Predictive Variables of Faculty Retention in the Counselor Education Field: A Gender Perspective

Faculty, especially women faculty, continue to leak from the academic pipeline. Considering the costs related to faculty turnover, researchers have recommended continued exploration of faculty experiences with occupational satisfaction and retention. Accordingly, utilizing quantitative methodology and content analysis, the authors examined the reported experiences of 218 professors in the Counselor Education field. Predictive variables related to faculty retention were identified. These variables included lack of support from administration, lack of support from colleagues, having to work harder than colleagues to be taken seriously, gender, and partner job status. Data also revealed significant differences between gender and partner job status, gender and workload variables, and gender and retention variables. These results provide information on the experiences of both men and women faculty, and highlight gender inequities in higher education.


Introduction
For several decades, researchers have explored gender inequities in the academy (American Council on Education, 2005;Mason & Goulden, 2004;Misra et al., 2011). Scholars have used a pipeline metaphor to describe progression in academic careers, noting that women leak from the academic pipeline more often than men (Mason et al., 2013). Researchers have identified some of the factors that contribute to the leaky pipeline for women in the academy, including that women provide more teaching and service than men-work that is less valued in the university reward system (Hermann, Haskins, Neale-McFall, Ziomek-Daigle, & Eckart, in press;Sallee, 2014;Webber & Rogers, 2018). Furthermore, resources that support heavily rewarded research activities are more likely to be distributed to men (Hermann et al., in press).
The reward system in academia also continues to reflect the ideal worker standard, a professor who can devote all of his time to work-related tasks (Halpern, 2008;Philipsen & Bostic, 2010;Ward & Wolf-Wendel, 2012). This reward system was designed for a different generation of faculty, a generation who had the support of secretaries at work and wives at home (Philipsen & Bostic, 2010;Seltzer, 2015). In the current generation, 72.3% of women with young children are in the workforce in the United States (Bureau of Labor Statistics, 2020). Yet, women still engage in more childcare and housework than their partners (Hermann et al., 2019;Hochschild, 2012;Slaughter, 2015). This inequitable, gendered division of childcare and household labor has been found to negatively impact women faculty (Hermann et al., in press;. Philipsen & Bostic, 2010). Technological advances have extended the faculty workday as well (Berg & Seeber, 2016).
As researchers have examined the leaky pipeline for women academics, they have explored various factors that correlate to professors' occupational satisfaction (Hermann et al., in press;Alexander-Albritton & Hill, 2015;Hill et al., 2005;Magnuson et al., 2009;Neale-McFall et al., 2018). Factors that negatively impact faculty's occupational satisfaction include lack of support from colleagues and administrators (Hermann et al., in press;Hill et al., 2005;Magnuson et al., 2009;Neale-McFall et al., 2018). Having to work in a toxic environment also correlates with professors' low job satisfaction (Hill et al., 2005;Magnuson et al., 2009). Hermann and Neale-McFall (2018) recommended that researchers continue to identify barriers that inhibit women faculty's success in the academy. These scholars also suggested that future research include information on men's experiences in academe. Researchers have also noted that considering the cost of replacing faculty, further study on occupational satisfaction and faculty retention is warranted (Mason et al., 2013;Webber & Rogers, 2018). This study was designed to address these recommendations through the exploration of factors related to the occupational satisfaction and retention of men and women faculty. And, as faculty experiences vary by academic discipline (Wolf-Wendel & Ward, 2006), the focus of the study was on the experiences of professors in the researchers' field of study, Counselor Education.

Purpose of Study
The purpose of this study was to explore the nature of men and women counselor education professors' work and to assess the relationship among demographic variables, workload variables, and retention. Retention was broadly defined as whether faculty were considering leaving their institutions. Workload variables included number of hours per week spent doing activities related to academic work, and work activity preferences (i.e., teaching, research, service). Research Questions for this study included: 1. Is there a significant relationship between gender and partner job status? 2. Is there a significant relationship between gender and workload variables?
3. Is there a significant relationship between gender and retention variables? 4. What variables predict consideration of leaving a university faculty position?

