Exploring the contribution of mentoring practices to mentee learning in a Malaysian youth development programme

Abstract

Within the mentoring relationship, it is generally accepted that learning takes place through the sharing and transmitting of knowledge between mentor and mentee, where mentors employ practices such as coaching, facilitation, guidance and reflection on experiences. Despite the number of studies on mentoring in the context of youth development programmes, mentoring relationship processes in youth development contexts is not well understood. Few studies have examined the variety of practices used by mentors and explored their contributions to different types of mentee learning. In an attempt to better understand this relationship, an exploratory correlational study was conducted on a group of youth development programme participants from Malaysia (N = 90) to identify how various mentoring practices enhance learning. The results showed that overall, facilitation and coaching contributed the most to new learning. While coaching showed the highest contribution to skill acquisition and knowledge learning, the contribution of facilitation was mainly towards attitude learning. Implications of the findings are discussed.

Keywords::

• mentoring
• mentor–mentee relationship
• learning outcomes
• youth development programmes
• Malaysia

Introduction

The fundamental process and primary purpose of mentoring is learning, which includes the transmission of knowledge as its core element (Bozeman & Feney, 2007; Zachary, 2000). Bozeman and Feeney (2007) state that mentoring is an informal, face-to-face communication over a sustained period of time between a person perceived to have greater knowledge, wisdom or experience (the mentor) and a person perceived to have less (the mentee). Hezlett (2005) pointed out that the mentoring relationship enables an increase in cognitive, affective and skill-based learning outcomes (St-Jean & Audet, 2008). While most literature describes mentoring in the context of a partnering relationship between a more experienced and knowledgeable adult or peer (mentor) with a more junior and inexperienced subordinate (mentee), mentoring goes beyond the establishment of a developmental relationship. As Zachary (2000) suggests, mentoring is a relationship that at its core is about learning and the process towards it. It is, therefore, an educational approach to impart new knowledge, skills, attitudes and competencies.

Part of the uniqueness of this special relationship is that simultaneously the mentee takes responsibility for his (or her) own learning through encouragement, guidance and the example shown by the mentor. Through this motivation, learning takes place when both parties realise the need and interest that learning will satisfy, and when there is facilitation from the mentor (Knowles, Holton, & Swanson, 2005). According to Knowles et al. (2005), the goal of learning is about change and ‘… the acquisition of habits, knowledge, and attitudes’ (p. 11). The acquisition of new knowledge, skills and/or attitudes resulting from learning forms a ‘new’ person of greater maturity who is better equipped to deal with the challenges of his or her environment.

Mentoring

The vast majority of studies on mentoring define the practice within a workplace learning and professional development context. Within this theoretical frame, the fundamental purpose of mentoring is personal and professional development (Brockbank & McGill, 2006). Several authors have also mentioned its core functions as: (1) learning technical skills and knowledge; (2) learning about the current job; (3) learning about organisational culture; (4) learning organisational policies and (5) preparation for a future job (Caruso, 1992, cited in Brockbank & McGill, 2006). Others, like Johnson and Ridley (2008), take a more general view by describing a mentor as a teacher and coach who provides encouragement and support, shapes behaviour using reinforcement, stimulates growth with challenging assignments, narrates growth and development, nurtures creativity and acts as an intentional model.

In the context of mentoring within youth development contexts, Dunphy et al. (2008) define mentoring as:

the process by which a more experienced, trusted guide forms a relationship with a young person who wants a caring, more experienced person in his/her life, so that the young person is supported in growth towards adulthood and the capacity to make positive social connections and build essential skills. (p. 9)

They further explain that the mentoring relationship can be ‘informal’ or ‘formal’. In the informal relationship, the mentoring bond develops when the young person receives guidance, encouragement and emotional support, whereas in the formal relationship, there is an organised mentoring programme for a one-to-one relationship that can take place at community and youth clubs. There are, however, other forms such as one-to-one mentoring, group mentoring, team mentoring and peer mentoring. Regardless of the form, all mentoring formats share the need for a relationship to first be established between the mentor and mentee before they proceed further with the mentor giving guidance, support and encouragement to the mentee (Kay & Hinds, 2005).

