Marcy Young Illies (Saint Cloud State University) Roni Reiter-Palmon (University of Nebraska Omaha)
One hundred and forty-six protégés with a mentor in their profession responded to a survey exploring how value similarity affects mentoring success (career support, psychosocial support, and satisfaction with the mentor) and organizational outcomes (organizational commitment, career success, and job satisfaction). Results revealed that protégés who perceived their values to be similar to those of their mentor had more mentoring success. Mentoring success correlated with more positive organizational outcomes. Finally, it was found that mentoring success mediated between perceived value similarity and important organizational outcomes (job satisfaction and organizational commitment).
values, mentoring, protégé outcomes, mentor satisfaction, mentor support
Accepted for publication: 15 January 2018 Published online: 1 February 2018
© the Author(s) Published by Oxford Brookes University
Mentoring is an important topic in organizations because it benefits the protégé, mentor, and the organization. Research has found mentoring to be related to positive behavioural, attitudinal, health, motivational, and relational outcomes as well as positive career outcomes for protégés (Eby, Allen, Evans, Ng, & Dubois, 2008). However, not all mentoring relationships result in positive outcomes (Burk & Eby, 2010; Eby, McManus, Simon & Russel, 2000; Washington & Cox, 2016). The benefits or positive outcomes that may come from mentoring may depend on the relationship that is formed between the protégé and mentor. One way to possibly enhance the success of the relationship may be to evaluate similarity between the mentor and protégé. Research indicates that similar characteristics between the mentor and the protégé may play a role in mentoring success (Eby et al., 2013; Ensher, Grant-Vllone, & Marelich 2002; Enshner & Murphy, 1997; Turban, Dougherty, & Lee, 2002). One characteristic that might be important is value similarity. Research indicates that a mismatch in values may affect mentoring success (Eby, Butts, Lockwood, & Simon, 2004). This research will attempt to show that value similarity between mentors and protégés will lead to greater mentoring success, which will ultimately lead to more benefits or positive outcomes.
Mentoring is defined as a relationship between a senior, more experienced person and a junior, less experienced person, where the senior member provides guidance and support to the junior member (Kram, 1984). Mentors provide protégés with the opportunity to learn and develop from more experienced individuals in the organization. Kram (1984) and Noe (1988) suggested that the career and psychosocial support provided to the protégé by the mentor helps the protégé succeed in the organization and possibly adjust much faster than employees without mentors. The career support function of mentoring includes helping protégés adjust to the organization by protecting protégés, giving protégés information to orient them to the organization, and giving protégés assignments to improve skills and be visible to others. The psychosocial aspect of mentoring includes helping protégés adjust at a more personal and individual level. Mentors befriend the protégé and socialize with him or her. The mentor also provides emotional support and is a role model for the protégé. These benefits highlight the learning and support aspects that a protégé receives; however, the benefits of mentoring, particularly for protégés, extend further than just support and learning the ropes. Others found that outcomes associated with mentoring may include promotions, higher salaries, and increased career satisfaction (Kram, 1984; Scandura, 1992).
A meta-analysis done by Allen, Eby, Poteet, Lentz, & Lima (2004) indicated that mentored individuals reported more positive work outcomes than non-mentored individuals. Mentored individuals reported higher compensation and more promotions, were more satisfied with their careers, and were more committed to their jobs. These results are similar to a meta-analysis done by Underhill (2005) that found that protégés reported increased job satisfaction, increased self-esteem, perceptions of career advancement opportunities, higher organizational commitment; and decreased work stress, and work-family conflict. More specifically, researchers have found mentoring to positively affect career success (Blickle, Witzki, Schneider, 2009; Higgins, Dobrow, Chandler, 2008; Kammeyer, Mueller & Judge, 2008; Peluchette & Jeanquart, 2000; Turban & Dougherty, 1994; Wallace, 2001) job satisfaction (Lankau & Scandura, 2002; Robinson & Reio, 2012; Seibert, 1999) and organizational commitment (Joiner, Bartram, & Garreffa, 2004; Lankau & Scandua, 2002; Payne & Huffman, 2005; Robinson & Reio, 2012; Seibert, 1999). These will be the outcomes on which this paper will focus.
