Mentoring Support and Relational Uncertainty in the Advisor–Advisee Relationship

Mentoring, which is defined as a “communication relationship in which a senior person [i.e., mentor] supports, tutors, guides, and facilitates a junior person's [i.e., protégé's] career development” (Hill, Bahniuk, & Dobos, 1989, p. 15), is vital to graduate student success (Myers & Martin, 2008). Throughout their graduate programs, doctoral students participate in several different mentoring relationships with peers, professors, and departmental staff members (Luna & Cullen, 1998; Myers, 1995), and of these mentoring relationships, none is more important than the advisor-advisee relationship (Foss & Foss, 2008).

Once the advisor–advisee mentoring relationship has been initiated, graduate students expect their mentors to provide guidance and support (Cronan-Hillix, Gensheimer, Cronan-Hillix, & Davidson, 1986). They also expect their mentors to be honest and genuine, dedicated and loyal, empathic and compassionate (Cronan-Hillix et al., 1986), collegial, and involved (Schlosser, Knox, Moskovitz, & Hill, 2003). These imperative mentor behaviors facilitate graduate students' socialization into the academic department (Austin, 2002; Myers, 1998). Doctoral advisors' provision of mentoring support also enables students to complete their academic program and dissertation (Golde, 2005; Hepper & Hepper, 2003; Madsen, 1993), and it enhances students' research productivity (Green & Bauer, 1995), their perceptions of the academic climate (Kelly & Schweitzer, 1999), and their attitudes toward graduate school (Lyons, Scroggins, & Rule 1990). However, mentoring partnerships, such as the advisor–advisee relationship, involve relational conflicts, power inequalities, and performance expectations (Hawkins, 1991; Kalbfleisch, 2002), which may explain the reason advisees often experience uncertainty about their own interactions with their advisor (Foss & Foss, 2008).

Uncertainty refers to an “inability to predict behaviors, attitudes, or outcomes” or to interpret the meaning behind behaviors, attitudes, or outcomes (Afifi & Afifi, 2009, p. 1). According to Knobloch and Solomon (1999), individuals may experience three types of uncertainty: self, partner, and relationship. Self uncertainty refers to individuals' inability to describe, explain, or predict their own attitudes or behaviors; partner uncertainty refers to individuals' inability to describe, explain, or predict their relational partners' attitudes or behaviors; and relationship uncertainty refers to individuals' doubts about the status and the future of the relationship (Knobloch, 2008; Knobloch & Solomon, 1999, 2002). Knobloch and Solomon concluded that self and partner uncertainty includes aspects of desire (i.e., feelings and commitment), evaluation (i.e., value and definition), and goals (i.e., future objectives of the relationship) whereas relationship uncertainty includes behavioral norms (acceptable and unacceptable), mutuality (i.e., emotional reciprocity), definition (i.e., current status), and future (i.e., long-term outcomes).

Some researchers (e.g., Afifi & Weiner, 2004; Solomon & Knobloch, 2001) have suggested that uncertainty may ebb and flow throughout interpersonal relationships. However, in established relationships, the nature of uncertainty likely changes from self and partner uncertainty to relationship uncertainty (Afifi & Reichert, 1996; Knobloch & Solomon, 1999). Thus, most of the uncertainty research conducted to date has focused on relational uncertainty, not self and partner uncertainty. These studies indicated that uncertainty can be detrimental to interpersonal relationships (Dainton, 2003) as it limits the number of topics the partners discuss (Afifi & Burgoon, 1998) and may prevent them from developing and restoring relational intimacy (Emmers & Canary, 1996). Others have established a negative relationship between relational uncertainty and the use of relational maintenance behaviors in romantic relationships (Dailey, Hampel, & Roberts, 2010; Dainton, 2003; Dainton & Aylor, 2001) and in cross-sex friendships (although these findings are less conclusive) (Weger & Emmett, 2009).

Taken together, two general conclusions emerge from the mentoring and relational uncertainty research. First, doctoral students depend on their advisors to successfully complete their academic programs (Golde, 2005; Hepper & Hepper, 2003; Madsen, 1993). Second, doctoral students may experience uncertainty about their advisor–advisee relationships (Foss & Foss, 2008), which may, in part, be due to a lack of received mentoring support from their advisors.