Learning within the mentor–mentee relationship

Mentors employ various approaches and techniques to enable learning to take place (Cranwell-Ward, Bossons, & Gover, 2004). Among them are coaching, guiding, counseling, supporting and the use of reflective dialogue on experiences (Zachary, 2000). Through these approaches, the mentee learns experientially from the lessons offered by the mentor, as well as from reflection on his or her own experiences. Drawing on Bandura's social cognitive and social learning theories, to be effective in enabling learning, the mentor models behavior and becomes ‘someone who, with regard to a particular area of life, is both trusted and respected … where trust and respect are based on a significant level of experience and knowledge what the latter believes the former to have’ (Stebbins, 2006, p. 5). For the relationship to be sustainable, the mentor must also be concerned about the mentee's personal growth and development. According to Cranwell-Ward et al. (2004), the role of a mentor is that of a coach, counselor, facilitator as well as that of an evaluator. Evaluation is necessary for the mentor to determine the progress of the mentee and hence to provide necessary feedback. Thus, there must be a certain bond shared between the mentor and mentee without which misunderstanding can occur and derail the learning process. The mentoring relationship thus requires a high degree of trust and mutual respect so that the mentor can help the mentee to realise his/her potential (Garvey & Langridge, 2006) through shared activities, information and encouragement (Delgado, 2002).

In the context of mentee learning, Harrison, Lawson, and Wortley (2005) proposed a model that considers critical reflective practice within experiential learning, suggesting that reflection and experiential learning are key parts of the mentoring-learning process. In addition, coaching is often employed in mentoring to enable the mentor to help the mentee in enhancing the mentee's performance and capability in developing skills and knowledge (Somers, 2008). While mentoring is aimed more towards long-term development, the focus of coaching is on immediate learning and development needs. Somers (2008) further added that the mentor has to coach the mentee on how to learn independently by giving guidance and encouraging reflection so that the mentee can synthesise what has been learned through his or her experiences. Finally, the mentor facilitates the mentoring process by encouraging the mentee to value, develop and express him or herself to work towards their goals (Hunter, 2007). In facilitation, the mentor does not direct or control the mentee, but instead helps the mentee to arrive at understanding and make his or her own decisions. As obstacles and problems arise, the mentor employs guidance to help the mentee navigate, explore different approaches and make necessary changes to overcome the challenges faced (Hunter, 2007).

Mentoring as a learning approach should contribute to the acquisition of knowledge, attitude and skill as posited by Bloom, Engelhart, Frust, Hill, and Krathwohl (1956), Krathwohl, Bloom, and Masia (1964) and Knowles et al. (2005). In planning youth development programmes, Edginton and Edginton (1994) have suggested that youth needs and development factors should be considered. And since youth development includes change and learning behaviours, Edginton and Edginton (1994) have suggested that such behaviours should be in accordance with Bloom et al.'s (1956), Krathwohl et al.'s (1964), and Bush's (1972, cited in Edginton and Edginton, 1994) learning taxonomies of ‘thinking, feeling and action’ behaviours (p. 143). They argued that the different levels of learning in the cognitive, affective and psychomotor domains be inclusive of the performance objectives of youth. As such, learning should be embedded within the programme goals of all youth development programmes.

The current study

Programme Remaja Perdana Rakan Muda (PRPRM) is the Malaysian version of the Duke of Edinburgh's International Award programme, an internationally based youth development programme and the only such programme in Malaysia that engages young people in intensive one-on-one mentoring relationships towards the achievement of positive developmental outcomes. It is part of the Ministry of Youth and Sports Malaysia's umbrella Rakan Muda (Young Friends) programme for developing healthy lifestyles among young people within the age group of 14–25. PRPRM is a voluntary, non-competitive programme that requires sustained effort over time by encouraging young people to design their own array of activities, set their own goals and challenge themselves to achieve their aims.

The programme aims to achieve five core outcomes: make a positive contribution to society and demonstrate social responsibility; enjoy healthy lifestyles for good mental and physical well-being; develop talents, broaden abilities, increase self-confidence and improve employability; gain a deeper understanding of the environment and build new relationships with other youth (The Duke of Edinburgh's International Award Foundation, 2012).