Research has suggested that the support (career and psychosocial) that protégés receive may affect personal and organizational outcomes. For example, Joiner et al. (2004) found that mentoring support correlated with intentions to leave the organization; those with less mentoring support planned on leaving the organization in the near future. Mentoring support was also associated with perceived career success; those who had more support felt they were more successful. The association between mentoring support and certain outcomes may be clearer when looking at the functions associated with mentor support. For example, helping protégés adjust to the norms of the organization would possibly prevent them from leaving the organization. However, it is also possible that less obvious aspects of support such as befriending a protégé may also affect certain outcomes such as organizational commitment. Overall, protégés who receive smaller amounts of support may ultimately receive fewer benefits than those that received larger levels of support.
However, research suggests that mentoring support may be somewhat contingent on the quality or satisfaction of the mentoring relationship (Eby & McManus, 2004; Feldman, 1999). For example, protégés may not obtain the psychosocial and career support necessary if the relationship is of poor quality. A meta-analysis by Eby et al (2014) addressed these three constructs finding that relationship quality/satisfaction correlated positively with instrumental/career support and psychosocial support indicating that those with higher quality relationships received more support or vise versa, those receiving more support felt the relationship was more satisfactory.
Researchers have also explored unsatisfying mentoring relationships, termed “negative mentoring” (Burk & Eby, 2010; Eby et al., 2004; Eby, Durley, Evans, & Ragins, 2008; Washington & Cox, 2016). Eby et al. (2000) suggested that negative mentoring relationships decrease the benefits of mentoring and cause emotional and social drawbacks, as with any dysfunctional relationship. Eby et al. (2004) suggested several things that may contribute to negative mentoring experiences such as distant behavior, manipulative behavior, lack of mentoring expertise, and mismatch of dyads. They found that a mismatched dyad was the most commonly reported factor of negative mentoring. A mismatch between protégé and mentor could be caused by dissimilarities in personality, values, jobs, gender, ethnicity, age, or many other individual differences.
Researchers have known for years that similarities among relationship participants may have a positive effect on the relationship. Numerous studies have addressed similarity and relationship satisfaction (Blum & Mehrabian, 1999; Guant, 2006; Shiota & Levenson, 2007; Russell & Wells, 1991). Researchers have suggested that certain aspects of similarity may be more beneficial than others. For example, Gaunt (2006) suggested that similarity in personality and values is more important than attitude and religious similarity in spousal relationships. However, Shiota and Levenson (2007) and Gattis, Berns, Simpson, and Christensen (2004) found that personality similarity may not predict marital satisfaction. Overall the research indicates that similarity is important, but the specific characteristics that need to be similar are debatable.
Along the lines of relationship research, attraction researchers have suggested that perceived similarity among relationship participants is one of the most important predictors of attraction (Fitness, Fletcher, & Overall, 2003). In fact, perceived similarity seems to be more important than actual similarity in certain aspects of relationship quality. Research has shown only a weak relationship between actual similarity in attitudes/personality and relationship quality (Karney & Bradbury, 1995; Strauss, Barrick, Connerley, 2001). For example, Strauss, Barrick, and Connerley (2001) examined actual and perceived similarity in performance ratings of peers and supervisors. The results of their study showed that actual personality similarity did not relate to performance ratings, however perceived personality similarity was strongly related to performance ratings. One reason perceived similarity may be more important than actual similarity is that people make assumptions and decisions based on their perception rather than actuality. When assessing character similarity, people do not have ready access to actual data. Perceived similarity may be important in relationships because individuals see familiarity as being safe, whereas novel stimuli may unconsciously be seen as dangerous (Berscheid & Reis, 1998). Similarity may also be important for self-evaluation. Tesser (1988) suggested that individuals need to have a positive evaluation of themselves, and their relationships with others influences this self-evaluation. Specifically, others who are similar to oneself may help maintain a positive self-evaluation.