Mentoring support typically applies to career mentoring or psychosocial mentoring. Career mentoring refers to behaviors intended to advance a protégé's job skill development. Psychosocial mentoring involves behaviors intended to enhance a protégé's confidence and self-perceived effectiveness (Kram, 1983, 1988). However, researchers (Hill, Bahniuk, Dobos, & Rouner, 1989; Schrodt, Cawyer, & Sanders, 2003) have identified several different types of career mentoring (i.e., research assistance, promotion, protection, and career support) and psychosocial mentoring (i.e., collegiality, friendship, collegial, social, and collegial task) provided to protégés.

Through research assistance, mentors work collaboratively on projects with their protégés to promote them within a field of study. Through promotion mentors offer strategies and suggestions intended to help their protégés achieve career goals, while protection behaviors safeguard them from people and situations exerting a negative impact on their burgeoning careers. Career support refers to mentor time devoted to giving advice and information about both formal and informal departmental policies. Mentors socializing with their protégés project collegiality, and those who provide support and encouragement to develop a trusting relationship exhibit friendship. Collegial social mentoring refers to socializing by sharing personal information, and in collegial task mentoring, protégés work jointly with mentors on projects (Hill et al., 1989; Schrodt et al., 2003).

Collectively, these eight types of mentoring support enhancement of protégés' skills, advance their careers, and make them feel accepted, appreciated, and confirmed. Thus, when doctoral students receive mentoring support from their advisors, they form favorable attitudes toward the relationship (Cavendish, 2007). By contrast, then, a lack of received mentoring support may be associated with unfavorable perceptions of the relationship, such as relational uncertainty, which in some cases contribute to doctoral students' decisions to discontinue their academic programs and their advisor–advisee mentoring relationships (Golde, 2005). To test these ideas in the context of the advisor–advisee mentoring relationship, we posit the following hypothesis: Advisees' received career mentoring and psychosocial mentoring from their advisors will be negatively related to their advisor–advisee relational uncertainty.

Students (N = 378, 148 males and 230 females with a mean age of 30.88 years (SD = 7.08) enrolled in PhD and EdD programs participated in this study. The participants had been enrolled in their current academic program between 1 and 100 months (M = 33.20, SD = 20.87), and they had been involved in their current advisor–advisee mentoring relationships between 1 and 12 years. The participants indicated that their advisors served as assistant professors (17.5%), associate professors (29.6%), and full professors (50.6%); most advisors (60.7%) were male.

We recruited participants using both convenience and volunteer sampling techniques via campus mail at a local university and via e-mail announcement messages sent to Communication Research and Theory Network subscribers. The participants completed the Academic Mentoring Behaviors Scale (AMBS) (Schrodt et al., 2003), the Mentoring and Communication Support Scale (MCSS) (Hill et al., 1989), and the Relational Uncertainty Scale (RUS) (Knobloch & Solomon, 1999). When scale items needed adjustments, we rephrased them to best reflect the advisor–advisee relationship. Table 1 contains the Cronbach's α coefficient, mean, and standard deviation for each scale used in this study.

The AMBS, a 15-item, five-factor instrument, measures protégés' reports of their mentors' provision of research assistance, protection, collegiality, promotion, and friendship. Responses were solicited on a 5-point Likert scale ranging from strongly disagree (1) to strongly agree (5). Reliability coefficients ranging from .65 to .82 have been reported for the five factors of this scale (Schrodt et al., 2003).

The MCSS, a 15-item, three-factor instrument, measures protégés' reports of their mentors' provision of career mentoring, collegial social support, and collegial task support. Responses were solicited on a 5-point Likert scale ranging from strongly disagree (1) to strongly agree (5). Reliability coefficients ranging from .65 to .86 have been reported for the three factors of this scale (Myers, 1998).

The RUS, a 16-item, four-factor instrument, measures the extent to which the respondents express uncertainty about the behavioral norms, mutuality, definition, and future of the relationship. In this study, we eliminated one irrelevant statement: whether or not this is a romantic or platonic relationship. Responses were solicited on a 6-point Likert-type scale ranging from completely or almost completely certain (1) to completely or almost completely uncertain (6). Reliability coefficients ranging from .73 to .89 have been reported for the four factors of this scale (Knobloch & Solomon, 1999).