Mentoring in the Duke of Edinburgh Award for Young People was outlined as one of the roles of the Award leader (The Duke of Edinburgh's Award International Association, 2004). It is not detailed out or defined, however, how mentoring is to be conducted, nor is any training provided for the Award leaders/mentors on mentoring specifically. The training provided is limited to leadership and administration of the programme. This study was therefore partly aimed at determining how mentoring is conducted in the Award programme to help programme leaders to better understand the quality and nature of mentoring occurring within the programme.

In Malaysia, the use of mentoring as a professional development strategy is becoming more prevalent in the fields of business, education and higher education (Ismail & Arokiasamy, 2007; Lo & Ramayah, 2011). There is little evidence, however, that mentoring has been pointed to as a strategy in other youth development programming contexts. The extant mentoring literature in Malaysia is mainly limited to career, professional or competency development in the aforementioned contexts. These studies report on mentoring practices that are dependent on the goals of the organisations or programmes and assimilated into the organisations' normative practices.

While the literature has acknowledged that mentoring has learning and personal development as its core purposes (Brockbank & McGill, 2006; Zachary, 2000) and the transmission of knowledge (Bozeman & Feeney, 2007), most studies on mentoring have focused on the development of the interaction between mentors and mentees (Rhodes, 2002), the process of mentoring (Cranwell-Ward et al., 2004; Dolan & Brady, 2012; Kay & Hinds, 2005) and the development of trust and respect in the partnership (Garvey & Langridge, 2006; Stebbins, 2006). Few studies on mentoring within the context of youth development programmes in particular have linked mentoring processes with learning outcomes, and, more specifically, the extent to which different mentoring practices associate with specific forms of learning. The current study, therefore, aimed to address this gap by identifying how the various practices that mentors employ to ensure learning takes place – defined in this study as bonding, coaching, facilitating and reflection – associate with different learning outcomes. The findings of the study would then enable trainers of youth mentors and the mentors themselves to assess their training or mentoring approaches in meeting the desired learning outcomes, and realise the possible strengths and weaknesses in their mentoring work.

Method

Research design

Correlational study was employed for this study to determine the association between mentoring practices and learning outcomes in the school-based Award programme. The study was designed to examine the associations between mentoring practice and learning outcomes in the form of knowledge, attitude and skill, as well as to provide a general description on the relationship of these two phenomena (Singleton & Straits, 1998). We did not seek to determine causes of the phenomenon in question.

Participants

The PRPRM programme has three levels of participation, differentiated according to the amount of time and level of challenge of the activities. The three levels include Bronze, Silver and Gold with a minimum time period of 6, 12 and 18 months of involvement, respectively. Respondents for the current study included Gold Award winners only, as only this was the only group that underwent at least 12 months of mentoring. Prior research has reported that mentoring is most effective over longer periods of time such as of at least 12 months (Rhodes in DuBois and Karcher, 2005, p. 33). Based on this criterion, we felt that greater insight on the learning process could be acquired by limiting the sample to those having had at least 12 months of mentoring experience. Study respondents included 90 students (M age = 17.30, SD = 0.46; 60% male; 91.1% Malay) from one suburban boarding secondary school in Malacca, Malaysia, and included all Gold award winners for the year 2011.

The sample was predominantly Malay. As this study was focused solely on exploring the practice and process of mentoring within a specific programmatic context, we did not take into consideration differences in cultural or social norms of the sample and therefore any such implications were not investigated. Moreover, the administrators of the Award programme concentrate on the development of the young people based on the criteria outlined by the International Award programme regardless of race or ethnicity. Although the majority of our sample was Malay, we are unaware of any differences in how the programme was implemented, nor whether the racial make-up of the sample had any implication on the outcomes or delivery of the programme. However, future research might explore these issues further.