Researchers have evaluated similarities in mentoring relationships for the purpose of creating more productive relationships. These include evaluating characteristics such as gender similarity (Noe, 1988; Ragins & Cotton, 1999; Scandura & Williams, 2001; Sosik & Godshalk, 2000) and racial similarity (Hu, Thomas, & Lance, 2008; Johnson-Baily & Cervero, 2004; Oritz-Walters & Gilson, 2005; Thomas, 1990) and similarity on more personal characteristics such as personality, values, and attitudes. The research on demographic similarity (race and gender) is somewhat mixed. While some studies have found racial and or gender similarity among mentor and protégé beneficial other studies have not. In general, demographic similarity is harder to study as there are fewer women and people of color in mentoring roles. The benefits of demographic similarity may largely depend on the organization and goals of the mentoring relationship.
Research addressing, characteristics such as personality, attitude, values, and decision style (Ensher, et al., 2002; Enshner & Murphy, 1997; Hernandez, Estrada, Woodcock, & Schultz, 2017; Hirschfeld, Thomas, & Lankau, 2006; Hu, Baranik, & Wu, 2014; Lankau, Riordan, & Thomas, 2005; Menges, 2016; Mitchell, Eby, & Ragins 2015; Owen and Solomon 2006; & Turban, et al., 2002) seems more promising. For example, Enshner and Murphy (1997) found that protégés who perceived themselves to be similar to their mentors in terms of outlooks, perspectives, and values, tended to be more satisfied with the mentoring relationship. Turban et al. (2002) found that protégés who perceived themselves as similar to their mentors (defined as similar perceptions, working styles, values, attitudes, and career aspirations) felt they received more mentoring (career and psychosocial support) than those who did not perceive themselves as similar. Last, Ensher, et al. (2002) found that, in dyads that had similar attitudes, protégés reported more vocational support, psychosocial support, and role modeling than dyads that did not have similar attitudes.
Although these studies suggest that personal characteristics may affect mentoring relationships, particularly from the perspective of the protégé, many of these studies tend to combine various constructs such as personality, attitude and values together into one overall similarity measure. Combining constructs makes it impossible to determine the effect a specific personal characteristic has on the mentoring relationship. Knowing more about a specific personal characteristic is important for furthering our understanding of how a specific characteristic affects mentoring relationships and allows us to make more advances in the field of mentoring.
Researchers have expressed concern for combing constructs specifically in reference to personal values, explaining that values seem to have many different definitions and there seems to be many concepts that are interchanged with values, making this research less clear and concise (Rohan, 2000; Wach & Gosling, 2004). Wach and Gosling (2004) addressed the difference between values, attitudes and interests, indicating that interests are more narrowly defined than values, which are more abstract. Values also take into account society’s influence rather than just personal preference as is the case with interests (Sagiv, 2002). Values are also more central than attitudes, which can change more easily. As Schwartz (1992) would suggest, a person may have many more attitudes than values. In general, values seem to be more fundamental to our motivations and are broader in perspective allowing for many different interests or attitudes to stem from the same value. The central nature of values may allow individuals to make more holistic comparisons to others, suggesting that values may be an important topic to address singly among protégés and mentors.
Rokeach (1973) defined a value as “an enduring belief that a specific mode of conduct or end-state of existence is personally or socially preferable to an opposite or converse mode of conduct or end-state of existence” (p. 5), indicating that values are internal, affect our behavior, and are affected by people and society. Rokeach (1973) suggested that people typically have few values that are similar, though the importance they place on those values varies. Schwartz (1992) furthered Rokeach’s work, suggesting that people’s values can be combined into categories based on basic needs, interaction with people, and survival within a society. Knafo and Schwartz (2004) argued that values are guiding principles in one’s life that are used to “select and justify actions and to evaluate people and events” (p. 440).
