Abstract
The purpose of the present study was to determine whether students’ views of mentors have changed as a function of the increased number of faculty members conducting research and the inclusion of undergraduate students in faculty mentors’ research teams, using reports from current students. The participants were 227 undergraduate, master’s, and doctoral students at two large, Western public universities located in the United States. One institution was a research-intensive university and the other was not. Students were asked to complete a questionnaire about whether they had a mentor, the characteristics of their mentors, and their perceptions of their mentors. The findings indicated that 28.5% of undergraduates and 95% of graduate students had mentors. Undergraduate students were significantly more likely to choose mentors for being inspiring instructors, and graduate students were significantly more likely to choose mentors because of interest in their research. The most important characteristic of both good and bad mentors was personality. Students at all levels perceived their mentors as very interested in their futures. Mentor satisfaction was high among students at all levels. The findings are encouraging, and they provide evidence that psychology has adapted well to the increased number of faculty conducting research and to the inclusion of undergraduate students in research.
Psychology has a long history of mentoring, beginning in 1879 when Wilhelm Wundt established the first psychological laboratory (Balance & Evans, 1975). Wundt oversaw the theses of 186 doctoral students (Tinker, 1932)—116 of whom were studying psychology (Fernberger, 1933)—and mentoring has been remained a staple of educational training in psychology since. Given the commonality of mentoring in psychology, it has been studied by investigators over the past three decades (Clark, Harden, & Johnson, 2000; Petrella & Jung, 2008).
The long history of mentoring in psychology has included various shifts in its practice. Before 1990, many higher education public institutions that were not considered research-intensive institutions expected faculty to focus their work on teaching and mentoring students. When the economic recession occurred in the 1990s, states were forced to cut their funding for higher education. This forced university administrators, particularly those in nonresearch-intensive institutions, to seek alternative sources of funding (Zusman, 2005) such as extramural research support and fundraising in the community. The need for alternative sources of funding was one factor that changed the role of many faculty members who were hired for tenure-track positions (Litwin, 2009; Tien, 2008). Newly hired faculty members were expected to conduct focused programs of research that were supported by their research grants; research grants were expected to include funding for student research assistants. University faculty were also encouraged to offer academic credit for nonfunded students to work in their research programs. Having students involved in research was considered mutually beneficial. As a result of these changes, research has become one of the most important functions of universities, particularly for research-intensive institutions but also for some nonresearch-intensive institutions. Increasing research activity has increased the number of faculty who serve as research mentors (Cornelius, Wood, & Lai, 2016; Tien, 2008).
By mentoring students in their research, faculty members were able to be more productive and, as a result, could provide an array of opportunities for their students. Students were able to gain hands-on experience, learn about research methods and techniques, and, in some cases, author research presentations and publications. Thus, the role of a mentor in many institutions changed from offering pragmatic advice about a mentee’s strengths and weaknesses (Kwan & Lopez-Real, 2006) to a more dynamic and reciprocal relationship (Biaggio, 2001; Galbraith & James, 2004). As a result of this change, university retention rates for those involved in research increased (Nagda, Gregerman, Jonides, von Hippel, & Lerner, 1998), and involvement in research was related to higher grade point averages (Campbell & Campbell, 2000).
Graduate Students
Mentoring of graduate students is an important aspect of graduate psychology training because it is related to increased research productivity and enhanced satisfaction with doctoral programs (Clark et al., 2000; Hollingsworth, Fassinger, & Hansen, 2002). Clark, Harden, and Johnson (2000) examined the prevalence of mentoring in a psychology graduate setting. They found, in a retrospective survey of PhD graduates in clinical psychology, that 71% of the respondents reported having a mentor during graduate school.