Procedures

Data were collected using survey method. Due to limitations experienced by the researchers to access the respondents directly, data collection was administered through the cooperation of the school counsellor, who was responsible for the administration of the PRPRM programme. The survey questionnaires were given to the school counsellor who was responsible for delivering them to the respondents. The completed surveys were picked up after 2 months' time by one of the research team members. The data were analysed using IBM SPSS Statistics 19. Preliminary analyses were conducted to ensure no violation of the assumption of normality, linearity and multicollinearity.

Instrumentation

Given the lack of instrumentation on learning within mentoring relationships, a self-report survey questionnaire was developed. Two main scales were created for mentoring practice and learning outcomes. Items in the questionnaire for mentoring practices were drawn from various literatures as outlined earlier. Items for the learning outcomes scale were developed from the master list of PRPRM programme outcomes. The programme outcomes were first categorised according to keyword matching the three types of learning outcomes, i.e. knowledge, attitude and skill, as put forth by Bloom et al. (1956) and Anderson et al. (2001) (knowledge), Krathwohl et al. (1964) (attitude) and Simpson (1972, cited in Clark, 2010) (skills). Once the programme outcomes were matched with the taxonomic keywords, item statements were created. For both scales, tests for reliability and validity were conducted following an initial pretest of the instrument. As the items were originally identified and grouped according to theory, exploratory factor analysis (EFA) was also conducted to establish construct validity of the instrument and to identify the statistical dimensions (Hair, Black, Babin, Anderson, & Tatham, 2006).

Validation was conducted to measure the instrument's accuracy and reflection of the concepts considered (Babbie, 2001; Hair et al., 2006). To establish face validity of the instrument, individuals with expertise in the mentoring programme upon which the study was based were asked to provide feedback on the items. They included a practicing clinical psychologist and Senior lecturer from one Australian University who was a Gold Award recipient, the Asia-Pacific Regional Programme Manager of the Award programme and the Malaysian Director of the PRPRM programme. For content validity, several training consultants familiar with mentoring, facilitation and coaching practice provided feedback on the instrument constructs and survey items. In addition, previous winners of the youth Award programme in Malaysia were consulted for advice on the items based on their experience, as well as for translating the instrument into Malay language. On the final survey, all items were presented in both Malay and English.

To establish construct validity of the measures, EFA was conducted for both the mentoring practices and learning outcome measures. The item data were entered into SPSS using oblique (promax) rather than orthogonal rotation (varimax) to explore the inter-relatedness of the factors (DeJong, Faulkner, & Warland 1976). All items were entered into SPSS v. 21 and factored to explore whether they would empirically fall into the theoretical constructs upon which the inventory was based. For the mentoring practice scale, the Kasier-Meyer-Olkin measure of sampling adequacy coefficient was 0.881 and Bartlett's test of sphericity was 0.000, both indicating that factor analysis was an appropriate test to extract principal components from question items used in the study. In accordance with the design of the subscales, the analysis produced a four-factor solution (based on eigenvalues>1), which accounted for 76% of the variance. Results from the component correlation matrix showed low to moderate correlations between the factors ranging from 0.54 to 0.96, indicating that the scales were multidimensional, capturing four unique factors (Ang, 2005). The four constructs were bonding (four items; α = 0.91) (‘My mentor and I discussed about the relationship we are going to have’, ‘My mentor and I outlined the rules of our relationship’); coaching (four items; α = 0.90) (‘My mentor coaches me on how to clarify my objectives and expectations’, ‘My mentor coaches me to decide on the activities I should undertake in the Award programme’); facilitating/enabling (four items; α = 0.85) (‘My mentor has enabled me to see things from different angles’, ‘My mentor has enabled me to make decisions and solve problems by myself’) and reflecting (five items; α = 0.89) (‘My mentor helps me decide on the instructors I need for the activities that I undertake’, ‘My mentor requires me to describe in detail the experiences that I encountered’). Mentoring practice scale items were based on a 9-point Likert-like scale that ranged from 1 = ‘Totally Disagree’ to 9 = ‘Totally Agree’.