Research has shown that values play a role in mentoring relationships, particularly the selection of protégés or mentors. For example, research has indicated that in many instances mentors prefer protégés who show potential and drive to achieve because they perceive these protégés will bring more to the relationship and that the benefits will be greater for them, the mentor (Allen, 2004; Allen, Poteet, & Russell, 2000). However, research has shown that under certain circumstances, some mentors may prefer protégés that need their help over protégés who have great potential (Allen, 2004; Allen et al., 2000). This may suggest that mentors have certain values such as achievement or benevolence and look for ways to fulfill these needs through their protégés.
Values may be one characteristic where similarity is particularly important for successful mentoring relationships. Eby et al. (2004) suggested that a mismatch of values might lead to a negative mentoring experience, such as exploration, sabotage, deception, and communication difficulties. Hale (2000) suggested that in order to optimize learning in mentoring relationships, one must have some contrast in areas of behavioral style, learning style, strengths, and developmental needs. However, he also suggested that some similarities are also needed, particularity in the area of values, beliefs, and life goals.
One reason that values may be so important in the mentoring relationship is that values serve as a foundation for how we view people and situations. Those with similar values may be in agreement in how they evaluate people and situations. For example, those who value honesty may see people and situations differently than those who values craftiness. Similarly, Schein (1985) suggests that values may be related to communications systems. Those with similar values may have an easier time communicating, knowing what needs to be done and how it should be done and experience less frustration working together because they would evaluate things in a similar manner. In all, the congruence found by similar values may help build stronger interpersonal relationships. Mullen and Noe (1999) suggested that developing a strong bond is extremely important in creating a successful mentoring relationship. Thus, value congruency may help in developing a deeper bond between the protégé and the mentor than cannot be achieved with other forms of similarity, such as gender or interests.
This research will attempt to show that value similarity between mentors and protégés will lead to greater mentoring success (i.e., more career and psychosocial support and more satisfaction), which will ultimately lead to better career outcomes for the protégé (see Figure 1). The following hypotheses reflect the predicted model: (1) protégés with values perceived to be similar to their mentors’ values will report more mentoring success (i.e., more career and psychosocial support and more satisfaction with the mentor) than protégés with values perceived not to be similar to their mentors’ values, (2) protégés who perceive higher levels of mentoring success will have better organizational outcomes (i.e., organizational commitment, perceived career success, job satisfaction) than those with lower levels of mentoring success, (3) mentoring success will mediate the relationship between value congruency and the outcome variables.
Participants were professionals who had a mentor in their professional career and have/had been in that mentoring relationship for at least 6 months. One hundred and forty-six participants met the study requirements and completed the study. Participants were 19 and older with the average age of 36.16 (SD = 10.00). Most participants were female (70.5%), and most were Caucasian (90.4%), following by Asian (3.2%), Hispanic (2.5%), African American (1.9%), and two unidentified.
Participants were contacted through (a) a university alumni on-line newsletter, (b) social network pages, (c) e-mail, (d) postal mail, (e) university distribution list, and (f) the snowball approach. Analyses of variance indicated that there were no differences in dependent or mediator variables due to recruitment source. Demographic information such as age, company tenure, gender, and race was collected from all participants.
An on-line survey platform was used to collect data. Participants were told that the study was about evaluating mentoring relationships and that their responses would be confidential. Some participants were also asked to forward the recruitment information to others in their social or professional network.
Participants were given a definition of mentoring and asked if they had at least one person they perceived as a mentor in their career/profession. They were allowed to respond to this with a yes or no. Participants were then told to consider their current or most recent mentor and respond to questions with that particular mentor in mind. Participants also provided some demographic information about their mentoring relationship: (a) how long they had been in the relationship, (b) if the relationship was currently on-going, and (c) if it ended, and if so, how long ago it ended. Data were only used for participants who were in a current mentoring relationship or who had ended a mentoring relationship less than five years prior to the survey.