Other researchers have examined graduate student–mentor relationships. Johnson, Koch, Fallow, and Huwe (2000) conducted a retrospective study to examine psychology doctoral students’ perceptions of their mentoring relationships. Mentored students were asked to use a 5-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree) to rate their level of agreement with five functions of mentors. They found that students rated the order of importance as: mentor acceptance, direct training, role modeling, providing opportunities, and sponsorships. Similarly, Clark et al. (2000) found that doctoral students listed the most important characteristics of mentors as supportive, intelligent, and knowledgeable. Although the majority of the participants did not have negative experiences with their mentors, the most negative aspects of mentoring were lack of availability, difficulty in dealing with the termination of their relationship, and an inability to meet their mentors’ expectations.
In another retrospective study, clinical psychology graduates were asked to rate their relationships with their mentors. Ninety-one percent of the respondents rated their relationship with their mentor as positive, and mentored respondents reported that a mentoring relationship had greater importance than did nonmentored respondents (Clark et al., 2000). Taylor and Neimeyer (2009) found that students’ overall satisfaction with their mentoring relationships was positively related to support from their mentors. However, the level of student satisfaction with mentoring relationships decreased as the number of years with the mentor increased. Other researchers found that those with mentors reported greater satisfaction in their graduate programs and a higher likelihood of mentoring others in the future than those who were not mentored (Johnson, Koch, Fallow, & Huwe, 2000).
Researchers have also examined psychology graduate students’ views of mentoring (Cronan-Hillix, Gensheimer, Cronan-Hillix, & Davidson, 1986). Fifty-three percent of the students had mentors. The most important characteristics of a good mentor listed in order of importance were interested/supportive, personality characteristics, knowledgeable/competent, sharing/giving, and not exploitative, involved in research/resourceful and attitudes toward students. The most important characteristics of a bad mentor listed in order of importance were uninterested/unsupportive, personality characteristics, lack of knowledge/incompetent, exploitative, attitudes toward students, and unavailable/inaccessible.
Undergraduate Students
Researchers have also examined the effects of mentoring among undergraduate students. Kardash (2000) found that undergraduates who were mentored in research had stronger professional and critical thinking skills than those who were not mentored in research. In addition, Petrella and Jung (2008) found that undergraduates involved in research reported a better understanding of research, improvements in data interpretation, and a better ability to apply scientific concepts. Mentored students also reported increased confidence and independence.
Undergraduate students who had research mentors reported that the most valuable element of their research experience was the relationship they developed with their mentor (Malachowski, 1996). Koch and Johnson (2000) mailed surveys to university alumni who were involved in undergraduate research. A total of 36% of the alumni responded. Mentored respondents reported a greater satisfaction with their psychology program than those who were not mentored. Thus, undergraduate students indicated benefits from being mentored in research.
Research on mentorship in psychology has been chiefly retrospective, conducted using either graduate or undergraduate students only. The purpose of the present study was to assess current students’ views of mentors including undergraduate-, master’s-, and doctoral-level students. In this study, we sought to explore the dynamics surrounding the predictors and effects of mentorship across educational levels and evaluate the traits that students perceived as most critical to their perceptions of their mentors as being “good” or “bad” mentors.
Method
Participants
The participants were 227 students majoring in psychology, 186 women and 41 men, enrolled at one of the two large Western universities during the 2017 academic year. The mean age was 24.05 years (range 18–56). The participants were undergraduate (n = 161), master’s (n = 31), and doctoral (n = 35) students. Among students enrolled in graduate programs, 44% were enrolled in clinical, 5% in developmental, 11% in learning and cognition, 19% in social, 8% in industrial/organizational, and 13% other.
Measures
A Qualtrics© (2018; Qualtrics, Provo, UT) Questionnaire was developed, based on the measures used in the Cronan-Hillix, Gensheimer, Cronan-Hillix, and Davidson (1986) study to assess psychology students’ involvement in research, perceptions of mentoring, and their relationships with their mentors. Students were asked to provide the following demographic information: major, university, sex, ethnicity, age, employment and educational status, the number of years worked on their current degree, and whether they had a mentor.