For the learning outcomes scale, the Kasier-Meyer-Olkin measure of sampling adequacy coefficient was 0.875 and Bartlett's test of sphericity was 0.000. The analysis produced a three-factor solution (based on eigenvalues>1), which accounted for 69% of the variance. Results from the component correlation matrix showed low to moderate correlations between the factors ranging from 0.48 to 0.94. The three constructs were knowledge learning (five items; α = 0.86) (‘As a result of the program I am able to explain the process that I need to undergo when doing my activities’, ‘As a result of the program I am able to organize my activities in an orderly and systematic way’); attitude learning (five items; α = 0.86) (‘As a result of the program I am able to be more self-reliant’, ‘As a result of the program I am able to be more independent’) and skills learning (five items; α = 0.88) (‘As a result of the program I am able to excel in my presentation skills’, ‘Build on my interpersonal skills’). A 9-point Likert-like scale ranging from 1 = ‘Totally Not Me’ to 9 = ‘Totally Me’ was used for the responses.

Results

To assess the associations between mentoring practice and learning outcomes, we conducted bivariate correlation and linear regression analyses (Table 1). The findings from Table 1 on mentoring practices range from reflecting with a mean score of 6.75 to facilitating/enabling with a mean of 7.13. All three outcome variables indicated relatively high means of 7.20–7.23. In the correlation matrices, all correlations were found to be significant. Reflecting indicated the lowest correlation with attitude and skill learning, where it showed a moderate correlation with Knowledge learning. This suggests that reflecting was more closely associated with knowledge learning than in the development of attitude and skill. Coaching and facilitating/enabling, respectively, indicated the strongest associations of the four mentoring practices. Coaching was most highly associated with knowledge and skill learning, while facilitating/enabling correlated with all three outcomes in a similar manner with r coefficients ranging from 0.51 to 0.54.

Table 1 Descriptive statistics and correlations of the test variables (N = 90).

				r
Variable	Range	M	SD	1	2	3
Learning outcomes
1. Knowledge learning	1–9	7.22	1.12	–	–	–
2. Attitude learning	1–9	7.20	1.20	–	–	–
3. Skill learning	1–9	7.23	1.22	–	–	–
Mentoring practices
4. Reflecting	1–9	6.75	1.42	0.43***	0.26*	0.26*
5. Coaching	1–9	6.94	1.54	0.62***	0.40***	0.62***
6. Bonding	1–9	6.42	1.64	0.34**	0.24**	0.42***
7. Facilitating/enabling	1–9	7.13	1.35	0.54***	0.51***	0.52***

Note: *p < 0.05, **p < 0.01, ***p < 0.001 (two-tailed).

Hierarchical linear regression analysis was conducted for each of the three learning outcome types. We used a two-step enter approach whereby control variables income, education level and place of residence (urban/rural) were entered in Step 1, followed by the mentoring practice variables in Step 2. The β coefficients indicate that Coaching was the main contributor to knowledge (β = 0.56; p < 0.001) and skill learning (β = 0.69; p < 0.001); facilitating/enabling contributed to attitude learning (β = 0.48; p < 0.01) while reflecting showed a negative contribution to skill learning (β = –0.48; p < 0.001). Bonding was not associated with any of the learning outcomes as indicated in Table 2 below.

Table 2 Standardised β coefficients for knowledge, attitude and skills learning.

	Knowledge learning		Attitude learning		Skills learning
Variable	Step 1	Step 2	Step 1	Step 2	Step 1	Step 2
Demographics
Income	0.08	0.05	− 0.07	− 0.11	0.03	− 0.04
Education	0.05	− 0.01	0.04	− 0.01	0.01	− 0.05
Residence	− 0.10	− 0.03	− 0.09	− 0.02	− 0.17	− 0.13
Mentoring practices
Bonding	−	− 0.03	–	0.07	–	0.16
Coaching	–	0.56***	–	0.20	–	0.69***
Reflecting	–	− 0.08	–	− 0.24	–	− 0.48***
Facilitating/enabling	–	0.20	–	0.48**	–	0.25*

Note: *p < 0.05, **p < 0.01, ***p < 0.001.