Job satisfaction was measured using the Hackman and Oldham (1974) general job satisfaction survey. The survey includes three items (e.g., “Generally speaking, I am very satisfied with this job”). Participants were asked to answer questions during the time they were mentored. Items were measured on a seven-point scale. The items had good internal consistency, a = .94.
Perceived career success was measured with a four-item measure developed by Turban and Dougherty (1994). Sample items include “how successful has your career been” and “compared to your co-workers, how successful is your career.” The first three items were measured on a five-point response scale. The last question asked participants to gage their career success as either above, behind, or on-schedule. The questions were combined into one scale using standardized scores, a = .80. Participants were asked to answer questions about perceived career success during the time that they were mentored.
Organizational commitment was measured using the short form organizational commitment scale developed by Mowday, Steers, and Porter (1979). The short form consists of nine items that relate to commitment behaviorus and attitudes. An example question is “I really care about the fate of this organization.” Items were measured on a seven-point response scale. Internal consistency for this study was good, a = .94. Participants were asked to answer the nine items during the time that they were mentored.
Support was measured using a modified version of Thomas’ (1990) mentoring support scale. The measure is composed of 11 items that assess career and psychosocial support. Six items assessed career support and had good internal consistency, a = .87. Five items assessed psychosocial support and also had good internal consistency, a =.81. An example of a psychosocial support question is “To what extent did the person act as a professional role model for you?” An example of a career support question is “To what extent did the person advocate for you to get a promotion?” The measure uses a five-point response scale. Participants were asked to respond to these questions with the mentor they chose in mind.
Satisfaction with the mentor was measured with a scale modified from Enshner and Murphy (1997). The measure is composed of three items with five-point response scales. An example question is “I feel satisfied with my mentor.” For this study, internal consistency was high, a = .90. Again, participants were asked to respond to these questions with the same mentor they chose in mind.
Values were measured using a modified version of the Lindeman and Verkasalo (2005) Short Schwartz Value Survey. A structured value scale was used to define more clearly specific values and allow for more precise measurement. This is an improvement over previous studies that evaluated value similarity without actually measuring specific values. The survey consists of 10 values: power, achievement, stimulation, self-direction, universalism, benevolence, tradition, conformity, security, and hedonism. Participants were given a definition of each value. An example is “Power, that is, social power, authority, wealth.” Participants were asked to rate each of the values with a six-point scale ranging from 0 (not important) to 5 (of supreme importance), which was modified from Lindeman and Verkasalo’s original 7-point scale which included -1 (being against my principles). The modification was done to eliminate negative numbers for difference score comparisons of values. Participants were asked to rate what they perceived their mentor’s values to be using the same survey. Perceived value similarity was used over actual value similarity because research has shown that in some instances, particularly with matching, perceptions may be more important than reality (Edwards, Cable, Williams, Lambert & Shipp 2006). To determine value congruency, difference scores were calculated between the protégé’s value and the perceived mentor’s value. Protégé scores were subtracted from mentor scores and then squared to eliminate negative numbers. The difference scores were then averaged to create on overall value difference score between protégé and mentor. A linear transformation was done by subtracting the maximum possible value difference from each overall difference score so that larger scores indicated similarity rather than difference in values.
Descriptive statistics and correlations were analyzed for the variables of the model portrayed in Figure 1 (see Table 1). T-tests indicated that gender was the only demographic variable related to value similarity. Females had values more similar to their mentor than did males, t (145) = -2.13, p = .037, d = -0.56. Hypothesis one, stating that protégés who perceived that they had values similar to those of their mentor would report more mentoring success, was supported. Protégés whose values were perceived to be similar to those of their mentor received more psychosocial support, r (146) = .30, p <.001; were given more career support, r (146) = .18, p =.029; and reported more satisfaction, r (146) = .26, p <.001. These variables were also inter-correlated, indicating that protégés who experienced one aspect of mentoring success also experienced the others (see Table 1).