Students who had a mentor were asked a series of questions, including responding “yes” or “no” to a list of six reasons for choosing their mentor or to indicate the reason if it was not listed. Students were also asked what a mentor meant to them, the sex of their mentor, how many years their mentor had been at the university, how many years they had known their mentor, how long they had worked with their mentor, and the mentor’s academic level. Students were also asked to indicate “yes” or “no” to five listed reasons for choosing their mentors: (1) mentor was an inspiring instructor, (2) interested in mentor’s area of research, (3) mentor sought out student to work with him or her, (4) mentor recruited student to the program, and (5) student sought out mentor because of similar interests. Students were asked to use a 10-point semantic differential scale ranging from 1 (no interest) to 10 (a great deal of interest) to rate their mentor’s interests in (1) their future, (2) helping them get into a graduate program, (3) helping them obtain an internship, (4) helping them obtain a postdoctoral position, and (5) helping them find a good job in their field. Students were also asked to list the one most important thing that their mentor did for them. Students were asked to use a 10-point Semantic Differential Scale ranging from 1 (extremely dissatisfied) to 10 (extremely satisfied) to rate their satisfaction with their mentor and to rate the extent to which they would like to be like their mentor (1 = not at all to 10 = very much). Finally, students were asked to rate the importance of your mentor to you using a 6-point scale ranging from 1 (not at all important) to 6 (extremely important).
Students with mentors were asked the following questions, and if the answer was yes, the student was asked whether the work was with their mentor: Are you working on any publications (no or yes); if yes, how many publications are you working on? How many publications have you authored? On how many publications are you first author? How many conference presentations/posters have you authored? If yes, how many conference presentations/posters have you authored? and, Are you first author on any conference presentations/posters? If yes, how many first author conference presentations/posters do you have?
Students without mentors were asked to respond “yes” or “no” to four listed reasons for not having a mentor: (1) have not been at the university long enough, (2) do not feel the need for a mentor, (3) the program does not encourage mentoring, and (4) have not been able to find a satisfactory mentor. Students were also asked to use the same scale to indicate whether they had ever been involved in research.
Procedure
The undergraduate advisor and the graduate program coordinators from both institutions forwarded a letter, via e-mail, to all undergraduate, master’s, and doctoral students, inviting them to participate in the study. The letter to the students identified the investigator as a professor in the department who was conducting a study on mentoring. Students were told that the survey would take about 10 min, that the information they provided would be anonymous, and that their information would be collected through Qualtrics (2018). The link for the survey was included in the attached letter, and students were told that, as an incentive, four participants would be randomly selected to receive a US$25 Starbucks gift card.
Results
Over half (52.2%) of the students did not have a mentor. However, these percentages differed greatly by educational level, χ2(2) = 83.98, p < .001. Among graduate students, 95.45% reported having a mentor. Two of the students without mentors were master’s-level students. One doctoral student reported not having a mentor. Among undergraduate students, 28.57% reported having a mentor. There was not a significant difference between the proportions of students with a mentor by sex, χ2(1) = 1.72, p = .190. Nor were there differences in educational level by sex or in the proportions of mentorship by sex within educational levels: undergraduate, χ2(1) = 0.06, p = .804; master’s, χ2(1) = 0.88, p = .350; and doctoral, χ2(1) = 0.26 p = .612.
Reasons for Choosing a Mentor
There were significant differences in two of the reasons reported for choosing a current mentor—being an inspiring instructor, χ2(2) = 12.49, p = .002, and being interested in the mentor’s research, χ2(2) = 29.79, p < .001. Undergraduate students were significantly more likely to choose mentors for being inspiring instructors, and graduate students were significantly more likely to choose mentors based on interest in their research. No other reasons differed significantly by education level, sought out, χ2(2) = 0.552, p = .759; recruited, χ2(2) = 0.204, p = .903; similar interests, χ2(2) = 2.940, p = .230; and assigned, χ2(2) = 1.700, p = .428. Graduate students’ choices of reasons for choosing a mentor were interest in the mentor’s research (100%), similar interests (88.23%), being sought out by a mentor (25.49%), being recruited by or assigned to a mentor (23.53%), and finding a mentor to be an inspiring instructor (15.69%). Undergraduate students’ choices were similar interests (78.05%), interest in the mentor’s research (53.66%), finding a mentor to be an inspiring instructor (46.34%), being sought out or recruited by a mentor (26.83%), and being assigned to a mentor (17.07%).