Participants
Participants for this study (N = 218) were full-time counselor educators currently employed in the United States. Regarding gender, 147 participants identified as women (69.7%), 62 as men (29.4%), 2 as transgender (0.8%) and 7 did not respond (.03%). In terms of ethnicity, 46 participants identified as part of an ethnic or racial minority group (21.8%) and 161 identified as Caucasian (76.3%). Eleven participants (1.9%) did not respond.

Procedure
In order to identify counselor education faculty members for participation in the study, the researchers utilized the internet to search for counselor education programs in the United States. This search yielded websites of CACREP-accredited and non-CACREP-accredited programs, master's and doctoral programs, and programs providing training in a broad range of counseling specializations. Based on this information, a database with a list of email contact information for counseling program faculty members across the U.S. was created. A total of 1,804 email addresses were obtained.
After receiving IRB approval, the first author sent an email informing prospective participants of the parameters for participation in the study, purpose of the study, risks, benefits, and contact information. A secure link to the survey instrument was included in the email. The participation request was sent out a total of three times over several months. A total of 218 participants completed the survey, yielding a return rate of 12.1%. For demographic descriptive data, the total number of participants equals 218. Due to incomplete surveys, 187 cases were utilized for ANOVA and regression analysis.

Instrument
The workload survey is a 48-item measure that consists of three sections: demographics, workload, and retention items. In creating the survey, the researchers established content validity by utilizing a comprehensive review of literature (e.g., Hermann et al., in press;Alexander-Albritton & Hill, 2015;Hill et al., 2005;Magnuson et al., 2009;Neale-McFall et al., 2018) and an expert panel to generate and refine the items on the survey. The expert panel included two men and six women counselor educators who have previously conducted research and presented on workload in academia.
The demographics section of the workload survey consisted of categorical status variables (i.e., gender, ethnicity/race, marital/partner status, job status of partner, CACREP/non-CACREP masters/doctoral level, and academic rank). The workload section of the survey assessed institutional variables (e.g., average number of hours per week spent doing activities related to work as a faculty member, work activity preferences). The retention items section assessed participants' perceptions related to support (e.g., from departmental colleagues regarding work, from administration regarding work) and workload equity (e.g., having to work harder than colleagues to be taken seriously). Whether a participant had considered leaving their university in the last year was also included in this section.
An open-ended question was included on the survey. This question asked participants to list the main reason(s) they might leave their institution. The question was designed to allow participants to elaborate on the factors that influence the retention of counselor educators.

Data Analysis
RQ1: ANOVA was conducted to compare differences between gender and partner job status (works full-time, works part-time, does not work). There was a significant effect of partner job status and gender at the p<.05 level for the three conditions, F (2, 134) = 3.378, p = 0.037; women participants (M=1.31, SD = .656) were more likely than men participants (M=1.64, SD = .821) to have partners who worked full-time. Further analysis indicated that 80% of the women had partners that worked full-time and 57% of men had partners that worked full-time; 9% of women had partners that worked part-time and 21% men had partners that worked part-time; 10% of the women had partners that did not work and 21% of men had partners that did not work.

RQ2:
ANOVA was conducted to compare differences between gender and workload variables. For gender and average number of work hours, there was a significant effect at the p<.05 level, F (1,180) = 6.50, p = 0.012; women participants indicated a greater number of work hours (44.48 vs. 38.76) per week. For gender and how one prefers to spend time (i.e., teaching, research, service), results were not significant, F (1,179) =.446, p = 0.505; however, mean scores indicate men were more likely to rank teaching as their preferred way to spend time: 61.4% (n=35) vs. 49.2% (n=61), women were more likely to rank research as their preferred way to spend time: 24.2% (n=30) vs. 14% (n=8), and neither group indicated a strong desire for service as their preferred way to spend time: 3.22% for women (n=4) vs. 1.8% men (n=1). When asked, Do you feel like you typically spend more time on service requirements than required, both women and men indicated they Agreed or Completely Agreed with the statement: 83.9% (women) and 73% (men), although results were not significant by gender, F (1,180) = 3.350, p = .069.