Variables
Mean
SD
2
3
4
5
6
7
Value similarity
5.76
1.39
.26*
.18*
.30*
.07
.10
.06
Satisfaction with mentor
4.42
0.75
-
.57*
.75*
.25*
.36*
.27*
Career support
3.87
0.92
.63*
.32*
Psychosocial support
4.20
0.72
.34*
.20*
Org. commitment
5.78
1.17
.59*
.23*
Job satisfaction
1.16
.24*
Perceived career success +
0.00
0.80
Note. n = 145, + z score
*p < .05
Hypothesis two stated that protégés who perceived higher levels of mentoring success would have more beneficial organizational outcomes than those with less mentoring success. This hypothesis was also supported. Protégés with more career support reported higher job satisfaction, r (146) = .36, p <.001; higher organizational commitment, r (146) = .26, p <.001; and more perceived career success, r (146) = .32, p <.001. Protégés with more psychosocial support also reported higher job satisfaction, r (146) = .36, p <.001; higher organizational commitment, r (146) = .34, p <.001; and more perceived career success, r (146) = .20, p <.001. Finally, protégés who perceive higher levels of satisfaction with their mentor had higher organizational commitment, r (146) = .25, p <.001; higher job satisfaction, r (146) = .36, p <.001; and more perceived career success, r (146) = .27, p <.001.
To analyze if mentoring success mediated the relationship between value congruency and the outcome variables a multiple mediation analysis was performed. Mediation tests the effects a predictor variable has on a criterion variable through an intervening variable (see Figure 1). Multiple mediated analyses allow researchers to (a) test if multiple mediator variables jointly mediate between the predictor and criterion variables, and (b) determine to what extent each individual mediator variable mediates between the predictor and criterion variable (Preacher & Hayes, 2008).
Mediation analyses were performed using Preacher and Hayes (2008) indirect macro program. To determine if mediation occurred, Preacher and Hayes’ bootstrapping procedure was used. This procedure estimates direct and indirect effects. Indirect effects address how variable X affects variable Y through the mediator. Direct effects address how variable X directly affects variable Y (see Figure 1). Bootstrapping uses a non-parametric resampling procedure that is appropriate to use with a small sample size and is more robust to non-normal samples. The procedure tests if the direct effect of a variable on another variable is different from zero and if the indirect effect of a variable on another variable is different from zero. For this study, we were interested in only the indirect effects as there was little evidence to suggest a direct relationship as the correlations between the outcome variables and value similarity were not significant (see Table 1). Ten thousand bootstrapped estimates were requested and each outcome variable (job satisfaction, organizational commitment, and perceived career success) was analyzed individually. Bias corrected and accelerated bootstrap confidence intervals were used as recommended by Preacher and Hayes (2004) and Efron (1987). Due to the fact that gender was related to value similarity, gender was used as a covariate for all analyses.
In multiple mediation, the effects of variable X on variable Y is assessed through many mediators. This analysis takes into account the relationship that the mediators have with each other. Testing the total indirect effect of the combined mediators eliminates the need to parse out the effect that one mediator has over another. However, analysis can be done to determine the effect of each unique mediator. Preacher and Hayes (2004) suggest testing the total indirect effect of the combined mediators and the indirect effect of each specific mediator. Testing the indirect effect of a specific mediator in a model that also contains additional mediators is not the same as testing the effect of a mediator in a single mediated model, as the model controls for the other mediators. Preacher and Hayes (2008) noted that strong relationships among mediators may attenuate analyses of each specific mediator. For this study, additional regression analyses were analyzed to test for collinearity among the meditator variables. Collinearity tolerances for the mediator variables with each dependent variable were within the acceptable standards, ranging from .38 and .59 (a tolerance value less than .10 is often considered to indicate collinearity is a problem [Cohen, Cohen, West, & Aiken, 2003]).