Perceived Interest in Students’ Futures
Students at all levels perceived their mentors to be very interested in their futures, F(2, 89) = .19, p = .824 (doctoral: M = 8.46, SD = 1.86; master’s: M = 8.57, SD = 1.65; and undergraduate: M = 8.29, SD = 1.75). Doctoral students perceived mentors as being most interested in helping them obtain postdoctoral positions (M = 8.07, SD = 1.51) and jobs (M = 8.11, SD = 1.50). Master’s students perceived mentors as being most interested in helping them get into a graduate program (M = 8.48, SD = 2.13) and obtain a job (M = 7.78, SD = 2.28). Undergraduate students perceived mentors as being most interested in helping them get into a graduate program (M = 8.10, SD = 1.90), obtain a job (M = 7.71, SD = 2.50), and obtain an internship (M = 7.23, SD = 2.64). There were no significant differences between the three groups with respect to perceived mentor’s interest in obtaining an internship, F(2, 83) = .28, p = .757, or job, F(2, 86) = .29, p = .753. There were significant differences by educational level in the perceived interest the mentor had in helping the student get into a graduate program, F(2, 81) = 6.05, p = .004. Tukey Honest Significant Difference (HSD) post hoc tests with family-wise Bonferroni corrections were conducted to evaluate significant pairwise differences (α = .025). These indicated that mentors were perceived to be significantly more interested in helping students get into a graduate program among undergraduate (M = 8.10, SD = 1.90) and master’s (M = 8.48, SD = 2.13) students than among doctoral (M = 6.32, SD = 2.92) students, ps = .011 and .006, respectively. Similarly, there were significant differences in perceived interest in helping a student obtain a postdoctoral position by educational level, F(2, 83) = 3.33, p = .041. No pairwise comparisons remain significant after a family-wise Bonferroni correction is applied: undergraduate (M = 6.38, SD = 3.07) and master’s (M = 6.95, SD = 2.94) students had descriptively lower average endorsement than did doctoral (M = 8.07, SD = 1.51) students, ps = .032 and .308, respectively. The polynomial contrast for a linear increase in perceived mentor’s interest in helping a student obtain a postdoctoral position was significant, p = .012.
Qualities of Good and Bad Mentors
There were no significant differences by education level in evaluations of the most important thing the mentors do for their students, χ2(12) = 11.272, p = .506. Among all students, mentors being interested in and supportive of students was the most frequently cited “most important thing” (39.77%), followed by being accessible (27.27%) and being involved in research and being resourceful (12.50%). Relatedly, students were asked to list the five most important characteristics of a good and a bad mentor (see Table 1). Final scores were computed by weighting the proportions of students who listed the given characteristic by their reversed rank (e.g., first place rank received a 5-point weight and fifth place rank received a 1-point weight). The most important characteristic for both positive and negative evaluation was personality. The second and third most important positive characteristics were shared interests and knowledge/competence, respectively. The second and third most important negative characteristics were a lack of accessibility and a lack of shared interests, respectively.
Characteristics of Good and Bad Mentors.
Note. Scores are the sum of reverse-scale weighted rank proportions. Superscripts numerals indicate ranks for top three characteristics in each domain.
The boldface values signifies the three most important Positive and Negative Characteristics of mentors.