RQ3
: ANOVA was used when analyzing retention variables and gender. Retention variables included: In the past year, have you considered leaving your current institution? How supported do you feel by departmental colleagues regarding work responsibilities? How supported do you feel by your administration regarding work responsibilities? And, I have to work harder than other colleagues to be taken seriously in my department. Results were significant for all analyses (see Table 1). A significant effect of gender was found for counselor educators who considered leaving their institution in the past year, F(1, 180) = 7.047, p= .009; women were more likely (70.2%) than men (49.1%) to answer 'Yes.' Regarding departmental support, a significant effect of gender was found for how supported counselor educators felt by department colleagues regarding work responsibilities, F(1,180) =4.71, p=.031; 49% (n=66) of women felt Supported or Very Supported vs. 70% of men (n=40) who felt Supported or Very supported. Similarly, with administrative support, a significant effect of gender was found, F(1,180) = 6.75, p = .010, with 30% (n=41) of women feeling Supported or Very Supported and 52% (n=32) of men indicating feeling Supported or Very Supportive. Perceptions of having to work harder than colleagues to be taken seriously in their department by gender was also significant; F(1,180) = 10.69, p= 0.00, with 60% (n=75) of women and 42% (n=24) of men agreeing with this statement.
RQ4: A significant regression equation was found F (5, 175) = 14.561, p < .001 regarding counselor educators' consideration of leaving his or her current institution in the past year; 38.2% (R 2 =.382) of the variance was predicted by the following five variables, respectively: Support from administration regarding work responsibilities, Feelings of having to work harder than colleagues in order to be taken seriously, Support from departmental colleagues regarding work responsibilities, Partner job status, and Participant Gender. Therefore, counselor educators who did not feel as supported from departmental and administrative colleagues, those who felt they had to work harder to be taken seriously, those whose partners worked full-time, and those who identified as women were more likely to think about leaving their current institution in the last year.

Content Analysis
Of the 218 counselor educator participants, 81 women and 25 men responded to the open-ended question on the reason(s) they might consider leaving their university. Participants' responses to this question were analyzed through content analysis (Hsieh & Shannon, 2005;Kondracki et al., 2002). In this approach, researchers review the responses and identify themes (Hays & Singh, 2012). Results provide further understanding of participants' experiences (Hsieh & Shannon, 2005;Potter & Levine-Donnerstein, 1999).
The first and second authors examined participants' responses independently, identifying themes and subthemes as they reviewed the text. For example, the subtheme "hostile work environment" represents comments related to marginalization such as "feeling disrespected" and feeling "marginalized." To determine how visible themes were in the data, the researchers used frequency counts (Hays & Singh, 2012;Hsieh & Shannon, 2005). The researchers arrived at similar themes and frequency counts, indicating coder consistency and providing evidence of trustworthiness (Hsieh & Shannon, 2005;Kondracki et al., 2002;Potter & Levine-Donnerstein, 1999).
While Kondracki et al. (2002) caution researchers not to use frequency counts to establish the magnitude of an issue, the number of men and women participants who identified similar concerns can help identify trends and further illuminate the quantitative data findings. Accordingly, frequency counts and percentages based on the number of the men and women who responded to the question are included. Quotes from participants further illustrate their experiences.
Themes from these data included challenges related to the work climate, experiences with an unreasonably heavy workload, and work-life balance issues. Two subthemes related to work climate also emerged.