The total mediated model analyzing the effect of value similarity on job satisfaction through mentoring success was significant (see Table 2). Value similarity affected mentor support and satisfaction with the mentor which then affected job satisfaction. When considering the individual mediators, only career support was found to mediate between value similarity and job satisfaction when controlling for the other mediators (see Table 2).
The total model analyzing the effect of value similarity on organizational commitment through mentoring success was also significant (see Table 2). The model indicated that value similarity affected organizational commitment through its effect on career support, psychosocial support, and satisfaction with the mentor. When evaluating the unique contributions of each mediator, only psychosocial support was found to mediate between value similarity and organizational commitment (see Table 2).
The last model analyzed the effect of value similarity on perceived career success through mentoring success (see Table 2). The total model was not significant. Therefore, value similarity did not mediate through the combined mediators of psychosocial support, career support and satisfaction with the mentor. Even when a total model is not significant, Preacher and Hayes (2008) suggested that specific mediators can still be analyzed. When considering the individual mediators in this analysis, career support was found to mediate between value similarity and perceived career success (see Table 2), indicating that value similarity may affect perceived career success through career support when controlling for the other mediators.
Dependent Variable
Indirect Effects
Bootstrapping
BCa 95% CI
Point Estimate
Standard Error
Lower
Upper
Job Satisfaction
Psychosocial Support
0.029
0.040
-0.045
0.117
Career Support
0.031
0.023
0.001
0.099
0.036
0.033
-0.004
0.156
Total
0.091
0.039
0.032
0.192
Organizational Commitment
0.082
0.051
0.014
0.234
0.016
0.022
-0.012
0.077
-0.007
-0.101
0.037
0.186
Perceived Career Success
-0.027
-0.104
0.021
0.004
0.024
-0.002
0.098
-0.006
0.093
Note. Bca = bias corrected and accelerated, CI = confidence interval, 10,000 bootstrap samples.
The results indicated that protégés who perceived their values to be similar to their mentors’ values experienced more mentoring success, meaning they experienced more career support and psychosocial support and were more satisfied with their mentors. This result extends previous research that indicates value similarity is important in mentoring relationships (Eby et al., 2004; Mitchell et al., 2015) by exploring personal values directly. This finding reveals that value similarity, aside from other aspects of similarity, may help develop successful mentoring relationships and that it is an important variable for organizations to consider when developing mentoring relationships. It may also be a variable that organizations could analyze if they are finding that the current mentoring relationships in their organizations are unsuccessful.
Results also indicate that protégés who specified higher levels of mentoring success were more committed to their jobs, had more perceived career success, and experienced more job satisfaction. This result supports previous research indicating that mentoring is beneficial to an organization and that mentoring can lead to additional protégé benefits (Allen et al., 2004; Underhill, 2005). The results may also indicate that protégés who do not receive support (career and/or psychosocial) and who are not satisfied with their mentor may not reap these additional organizational outcomes. Therefore, these outcomes are contingent on a mentoring relationship that results in support and satisfaction and organizations need to understand how to help protégés and mentors develop supporting and satisfying mentoring relationships. They may also need to understand the relationship between support and satisfaction and specific outcomes.
Results show that mentoring success mediated between perceived value similarity and job satisfaction. Those who perceived themselves to be similar to their mentor had more job satisfaction through mentoring success. The results indicate that career support was the mediator that is most important in this relationship. Other researchers have also found that those who receive career support, along with other types of support, are more satisfied with their jobs (Eby et al., 2013). It may be that aspect of career support such as learning the ropes and being given key assignments was the most important factor leading to job satisfaction. This may be especially true of those in professional careers where more of an emphasis is placed on professional development and advancement.