Perceived Importance of Mentors
All students rated their mentors as very important to them; there were no significant differences in the means of these evaluations by student level, F(2, 83) = 1.304, p = .277 (doctoral: M = 5.23, SD = 0.99; master’s: M = 5.61, SD = 0.50; undergraduate: M = 5.32, SD = 0.91), although there was significantly more variability in the responses of doctoral and undergraduate students than in master’s students, Bartlett’s χ2(2) = 10.57, p = .005. The ratings for undergraduates ranged from 2 to 6, and the ratings for doctoral students ranged from 3 to 6, but the ratings for master’s students ranged only from 5 to 6. There were significant differences in students’ desires to be like their mentor by education level, F(2, 87) = 5.17, p = .008. Undergraduate students reported a stronger desire to be like their mentor (M = 8.51, SD = 1.92) than doctoral students (M = 6.86, SD = 2.34), p = .006. Master’s-level students (M = 8.13, SD = 2.16) did not differ significantly from doctoral students, p = .088. There were also significant differences by educational level in mean levels of satisfaction with one’s mentor, F(2, 87) = 7.92, p < .001, as well as in the variability of these responses, Bartlett’s χ2(2) = 8.91, p = .012 (undergraduate: M = 9.18, SD = 1.32; master’s: M = 8.83, SD = 1.70; and doctoral: M = 7.5, SD = 2.24). Undergraduate students reported greater satisfaction with their mentor than doctoral students, p = .001. Master’s-level students were also significantly more satisfied than doctoral students, p = .022.
Mentorship and Productivity
Examining productivity as a function of mentorship in graduate students was not reasonable, given the general dearth of graduate students without mentors. However, productivity among undergraduate students as a function of mentorship was examined. Having a mentor significantly predicted the number of publications on which students were currently working, χ2(4) = 25.541, p < .001. Students with mentors were more likely to have publications in progress than expected, and students without mentors were less likely to have publications in progress than expected. In fact, all publications in progress among undergraduates were publications on which they were collaborating with their current mentor, and only one student without a mentor reported having a single publication in progress. The effect of having a mentor on whether a student already had a completed publication was not significant, χ2(1) = 3.636, p = .057. There was no effect of mentorship on the number of first-author publications among undergraduates, χ2(1) = 0.511, p = .475. (Only 15 undergraduates answered this question, and only four had a first-author publication.) Having a mentor predicted the number of conference presentations or posters students had, χ2(6) = 18.289, p = .006. Students with mentors were less likely to have no presentations and more likely to have presentations than expected, but the opposite pattern was true for those without mentors. The same patterns existed for the significant effect of mentorship on whether students had first-author conference presentations or posters, χ2(1) = 5.743, p = .017.
Racial Minorities and Mentorship
White was the most commonly reported specific racial identification (46.35%), followed by Hispanic/Latino (25.75%), Asian (13.73%), Other (10.30%), Black/African American (3.43%), and Native Hawaiian/Pacific Islander (.43%). Racial minority students (collapsed variable) were significantly less likely to have a mentor than White students, χ2(1) = 6.188, p = .013. Among those with mentors, 55.96% were White; however, among those without mentors, only 39.50% were White. This largely reflects significant differences in student level by ethnicity, χ2(2) = 16.95, p < .001. Undergraduates were more likely to be non-White and graduate students were more likely to be White. Within undergraduates only, there were no significant differences in the proportions of those with mentors by ethnicity, χ2(5) = 3.677, p = .597. Tests within graduate students are of limited use because nearly all graduate students had mentors. There were 21 (of 66) non-White graduate students. Among non-Whites, two (9.52%) were without mentors, but only one (2.33%) White student did not have a mentor.
With such few cases, inferences based on these data are not tenable. After controlling for educational level, student ethnicity (White vs. non-White) did not affect: (1) importance of mentor, (2) desire to be like mentor, (3) satisfaction with mentor, (4) perceived interest in future, (5) number of publications being worked on with mentor, (6) total publications authored, or (7) satisfaction with university.