Discussion

Mentoring has been defined in many ways. As suggested by Eby, Rhodes, and Allen (2007), it is a relationship that involves the acquisition of knowledge with the primary goal of personal growth and development. Eby et al. further explain that in mentoring relationships, the bond between mentors and mentees can vary considerably. In the case of mentoring relationships that involve teachers and students in particular, the bonding experience tends to be different than in other developmental relationships. In the current study, the mentor–mentee relationships in the school where the study was conducted were comprised of teachers and students in the school. The findings suggest that mentoring relationship between teachers and students might experience a certain distance, due to the pre-existing student–teacher relationship, that will affect the type of bonding that might occur in non-teacher–student mentoring relationships. This can be explained by the ethical necessity of the professional relationship that has to be maintained between teachers and students. Teachers who play a dual role of being an educator and a mentor must uphold the delicate balance of professional ethics and integrity – often marked by a certain professional ‘distance’ – with that of a close mentor–mentee relationship.

Despite bonding not being significant in the multivariate results, bivariate results indicate that it remains an important component of mentoring. When mentors and youth spend consistent time together, bonding that is characterised by mutuality, trust and empathy will have significant benefits for youth (Rhodes & DuBois, 2008). The importance of a strong bond between the mentor and mentee cannot be denied because there is a danger of more harm than good without it (Larose, Cyrenne, Garceau, Brodeur, & Tarabulsy, 2010). Hamilton and Hamilton (2010) pointed out that a close relationship is not necessary with coaching. Nonetheless, Keller (2007) argued that bonding plays an important role in mentoring when solidified and strengthened because it allows mentors to provide better advice and guidance to their mentees. The findings here would then suggest that bonding has more to do with the establishment of the mentor–mentee relationship and acts as an indirect, rather than direct contributor to learning, while the other facets of mentoring, i.e. coaching and facilitating, act as the bridge(s) for learning to occur.

The importance of facilitation as a predictor of attitude learning is consistent with Jonson (2008), who suggested that a main role of a good mentor is that of being a facilitator, rather than an authoritative figure, or that of a ‘guide on the side’ (ibid., p. 98). Such individuals encourage, involve, help, share and motivate. Mentors as facilitators contribute to youths' identity development by creating opportunities for them to engage in activities, acquire resources and meet their occupational needs (DuBois, Portillo, Rhodes, Silverthorn, & Valentine, 2011). Although little existing research points to the role of facilitation on attitude formation specifically, we can tentatively argue from the findings that facilitation successfully contributes to mentees' attitude learning by exposing them to a variety of opportunities and experience, as well as the attributes of mentee competencies, self-confidence, self-direction and professionalism (Jonson, 2008).

The findings support the role of coaching in helping to fill gaps in knowledge in the mentoring relationship (Zachary, 2000). Through coaching, the coach guides and encourages the mentee to find his or her own solutions to problems faced (Hook, McPhail, & Vass, 2007; Somers, 2008). During the facilitation process, one of the mentor's roles is to coach the mentee on how to improve the mentee's decision-making skills (Braakman & Edwards, 2002; Schwarz, 2002). It is also part of the mentor's role to help the mentee plan and implement his or her goals, as well as clarify issues of concern (Hunter, 2007). Nonetheless, while coaching is considered a part of mentoring, it is not mentoring (Somers, 2008). It could be posited that through coaching, the mentor further facilitates learning, making it an important part of the bridge linking the mentor–mentee relationship to learning.

Although bivariate results showed reflecting as a significant but weak predictor of skill learning, multivariate results showed a negative association. Reflection is an integral part of learning; however, the negative outcome coupled with the high β value for coaching suggests that coaching is more relevant and takes precedence over other forms of mentoring in the context of skill learning. Since skill learning primarily reflects the acquisition of psychomotor abilities, it is more mechanistic than other forms of learning and therefore has more to do with repetition rather than reflection. Moreover, the relationship between skill learning and cognitive learning is inherently negative in that it naturally progresses from the cognitive stage – where the learner is conscious about controlling movement – to the associative where the learner relies less on cognition and is able to associate the movements, and does not have to think about every detail, but where the movement becomes more natural without feeling awkward while doing it. This reflects an autonomous stage where the learner has almost completed the learning, but is refining the new skill through practice (Demand Media, 2011). At this final stage, the learner no longer needs feedback from the instructor (which in this case, includes the mentor) as the skill has become automatic when the mind and body become one.