Work Climate
Both men and women seemed to be experiencing challenges with their work environment, with 36 women (44.4%) and 8 men (32.0%) describing their concerns related work climate issues. Subthemes included experiencing a lack of support for colleagues and administrators and a hostile work environment. As with the presentation of percentages related to each theme, subtheme percentages are based on the number of men and women who responded to the open-ended question.
Lack of Support. Fifteen women (18.5%) and three men (12.0%) indicated they experienced a lack of support from their peers and the administration. Three women (3.7%) and two men (8%) explained that they were in non-tenure track positions and felt as though they were doing the work of tenure-track faculty for less pay and less respect. An additional six women (7.4%) described feeling undervalued. Four women (4.9%) expressed that budget issues at the university were impacting support for their programs and faculty lines. Ten women (12.3%) and two men (8.0%) added that low pay was an issue for them.
Hostile Work Environment. Twelve women (14.8%) and two men (8%) experienced working with difficult colleagues, harassment, and relational aggression. Five women (6.2%) and one man (4.0%) described feeling marginalized in their work settings. Six women (7.4%) and one man (4.0%) noted that gender discrimination led to an "inequitable distribution of resources" including women receiving lower salaries. These participants further described a "gendered workload" in which "women carry the majority of service commitments."

Unreasonably Heavy Workload
Twenty-one women (25.9%) and seven men (28.0%) indicated the "lack of reasonable workload expectations" was an impediment. Participants described overwhelming expectations. They reported that "demands far exceed resources" and the "do more with less" norms created challenges. One participant elaborated that she experienced a "requirement to be on campus all day and teach several nights a week." Ten women (12.3%) and four men (16.0%) specifically addressed the negative impact of their heavy service responsibilities.

Work-Life Balance Issues
Nine women (11.5%) and two men (8.0%) identified the work-life balance issues they experienced, including a woman participant who described a "weekend class schedule that interfered with family responsibilities." The women further explained that they were working at universities that were "not family-friendly" and had "no childcare" support. One woman elaborated on her colleagues' "lack of understanding of responsibilities of caring for a small child." And a man also reported that his work responsibilities conflicted with his "personal and family time." In terms of university location, five women (6.2%) and five men (20.0%) experienced challenges related to the geographic location of the university. These participants explained that the geographic location impacted their quality of life and extended their commute. Participants also noted that the location of the university negatively impacted their ability to live closer to family.