Comparison of 1986 and 2017 Studies
To evaluate the degree of change that has occurred over the past three decades, the data for graduate students from the present study (2017) were compared to the data from the Cronan-Hillix et al.’s (1986) study, using independent-samples tests of proportions and one-sample tests of means (because estimates of variability were lacking from the 1986 study, the means were used as expected values). No undergraduates were included in the 1986 study, so they were excluded for comparative purposes. Also, comparisons could not be made regarding the effects of having a mentor because only three graduate students in the 2017 study lacked a mentor.
In the Cronan-Hillix et al.’s study, there were 90 graduate students who were enrolled at the same large mid-Western research-intensive university; participants were in one of the six specialized programs: clinical (57%), ecological/community (17%), developmental (9%), industrial/organization (9%), social (6%), or experimental (3%). The proportions of students in each program corresponded closely to the proportions enrolled in each specialty for the year. Fifty-three percent of the students were male, the mean age was 29.7, with a range of 22–48.
Tests of proportions were performed to evaluate the differences between the 1986 study and the present study. Having a mentor was significantly more common in 2017 (95.45%) than it was in 1986 (53%), p < .001. The probability of having a female mentor was also significantly greater in 2017 (50.98%) than it was in 1986 (13%), p < .001. There were no significant differences between the proportion of mentors holding full professor rank in 1986 (71%) and in 2017 (58.82%), p = .212.
The proportion of students who reported choosing their professor because of similar interests or being sought out by, p = .865, assigned to, p = .422, or recruited by, p = .422, the mentor did not differ between 1986 and 2017. However, students were significantly less likely to report choosing a mentor for being an inspiring instructor in 2017 (15.69%) than they were in 1986 (46%), p = .001.
The roles of mentors also seemed to have shifted from 1986 to 2017. In the 1986 study, the 5-point scale was dichotomized to compare 3–5 (occasionally always) against 1–2 (never rarely). For comparison, this same dichotomization was performed with the 2017 data. Students reported that they were significantly more likely to discuss research (1986: 75%; 2017: 93.88%, p = .006), their program or plan of study (1986: 41%; 2017: 77.55%, p < .001), and their coursework (1986: 34%; 2017: 55.10%, p = .016) with mentors in 2017 than they were in 1986. There were no significant differences in the proportions of students in 1986 (49.7%) and 2017 (36.73%) who reported socializing with their professors, p = .132. According to a one-sample t test in which the reported mean from the 1986 study was used as an expected value (no measures of variability were reported to permit an independent-samples t test), students’ satisfaction with their mentors did not change significantly from 1986 (M = 8) to 2017 (M = 8.10, SD = 2.10), t(50) = .33, p = .74.
Discussion
The findings from the present study indicated that almost half of the student respondents had mentors. More graduate than undergraduate students reported having mentors. This is consistent with other research indicating that having a mentor was more common for graduate than for undergraduate students (Blake-Beard, Bayne, Crosby, & Muller, 2011; Clark et al., 2000). This is hardly surprising, as students have more time and opportunity to form a relationship with mentors as their educational level increases. However, undergraduate students were more satisfied and more likely to want to be like their mentors than graduate students.
Findings from the present study also indicated that undergraduate students were more likely to select a mentor because he or she was an inspiring instructor, while graduate students were more likely to select a mentor because of similar interests. This corresponds to the differences in the ways that different levels of students are selected by faculty. Undergraduate students are more likely to become a member of a mentor’s research team through enrollment in a mentor’s course or an informal selection process. On the other hand, graduate students usually submit formal applications to the department in which they would like to complete their degree, often motivated by mentorship match and research alignment (Kracen & Wallace, 2008). Faculty members then interview a number of potential graduate students before making decisions about which graduate student(s) they would like to admit to the program and have work with them.