These arguments suggest that reflective practice could be positive in skill acquisition at the initial phase of cognition, but that it is practice that later contributes to skill acquisition; and that a learner's past experience could determine whether reflection contributes to positive or negative outcomes. The nature of the present study sample must also be considered in relation to these findings. They were students in a secondary school where a more pedagogical (and less andragogical) learning approach is employed. Hence, it is possible that the teacher/mentor may not be engaging the students in much reflective practice.

In Hezlett's (2005) study, the author states that an important objective of the mentoring relationship is learning, and that the support received from mentors has been linked to enhanced mentee learning while negative experiences in mentoring relationships can result in a lack of learning. Such situations can also go hand-in-hand with the termination of the mentoring relationship altogether. In integrating the taxonomy of learning outcomes into formal mentoring, Hezlett (2005) supports the proposition that mentoring enhances mentee cognitive, skill-based and affective learning. Cognitive knowledge is comprised of organisational and technical knowledge; skills include technical, interpersonal, time management and self-organisational abilities; and affective changes mainly refer to increases in self-confidence and self-efficacy. However, in the current study, we found that cognitive, affective and skill learning occurred at similar reported levels. The scope of the study did not examine in-depth the different kinds of cognitive learning that could have taken place as suggested by Hezlett.

It is important to mention the role of power relationships in this study. The mentors in the current study were teachers and the mentees students. While most mentoring relationships involve a bond between mentor and mentee, the pre-existing relationships being one between teacher and student may have reduced the sense of bonding between the parties. This was supported by the study findings and might be due to the professional distance between the teacher-mentor with the student-mentee. In this case, the former has a certain level of power or influence over the latter. Ensher and Murphy (2005) argued that in mentoring, power should not rest with one party; the power relation has to mutually benefit both the mentee and mentor, which means that there is a reciprocal power relationship. As the mentee gains in new learning, emotional support, help and growth, the mentor also gains from the mentees' feedback, coaching skills, enhanced reputation and personal fulfilment (ibid.). In the context of the Duke of Edinburgh programme specifically, the mentee also gains from positive recognition in the form of receiving their Bronze, Silver or Gold Award. As a result, the mentees' success is also a reflection of the mentors' perceived competence in mentoring (Ragins, 1997). The potential downside of this relationship, however, according to Ragins (1997), is that if the mentee fails to meet his goals or performance, it reflects poorly on the mentor and could undermine his reputation and credibility.

This suggests that power relations in mentoring relationships in the Duke of Edinburgh programme might very well be two-way. Both parties have power over each other, but in different forms. In this study, the teacher-mentors' power in outlining the developmental direction of the student-mentees was balanced by the reality that if a student-mentee fails to meet the goals of the Award programme, it reflects on the teacher-mentor's credibility.

For a relationship to be successful and mutually powerful, Ensher and Murphy (2005) disagree with formal mentoring relationships. Their research found that spontaneously developed or informal mentoring relationships are on the whole more effective. They suggest that what is most important in the relationship is passion, and that both mentor and mentee can be dissimilar but have skills and needs that complement each other. Moreover, the authors suggest that the mentee should initiate the relationship by identifying who his/her prospective mentor or mentors should be. This is to ensure that the mentee can identify with the mentor-to-be in order to enable the mentee to meet his/her developmental and learning goals, and so that the two are compatible and comfortable with each other. This is not so in a formal mentoring relationship where the mentor-to-be is usually selected for the mentee and may not meet the mentee's expectations, such as in the current study.

Nonetheless, this correlational study was confined to ascertaining the relationship between different mentoring practices and the dependent variables of knowledge, attitude and skill learning outcomes. Although analysed individually, we recognise that in practice mentoring practices occur simultaneously. Mentors do not ‘bracket off’ their practices and approaches as neatly as the analysis here would imply to target different types of learning outcomes. Despite this limitation, we feel that as an exploratory study, the findings provide an initial understanding as to how mentoring practices associate with different types of learning outcomes. This has value not only for future research, but also as an indicator of the overall nature of mentoring within the Duke of Edinburgh programme as practiced in Malaysia.