Mentoring success also mediated between perceived value similarity and organizational commitment. Protégés who perceived themselves to be more similar to their mentor experienced more organizational commitment through mentoring success. Psychosocial support was the most important mediator in this relationship, indicating that aspects of psychosocial support such as befriending protégés seem to lead to more organizational commitment. Results suggest that organizations that are interested in using mentoring relationships to obtain more organizational commitment may want to specifically focus on developing strong psychosocial support between the mentor and the protégé.
Mentoring success did not mediate the relationship between perceived value similarity and perceived career success. However, career support alone did mediate in this relationship. It seems fitting that career support, which addresses tasks such as giving protégés assignments, would be the aspect of mentoring that would be most associated with perceived career success.
The results of this study extend previous research findings by further solidifying certain relationships among mentoring variables. This study also adds a new element to the mentoring research by evaluating value similarity independently of additional variables such as interests, personality or attitudes, a problem mentioned by Rohan (2000) and Wach and Gosling (2004). The evaluation of values independent of other variables helping us better understand the impact of value similarity in mentoring relationships. The study also incorporates a structured scale of values. By using such a scale, protégés and mentors may be more accurate in reporting their values and more information can be obtained about specific value similarity and dissimilarity. Although research has indicated that value similarity is helpful in creating successful mentoring relationships, little research has been done to explore similarity among specific values. This study may also be the first step in supporting a process through which organizations can pair mentors and protégés based on value similarity. This may be especially important for organizations where mentors are assigned a protégé and there is no basis for making that pairing.
This study also helps to support a framework for thinking about how mentoring variables relate to each other. This study helps explains how perceived value similarity affects organizational outcomes through mentoring success. More specifically, the study indicates which mentoring success variables seem to be most important in mediating the relationship between value similarity and each outcome, giving organizations and researchers more information on how mentoring relationships affect outcomes.
Although this research advances our knowledge of mentoring relationships, it does pose some limitations and needs for future research. A possible limitation is the use of multiple sampling methods. However, a heterogeneous participant pool also has its benefits in that it should increase the generalizability of the results. Future research may want to consider replication in additional samples to determine if different values need to be similar among different industries, occupations, and ages, in order to have effective mentoring relationships. This study also used difference scores to calculate protégé and mentor value similarity. Researchers have suggested that difference scores may be limiting and that other methods such as polynomial regression may be more effective. However, polynomial regression requires a larger sample size than what was available for this study in order to obtain adequate statistical power (Edwards, 2001).
Although this study addressed some important initial questions about perceived value similarity, additional research in the area of value similarity and mentoring should be considered. This may include studies on actual value similarity. Although perceived value similarity was the purpose of this study, and researchers argue that the perception of similarity may be more important than actual similarity, particularly in subjective matters (Edwards et al., 2006), there may be benefits to studying actual value similarity. Research that addresses both actual and perceived value similarity may be able to analyze how closely perceived value similarity is to actual value similarity. In addition, this study only addressed perceived value similarity from the perspective of the protégé and not from the perspective of the mentor. Future studies may want to evaluate perceived value similarity from the perspective of the mentor. It is possible that perceived similarity from the perspective of the mentor may account for greater differences in support and may also affect protégé outcomes.
Finally, future research should explore if there are negative consequences of value similarity. It is possible that there is a threshold for value similarity such that too much similarity is detrimental to a relationship. Too much value similarity may lead protégés and mentors to think and perceive in such similar ways that growth and development are hindered. In general, this study takes the first steps in analyzing how values affect mentoring relationships. The results indicate a relationship between perceived similarity and mentoring success (support and satisfaction), which ultimately affects protégé outcomes. This initial analysis of how values affect mentoring will hopefully lead to additional research that may further our understanding of the effect of values on mentoring relationships.
Marcy Young Illies teaches in the Psychology Department at Saint Cloud State University
Roni Reiter-Palmon is the Varner Professor of Industrial Organizational (I/O) Psychology and the Director of the I/O Psychology Graduate Program at the University of Nebraska Omaha