Discussion
The results of this study add to the literature on the occupational satisfaction and retention of faculty (e.g., Hermann et al., in press;Alexander-Albritton & Hill, 2015;Hill et al., 2005;Magnuson et al., 2009;Neale-McFall et al., 2018). Unique findings of this study include that having an employed partner is one of the variables that predicts whether a professor will consider leaving their university. Given historical cultural norms, it is not surprising that women counselor educators were significantly more likely to have partners who worked full-time. These results may help explain why women continue to leak from the academic pipeline.
Few researchers have reported statistics on the number of faculty who have employed partners. In the Hermann and Neale-McFall (2018) study, 9 of the 10 counselor educator mother participants had partners who worked full-time. Similarly, in this study, 80% of the women participants' partners worked full-time. The demographic findings on men faculty are also somewhat consistent with previous research. For example, in Sallee's (2014) study on faculty fathers, 67% of the men participants had partners who worked full-time. In this study, 57% of the men's partners worked full-time.
Women counselor educators having employed partners is consistent with research highlighting the inequities of a university system that rewards professors who can focus their energies on their work because they have a partner at home who can take care of the household and child rearing activities . These findings also align with research indicating that even when both partners work outside of the home, women counselor educators perform more of the housework and childcare than their partners (Hermann et al., in press;. And, though men are engaging in more childcare than in previous generations, faculty fathers often participate in more of the enjoyable parenting activities as opposed to the routine, practical activities required in childcare (Sallee, 2014). Similarly, women typically do more of the frequent, essential household tasks and men engage in more of the less frequent, discretionary household activities (Dugan & Barnes-Farrell, 2020). Furthermore, women provide the time-consuming management of household activities (Hermann et al., in press;Ciciolla & Luthar, 2019;Sallee, 2014). Women participants in the study reiterated that their colleagues did not understand the extent of their responsibilities related to managing a household and caring for children (Hermann et al., in press;. Women participants also indicated that their universities were not family-friendly and did not provide enough childcare support (Hermann et al., in press;. The content analysis further revealed that challenges significant enough for participants to consider leaving the university included increasing workloads. Participants' comments support previous research attributing the increase in academic workload in the past two decades to technological advances and decreasing numbers of faculty and staff (Berg & Seeber, 2016). Similarly, participants observed that budget cuts reduced university personnel, and that demands exceeding resources has led to overwhelming expectations to do more with less (Hermann et al., in press;Hermann et al., 2014;Hurtado et al., 2012;Seltzer, 2015).
In terms of respect, 60% of women and 42% of men believed they had to work harder than their colleagues to have their work taken seriously. This finding is important because this variable is predictive of whether professors will consider leaving their university. A similar variable, feeling undervalued, was a theme in the open-ended question responses, especially for participants in non-tenure track positions.
Lack of support from peers and administration was a prominent theme in both men's and women's responses to the openended questions. The quantitative data further provide that significantly more women than men felt that they were not supported in their work roles by their department colleagues and by their administration, though this lack of support was strikingly low for both groups, especially in terms of support from the administration. Less than one-third of women and only about one-half of men felt supported by their administration. The data also indicate that only 49% of the women faculty and just 70% of men faculty felt supported by their colleagues. These are important findings as both support from department colleagues and support from administration are variables that predict whether a professor will consider leaving their university. These results add to previous research in which support from colleagues and administrators was correlated to occupational satisfaction (Hill et al., 2005;Magnuson et al., 2009;Neale-McFall et al., 2018).
The content analysis further revealed that both men and women dealt with difficult colleagues, harassment, racism, and relational aggression. These issues have been themes in counselor education literature (Hermann et al., in press;Haskins et al., 2016;Magnuson et al., 2009;Trepal & Stinchfield, 2012). Working in a toxic environment has been linked to low occupational satisfaction for both men and women as well (Hill et al., 2005;Magnuson et al., 2009).
Another unique finding of this study is that men were more likely to rank teaching as their preferred work activity whereas women were more likely to rank research as their preference. Though these results were not statistically significant, they are also not consistent with previous research indicating men are more interested in research (Schuster & Finkelstein, 2006), and the more recent finding that both men and women faculty prefer research activities (Misra et al., 2011). The results also seem contrary to existing research findings that women faculty, including women counselor educators, provide more teaching than men faculty Hermann et al., 2014).
Neither men nor women ranked service as a preference for how they wanted to spend their time; yet 83.9% of women and 73% of men indicated that they exceeded expectations in terms of service activities. These results were supported by the open-ended question responses in which participants of both genders noted their dissatisfaction with their heavy service responsibilities. Yet participants also reiterated gender disparities as they commented that women provided more service than men (Hermann et al., in press;. Other findings on gender discrimination included the participants' comments on the inequitable distribution of resources, with women receiving fewer resources (Hermann et al., in press;Trepal & Stinchfield, 2012). Women also alluded to their low pay, a result found in previous research in which women counselor educators reported that they received lower salaries than men (Hermann et al., in press). These results are also supported by broader statistics including that women are paid only 81.4% of what men are paid in academia, a pay gap that has remained consistent for the past 10 years (American Association of University Professors, 2020). A unique finding from the content analysis is that some men found their low salaries to be an issue as well.
Another unique finding of this study is that there is a significant difference in the number of hours men and women counselor educators work, with women counselor educators working more hours than men counselor educators. This result conflicts with existing literature. Previous research has provided that men are privileged in the pursuit of the prevalent ideal worker standard, the standard in which professors are completely devoted to their academic work and work more hours than their colleagues (Halpern, 2008;Philipsen & Bostic, 2010;Ward & Wolf-Wendel, 2012).
Research on the challenges women experience in balancing home responsibilities and work in the academy (Hermann et al., in press; are supported by the results that women counselor educators are working more hours than men counselor educators, and women are less likely to have a partner who stays at home. And, though counselor education literature has generally focused on women's work/life balance challenges, it is noteworthy that men participants alluded to work-life balance issues as well. This result aligns with Sallee's (2014) finding that Generation X fathers engage in more parenting and are less likely to be willing to sacrifice family responsibilities for work responsibilities.
Variables related to the geographic location of participants' universities also emerged as challenge. As noted in the literature, in order to secure an academic position, faculty members often need to relocate (Hermann et al., in press;Wolfinger et al., 2008). Relocation separates academic parents from family members who may have been able to provide childcare support Wolfinger et al., 2008). Participants in this study expressed distress about not being able to live closer to family. The geographic location of the university also impacted participants' commute (Hermann et al., in press). Participants of both genders in this study indicated that the geographic location negatively impacted their overall quality of life. Conversely, in the Sallee (2014) study on faculty fathers, many participants indicated that they appreciated the locations of their universities as the location provided easy commutes, cultural benefits, and a low cost of living.
The findings that over two-thirds of women and almost one-half of men considered leaving their institution in the past year are particularly noteworthy. These numbers are consequential for both men and women, though statistically women were more likely than men to think about leaving their university. Considering these data, more research on variables that impact counselor educator retention is warranted.