Recommendations for future research

This paper was confined to examining the associations between mentoring practices and learning outcomes. Future research should re-examine the predictive relationship of mentoring practices to learning outcomes as originally outlined by Bloom, Anderson, Krathwohl, and Simpson with larger samples where structural equation models can be used to validate the strength of prediction of the variables. A multi-level modelling approach could also be employed to test each mentoring activity and compare their effectiveness and learning outcomes. Qualitative studies could also provide a more realistic data on the level of each learning outcomes that each respondent would have acquired. Furthermore, a future study on mentoring and learning taxonomies could enable Award leaders and mentors to customise their mentoring approaches to meet specific learning needs of the participants.

The findings of this study suggest that there were gaps in the mentoring relationship as indicated by the lack of significant contribution of bonding to the learning outcomes. This tentatively indicates that there was insufficient development and establishment of the mentoring relationship. The finding provides an area for potential future research. Even though the study has indicated that the other mentoring practices had relatively strong relationships and predictability to learning outcomes, they were neither strong nor conclusive. It is therefore important to bridge this gap in understanding the bonding aspect of the mentor–mentee relationship.

Previous research and the literature on mentoring have pointed out that a close bond is necessary for a strong mentor–mentee relationship, and that mentors and mentees need to be properly matched. If this is a weak link in the overall relationship, then the practice of matching mentors and mentees should be carefully executed to ensure compatibility between them. Therefore, before bonding takes place between the mentor and mentee, a period should be allowed to explore and determine their compatibility.

Finally, further research should look to assess the balance and effectiveness of power relations between mentors and mentees. This can be done in two contexts: one, in a formal school environment, and the other in informal or out-of-school contexts where a variety of learning strategies are employed. A comparative study could help to understand which of these two environments is more conducive to achieving targeted learning outcomes.

Additional information

Notes on contributors

Kwan Meng Lee

Lee Kwan Meng, PhD is a youth programme consultant (Programme Research and Publications) with the International Youth Centre, Kuala Lumpur, Malaysia. He has been involved in youth development programmes since 1975, but began his youth career in 1984 with the Malaysian Youth Council. This was followed with the Committee for ASEAN Youth Cooperation. He holds a doctorate in extension education and his doctoral thesis was on youth mentor learning. In addition to this doctoral degree and background in youth development, his other professional and academic interests include outdoor recreation where he has a Master of Science in Park and Outdoor Recreation from the Faculty of Forestry, Universiti Putra Malaysia (UPM), as well as a Master in Business Administration in International Management from Asia Pacific International Institute, New Zealand.

Steven Eric Krauss

Steven Eric Krauss, PhD is a research fellow with the Institute for Social Science Studies (IPSAS), UPM. He originates from the USA and has been living and working in Malaysia since 2001. Before joining IPSAS, he was an associate professor in the Department of Professional Development and Continuing Education in the Faculty of Educational Studies, UPM until 2010. He received his PhD from the Institute for Community and Peace Studies (PEKKA), UPM in 2005 in the field of youth studies. His research and teaching interests include positive youth development, youth work and Muslim youth religiosity and socialisation.

Turiman Suandi

Turiman Suandi, PhD is a professor in the Department of Professional Development and Continuing Education, Faculty of Educational Studies, UPM, and has served in the university for 30 years. He received his PhD in Extension Education from Ohio State University, Columbus, USA in 1991. His contributions in teaching, research and services are in the areas of extension education and human resource development, particularly leadership and motivation.

Azimi Hamzah

Azimi Hamzah, EdD is a professor and deputy director with the IPSAS, UPM. In addition, he also heads the Youth Development Studies Unit in the Department of Professional Development and Continuing Education, Faculty of Educational Studies, UPM. His specialisation is in youth resource development, adult education, extension education and continuing education. He received his Doctorate in Education from North Carolina State University, Raleigh, USA in Youth Resource Development and Adult Education in 1983.

Notes

^2. Email: [email protected]

^1. Email: [email protected]

^3. Email: [email protected]