We examined the extent to which doctoral students' received career mentoring and psychosocial mentoring from their advisors is related to their advisor–advisee relational uncertainty. In support of the hypothesis, a series of Pearson correlational analyses (see Table 2) indicated that advisees' reports of their advisors' provision of research assistance, protection, collegiality, promotion, and friendship as well as career, collegial social, and collegial task support showed a significant, negative correlation to all four dimensions of the advisees' relational uncertainty (i.e., mutuality, behavioral norms, definition, and future). That is, the more mentoring support doctoral advisees receive from their advisors, the more certain they tend to feel about their advisor–advisee relationship.

This finding not only provides practical implications for advisors, but also emphasizes the importance of continued examinations of the advisor–advisee mentoring relationship. Specifically, the negative association found between advisees' reports of received mentoring support and relational uncertainty bolsters the notion that advisees value mentoring support. Because relational uncertainty may discourage individuals from maintaining their relationships (Dindia, 2003), doctoral students who experience it may terminate their advisor–advisee relationship, often preventing them from completing their academic programs (Golde, 2005). To reduce doctoral students' uncertainty associated with their advisor–advisee relationships, advisors should provide mentoring support to their advisees. Advisors' provision of friendship, career mentoring, and collegial task support may be of particular importance as they are most closely associated with advisees' relational uncertainty. Moreover, the negative association found between advisees' relational uncertainty and their reports of received mentoring support corroborates Kram's (1998) claim that mentor effectiveness is a function of both career and psychosocial mentoring.

Effective mentoring relationships hinge on advisees' attempts to communicatively maintain their advisor–advisee relationships (Foss & Foss, 2008; Kalbfleisch, 2002) and warrant investigation in two areas. First, researchers need to examine the extent to which mentoring enactment theory (MET) (Kalbfleisch, 2002) can be used to explain doctoral advisees' use of relational maintenance behaviors with their advisors. MET explains the motivations of individuals entering into mentoring relationships, the ways they express interest in initiating mentoring relationships, and the reason for maintaining and repairing them.

The MET theory posits that those with positive past experiences as mentors will likely accept requests made by new protégés (Kalbfleisch, 2002). In academia, a doctoral student's choice of a particular faculty member as his or her advisor constitutes a major part of the educational journey (Hepper & Hepper, 2003). Thus, testing the propositions forwarded in MET that focus on the initiation phase of mentoring relationships may identify practical consideration for advisees looking to choose an advisor.

Second, future researchers might examine the extent to which advisees' relational characteristics (i.e., commitment, communication satisfaction, control mutuality, liking, relational satisfaction, and trust) relate to advisors' provision of mentoring support and the advisees' use of relational maintenance behaviors. These relational characteristics motivate partners to maintain their relationships (Stafford, 2003). Many researchers (Dainton & Stafford, 2000; Mansson, Myers, & Turner, 2010; Myers & Glover, 2007; Stafford & Canary, 1991, 2006; Weigel & Ballard-Reisch, 1999, 2001) have concluded that, across contexts, the use of relational maintenance behaviors, such as assurances, openness, positivity, shared tasks, and shared networks (and in some cases also advice and conflict management), is associated positively with relational characteristics. Moreover, advisees' reports of advisor–advisee relationships show that quality and satisfaction are functions of their advisors' mentoring support (Cavendish, 2007).

Limiting this study, we examined only the advisees' perspective, and students may not recognize all the mentoring behaviors that their advisors undertake on their behalf, such as promoting and complimenting their work to others in the department or in the larger field of study. Therefore, parties' perceptions of advisors' provision of career and psychosocial mentoring may differ. To address this concern, future researchers may want to examine the advisor–advisee relationship from the advisors' perspective and pair the data to examine the relationship between advisors' self-reported communicative behaviors and their advisees' perceptions of the relationship and vice versa.

Despite the one-sided viewpoint presented, our study corroborates and extends extant mentoring and relational uncertainty research, serving as a framework for future investigations. During the past four decades, a continuously growing body of mentoring research has focused on the positive outcomes associated with doctoral students' involvement in mentoring relationships while largely neglecting factors that may discourage doctoral students from maintaining their partnership with advisors. Because the advisor–advisee relationship is imperative to doctoral students' academic success (Hepper & Hepper, 2003) and poor advisor–advisee relationships lead to graduate student attrition (Golde, 2005), we encourage others to use these findings in investigations into additional factors that discourage advisees in their relationships with advisors.