Students at all levels reported perceiving their mentors as interested in their futures. Undergraduate- and master’s-level students perceived their mentors to be most interested in helping them get into graduate school. Doctoral students perceived their mentors as most interested in helping them obtain a postdoctoral position or job. These findings indicate that mentors are sensitive to the problems their mentees face at each stage of their careers, as their focus changes from the university to the outside world.
Good and Bad Characteristics of Mentors and Student Satisfaction
The results of the present study indicated that students reported personality characteristics as the most important positive and negative characteristics of a mentor. This finding is not surprising because the nature of a faculty member’s roles and responsibilities has evolved over time (Leech, Haug, Iceman-Sands, & Moriarty, 2015). Most faculty members in public higher education institutions are now required to conduct a significant program of research, thereby providing students with numerous choices for potential mentors. In addition, many faculty members with extramural funding may require involvement from multiple people. Thus, if faculty members want to be successful, it behooves them to be perceived as nice people who get along with their students. Students also report that having a mentor who is interested and supportive is very important (Johnson et al., 2000; Malachowski, 1996). Students who volunteer their time, and who are selected to work on a research team, want to interact and develop a relationship with their mentors. Mentors who fail to convey interest and support may not be able to retain students on their research teams. Shelton (2003) found that students who perceived greater faculty support were more likely to remain in their programs than those who perceive less support. Student respondents in the present study rated a mentor’s knowledge/competence as very important. The importance of having a mentor who was perceived by the student to be knowledgeable and competent was also found to be important in the Cronan-Hillix et al.’s (1986) study. Similarly, Shellito, Shea, Mueller-Solger, and Davis (2001) found that knowledgeable mentors were important to students. Faculty members who are perceived to be knowledgeable and competent may be more likely to be pursued as mentors.
In the present study, students also rated a mentor’s accessibility as an important characteristic. Other researchers have also reported that students want their mentors to be accessible (Shellito, Shea, Mueller-Solger, & Davis, 2001). The invention of the Internet has allowed mentors to become more responsive and reachable, and this may have increased the expectation that they should be more available, even if they are not available physically. Limited accessibility may reduce students’ interactions and opportunities for advancement in their profession. This is a difficult balance for some faculty members to maintain, especially those who travel often to give talks, serve on various grant review committees, or those who serve as editors for journals, and those who have families. Therefore, faculty members who are the most active in the field may be the least available. However, graduate students may not be in the best position for evaluating the merits of having a productive and well-known mentor who is less available versus one who is more accessible but not as productive.
Students indicated that they were highly satisfied with their mentors, but undergraduate and master’s students reported higher levels of satisfaction than doctoral students. Although there were differences, it is important to note that mean satisfaction levels were high for all educational levels. Some researchers have found that graduate students were satisfied with their mentoring experiences (Clark et al., 2000), while others have found that student satisfaction decreased as the number of years with the mentor increased (Taylor & Neimeyer, 2009). A possible explanation for this relationship may be that doctoral students who remain in their program for a longer period of time may be more frustrated as a consequence of their program’s heavier requirements (Van Vliet, Klingle, & Hiseler, 2013). Lower levels of satisfaction among doctoral students may also be a function of the smaller education gap between doctoral students and their mentors, suggesting that doctoral students may not learn as much from their mentors as undergraduate students. Another potential explanation for this difference is that there are more likely to be changes in the dynamics of the relationship for doctoral students and their mentors than for undergraduate students and their mentors. That is, the relationship between a doctoral student and a mentor is more likely to shift over time from perceiving the mentor as having more power and prestige to one that is more balanced and collegial. However, this change would be unlikely to occur with undergraduate students because they do not feel competent to criticize their mentor. Thus, doctoral students may exhibit a lower tendency toward “hero worship.”
Faculty members often prepare graduate students for prospective faculty positions (Austin, 2002; Boyle & Boice, 1998). However, they do not guide undergraduate students in the same way (Colbeck, 1998). Mentors have higher expectations for doctoral students than for undergraduate students, and thus may place more pressure on doctoral students to succeed, causing them to experience stress and frustration (Johnson, 2002).