Limitations and Implications for Future Research
Limitations of this study include that the data are self-reported, which can create concerns about participant bias. The use of content analysis also has limitations including the potential for researcher bias, especially considering that researchers could not ask clarifying questions related to participants' responses (Hays & Singh, 2012). And, additional research can further establish validity and reliability measures related to the survey instrument used in the study.
Though limitations exist, the results highlight challenges both men and women experience in academia. Future research on faculty retention can be informed by this study. Given the limited scholarship on partner job status and the result that a partner's full-time job status is a predictive variable in the consideration of leaving one's university, more research on this topic is needed. Future research can also provide additional perspectives on faculty workloads. And as participants reported they were disappointed in their salaries, future studies can provide further data on the pay of both men and women counselor educators.
In addition to establishing topics for future research, these data can be used to support the efforts of universities in creating a better working environment for all faculty. The results illuminate the importance of both colleagues and administrators providing a more supportive environment for faculty. Changing the reward system so that all faculty work is valued can contribute to this goal. Fostering faculty understanding and compassion by engaging in ongoing conversations related to colleagues' workloads and parenting experiences is indicated as well.
These results provide further evidence that universities need to prioritize family-friendly policies as these policies promote faculty recruitment and retention (Sallee, 2014). The provision of childcare on university campuses, adequate paid family leave, and automatic stop-the-clock tenure policies support faculty retention . Family-friendly policies will likely be even more necessary considering the potentially lingering impact of the COVID-19 pandemic on the academic work of faculty with care-giving responsibilities. Study results can provide important information to university administrators about faculty workload prior to the COVID-19 crisis, information that will be critical to ensure that the economic impact of the pandemic does not further increase workloads already significant enough for both men and women faculty to be considering leaving their institutions. Accordingly, if responsibilities in one area of a faculty's workload increase because of this crisis, an equivalent amount of work in another area should decrease. And reward systems need to be recalibrated to reflect these workload shifts. It is also important that administrators ensure that events like a pandemic do not perpetuate gender inequities in the academy.
Finally, study results further support the need for culture change. Philipsen and Bostic (2010) noted that universities are in an ideal position to promote culture change. Professors can challenge the differing expectations for men and women in the workplace and at home, and they can support the creation of a society in which men and women can be active workers and involved parents (Sallee, 2014). Thus, study findings reiterate Sallee's (2014) recommendation that faculty and administrators must work to dismantle gendered infrastructures in order to create a more equitable society for everyone.