodes 2002a). Plot sequences in narratives of ‘friendship for its own sake’ In contrast, mentors who narrate their relationships as ‘friendships for their own sakes’ explicitly reject the notion of a direct causal relationship between their actions and what become, in the narrative of ‘help’, the mentees’ reactions. One mentor states, for example, ‘I mean I can’t tell her, “Marlein, behave”, and expect her to behave.’ The agency attributed to the mentees in these accounts denies the possibility for mentors to directly affect change in their mentees’ lives. In addition, the primary value of these relationships is constructed in terms of the relationship process and the ‘time spent together’, rather than in terms of possible benefits external to the relationship itself. As one mentor states: ‘There’s no use drilling into her all the time about the future, about goals and this … What you have right now, work at it.’ The relative fluidity of the plot sequences in these relationships allows their meaning, and the roles of the mentors and mentees within them, to be constructed in multiple and less prescribed ways. The focus on process rather than outcomes allows the value of these relationships to be constructed in terms of events within which the partners derive mutual enjoyment and meaning, rather than being interpreted according to a preconceived agenda. The lack of a preconceived agenda further appears to enhance the mentors’ responsiveness to their mentees in multiple aspects of the relationships, such as decision-making about when and in what areas mentees require help, suitable topics for conversation, awareness of their mentees’ assessments of the relationships’ worth, and so on.

Journal of Child and Adolescent Mental Health 2008, 20(1): 21–32

31

Downloaded by [University of Bath] at 18:52 05 June 2016

Mentors are thus not subject to the same frustration and self-criticism that appear within the narratives of ‘help’, and are more satisfied with their relationships: I just think that describing how rewarding it is, is an understatement, you can never really say how much … Like I will tell myself I don’t have time to actually make, but every week I make time to go. The reward derived by these mentors thus appears further to motivate them to be consistent within their relationships, one of the key qualities identified by previous research as associated with positive change (Rhodes 2002a). The freedom these mentees report in communicating with their mentors about both positive and negative aspects of their lives indicates the kind of atmosphere identified by Morrow and Styles (1995) as more likely, in fact, to bring about positive change. All the relationships in this study are therefore, to some extent, constructed in terms of their purpose to help their mentees. However, differences in the primacy attributed to this overall goal, and in the ways in which help supplied is expected to unfold, account for substantial differences in the quality and functioning of these relationships. Conclusion This study has illuminated the underlying constructions of these relationships and the interactions of these with more overt relationship characteristics, arguing for the importance of these constructions in determining the qualities of the relationships that develop. The study has further shown how the programme and broader social context influences the relationships, making available particular narratives, in terms of which the mentorships are enacted. Due to the programme’s overall purpose to effect change in the mentees’ developmental trajectories, the narrative of ‘help’ may be the one most readily available to mentors and mentees, in terms of which to construct their relationships. This narrative further interacts in many of the relationships’ accounts with narratives from South Africa’s socio-political history, in terms of which mentors, and the social groups to which they belong, are constructed in stereotypical terms as different from their mentees with regard to the very characteristics, abilities and behaviours the programme is intended to change. Those relationships constructed in terms of the narrative of ‘friendship for its own sake’, however, provide evidence that the programmme’s overall purpose and socio-political context does not necessarily lead to the former negative constructions. The success of these latter relationships appears, in fact, to be dependent on the partners’ ability to move beyond the constraints of the above narratives, and to establish more personal relationships, based on equality and on the value of the relationship itself, rather than on the programme’s ultimate goals. While the small sample of mentorships is an obvious limitation of the study, it has enabled in-depth exploration of relationship qualities. This study was also unable to link findings about relationship processes to outcomes, due to the fact that outcomes data was not available at the time of the study. A further potential limitation is the reliance on participants’ accounts of their relationships, with no independent confirmation from other sources. However, it is believed that this study nevertheless offers valuable information about these relationships — in the words of Portelli (Stilles 2001: 2): ‘Oral sources tell us not just what people did, but what they wanted to do, what they believed they were doing, and what they now think they did’, providing a particular form of insight into the processes occurring in these relationships unobtainable through other means. In spite of the above limitations, the study has thus produced insights, through analysis of the relationships’ narratives, with important implications for the future training of mentors, highlighting the relevance not only of overt relationship practices but also of the relationships’ underlying constructions. Acknowledgements — We wish to thank the staff of the BBBS of South Africa program, who gave generously of their time, and always were willing to share information and be of assistance to us. We are particularly thankful to the mentors and youths who participated in the research. Public/Private Venture of Philadelphia, PA has shared many details of the Morrow and Styles (1995) study with us. This has helped us to conduct a much more thorough study than would have been possible otherwise.