Importance of Mentors
In the present study, mentors were very important to students of all educational levels. This finding is not surprising, because working on a faculty member’s research team provides students with opportunities for professional development, increased confidence, and role modeling, all of which could be valuable in helping students reach their academic or professional goals (Mangione, Borden, Nadkarni, Evarts, & Hyde, 2018). Furthermore, the mentoring dynamics in academia serve unique functions that are different from the mentoring dynamics of other environments (Galbraith & James, 2004). For instance, undergraduate students often volunteer their time to work in research, and they rely on their mentors to support their future aspirations and to guide them to their next steps. In traditional work environments, such as human resources, people do not volunteer their time, nor are the mentors expected to be preparing them for other positions.
Differences Between the 1986 and 2017 Studies
The Cronan-Hillix et al.’s (1986) study included only graduate students, and only 53% of the respondents in that study had mentors. In 2017, 95% of the graduate students had mentors. Explanations for these differences could be that they occurred as a result of changes in research over the past 30 years. Other explanations include that students have changed over time and that the universities where the 1986 and 2017 studies took place were different. In 2017, faculty members in psychology departments were more likely to have funding for their research. Publishing in the top-tiered journals has become more competitive (Certo, Sirmon, & Brymer, 2010), and more psychology graduate programs may rely on faculty-obtained grant funding to support their graduate students.
There were no significant differences in the academic rank of the mentors for graduate students over time, and the majority of faculty mentoring graduate students were full professors. However, there was one significant difference between the 1986 and 2017 studies in the reasons for choosing a mentor. In 1986, graduate students were significantly more likely to choose a mentor for being an inspiring instructor than in 2017. This difference may be because the selection procedures for graduate students have not changed much over the past 30 years. This may also reflect the devaluation of the teaching role in universities, the increased requirement to have a mentor in order to enter graduate school, and the increasing competitiveness of programs.
The roles of mentors have shifted from 1986 to 2017, as graduate students were more likely to report that they discussed research, their program or plan of study, and their coursework with their mentors in 2017 than in 1986. This may be because graduate students may have spent more time with their mentors in research in 2017 than in 1986. However, there were no differences in reported socializing with their mentors over time. Finally, there were no significant differences in graduate students’ reported satisfaction with their mentors over time.
Limitations
As in all studies, there were limitations in the present study. The questionnaire used in the present study had been used only in the Cronan-Hillix et al.’s (1986) study and thus had limited demonstration of reliability and validity. The measure was a self-report questionnaire and thus inherently suffers from self-report bias. Researchers need to determine the reliability and validity of measures for assessing students’ views of mentoring. Another limitation of the study was that it included two large Western public universities and may lack generalizability to other students or universities, particularly those at nonpublic institutions and those in other geographical locations with smaller class sizes and lower student to faculty ratios. In addition, the representativeness of the samples was different. Participants in the 1986 study were all enrolled in doctoral programs at a large mid-Western university. In the present study, the participants were enrolled in one of the two large Western universities and included undergraduate, master’s, and doctoral students. The comparisons of graduate students in the two studies differed in that in the 1986 study, all students were enrolled in doctoral programs; however, those in the 2017 study included both master’s and doctoral students. Therefore, the generalizability of the findings to all psychology students may be limited.
Future Directions and Conclusions
Future researchers should examine whether students’ views of mentoring vary as a function of whether the university is public or private. This distinction may be important for students who are making decisions about which type of university is better suited to their needs, their research goals, and their interests. Furthermore, researchers should continue to examine whether students’ views of mentors change as competition for research funding, competitiveness for publishing in high-impact journals, competition for graduate schools, postdoctoral fellowships, and tenure-track positions change. Finally, examining generational changes in students’ views of mentoring may assist faculty in better meeting the needs of their students.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.