32

Shelmerdine and Louw

Downloaded by [University of Bath] at 18:52 05 June 2016

References Barron-McKeagney T, Woody JD and D’Souza HJ (2001) Mentoring at risk Latino children and their parents: Impact on social skills and problem behaviours. Child and Adolescent Social Work Journal 18: 119–136 Beam MR, Chen C and Greenberger E (2002) The nature of adolescents’ relationships with their ‘very important’ non-parental adults. American Journal of Community Psychology 30: 305– 325 Billig M (1999) Freudian repression: Conversation creating the unconscious. Cambridge: Cambridge University Press DuBois DL, Holloway BE, Valentine JC and Cooper H (2002) Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology 30: 157–197 DuBois DL and Neville HA (1997) Youth mentoring: Investigation of relationship characteristics and perceived benefits. Journal of Community Psychology 25: 227–234 Dyan D and Katz E (1992) Scripting media events. Massachusetts: Harvard University Press Gee JP (1999) An introduction to discourse analysis: Theory and method. London: Routledge Grossman JB and Rhodes JE (2000) The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology 30: 199–219 Grossman JB and Tierney JP (1998) Does mentoring work? An impact study of the Big Brothers Big Sisters program. Evaluation Review 22: 403–426 Herrera C (1999) School-based mentoring: A first look into its potential. http://www.ppv.org/indexfiles/mentorindex.html [Accessed October 2002] Herrera C, Sipe CL and McClanahan WS (2000) Mentoring school-age children: Relationship development in community-based and school-based programs. http://ppv.org/indexfiles/mentor-index.html [Accessed October 2002] Hermans HJM and Kempen HJG (1993) The dialogical self: Meaning as movement. California: Academic Press Hutchby I and Wooffit R (1998) Conversation analysis: Principles, practices and applications. Cambridge: Polity Press Jucovy L (2002) Same-race and cross-race matching. http://ppv.org.indexfiles/mentor-index.html [Accessed October 2002] Kress G (1990) Critical discourse analysis. Annual Review of Applied Linguistics 2: 2–11 Liang B, Tracy AJ, Taylor CA and Williams LM (2002) Mentoring college-age women: A relational approach. American Journal of Community Psychology 30: 271–288 Morrow KV and Styles MB (1995) Building relationships with youth in program settings: A study of Big Brothers Big Sisters. Philadelphia: Public/Private Ventures National Research Council and Institute of Medicine (2002) Community programs to promote youth development. Washington DC: National Academy Press PANOS (1999) Giving voice: Practical guidelines for oral testimony research. Oral testimony programme. London: PANOS Patton M (1990) Qualitative research and evaluation (2nd edn). Newbury Park, CA: Sage Phillip K and Hendry LB (1996) Young people and mentoring – towards a typology? Journal of Adolescence 19: 189–201 Phillip K and Hendry LB (2000) Making sense of mentoring or mentoring making sense? Reflections on the mentoring process by adult mentors and young people. Journal of Community and Applied Social Psychology 10: 211–223 Rhodes JE (2002a) Keeping mentors and mentees together over time. http://www.mentoring.org/research_ corner/feb_background.adp [Accessed October 2002] Rhodes JE (2002b) What’s race got to do with it? http://www.mentoring.org/research_corner/mar_background. adp [Accessed October 2002] Rhodes JE, Bogat GA, Roffman J, Edelman P and Galasso L (2002) Youth mentoring in perspective: Introduction to the special issue. American Journal of Community Psychology 30: 149–155 Scheurich J (1997) Research method in the postmodern. London: The Falmer Press Shotter J (1993) Conversational realities: Constructing life through language. London: Sage Publications Stilles A (2001) Prospecting for truth amid the distortions of oral history. New York Times, 11 March 2001 Styles MB and Morrow KV (1996) Understanding how youth and elders form relationships: A study of four Linking Lifetimes programs. In: Sipe CL (ed.), Mentoring: A synthesis of P/PV’s research: 1988–1995. Philadelphia: Public/Private Ventures. pp 28–33 Tierney JP, Grossman JB and Resch NL (1995) Making a difference: An impact study of Big Brothers Big Sisters. Philadelphia: Public/Private Ventures