This project was made possible, in part, with support from the University of Nebraska-Lincoln Hixson-Lied College of Fine and Performing Arts’ Endowment Fund. Data analysis services were provided by the Nebraska Education and Research (NEAR) center.

Burnout, a long-lasting occupational crisis usually arising from chronic stress encountered at work, is included in the World Health Organization’s International Classification of Diseases (ICD-10, 1991) and has been identified as the biggest occupational hazard of the 21st century. Burnout is particularly problematic in human service occupations, with teaching widely regarded as one of the most stressful fields (Byrne, 1999; DuBois & Mistretta, 2020; Johnson et al., 2005). Burnout and its opposite condition, work engagement, among post-secondary professors across disciplines have been studied  in numerous countries—including China, Turkey, Canada, India, Iran, Ireland, Pakistan, Portugal, South Africa, Spain, the Netherlands, the United Kingdom, and the United States—documenting high levels of work stress and burnout (Gillespie et al., 2001; Han et al., 2020a; Han et al., 2020b, Han et al., 2020c; Kinman & Wray, 2013; Özgül & Polat, 2018; Sabagh et al., 2018).

Burnout among university educators involves feelings experienced within three distinct dimensions: (1) Emotional Exhaustion, (2) Depersonalization, and (3) Personal Accomplishment (Maslach et al., 2018). Emotional exhaustion (EE), often the first indicator of burnout among educators, involves feeling tired and drained of energy. Depersonalization (DP) refers to an indifferent or impersonal regard for students. Educators may exhibit depersonalization through negative responses to students, by using derogatory labels or generalizations about students, displaying distant or cold attitudes toward students, or physically distancing themselves from students. In contrast to the other two burnout dimensions, Personal Accomplishment (PA) refers to positive feelings of competence and achievement with higher scores being less consistent with burnout (Maslach et al., 2018).

Impacts of burnout across all occupations are well-described in literature. Burnout leads to an inability to regulate negative emotions (Golkar et al., 2014), decreased cognitive and fine motor functions (Savic, 2013), and diminished executive functioning, attention, and memory (Liston et al., 2009; Maslach et al., 2018). Other impacts include high job turnover; low morale; absenteeism; reduced productivity and creativity; poor work climate; interpersonal conflict; communication difficulties; and personal problems such as hopelessness, irritability, impatience, and substance abuse (Lowenstein, 1991; Magarell, 1982; Pines & Aronson, 1982). Burnout may not be limited to individuals. It can spread among work colleagues, particularly if work-related stresses are frequently discussed (Bakker & Schaufeli, 2000; Bakker et al., 2001; González-Morales et al., 2012). Successful college music teaching requires sustained high energy and professional enthusiasm to teach effectively while simultaneously meeting myriad service and/or research responsibilities. Faculty burnout is therefore likely to negatively affect the quality of instruction and job performance, impacting not only the individual but also the music institutions that employ them.

For employees in all professions, unremitting high job demands and stressors are the most important predictors of emotional exhaustion (Bakker & Demerouti, 2017). Conversely, providing job resources enhances work engagement (the opposite of burnout), with work engagement positively correlated to job satisfaction (Han et al., 2020c.). Gillespie et al. (2001) summarized stressors of university teaching in general, including low salaries; job insecurity; lack of funding, resources, and support services; task overload; poor leadership; and a lack of promotion, recognition, and reward opportunities. Music and other artistic disciplines tend to be particularly vulnerable to financial and other job resource reductions in climates of fiscal pressure, exacerbating these stressors. Additionally, Demerouti (2014) observed that having adequate recovery time from physical and emotional work demands after work ends and before it starts again is one of the most crucial factors in preventing burnout among all employees. Having adequate recovery time can be particularly challenging in academic music, where faculty frequently work long and irregular hours that may extend late into evenings or weekends.

Demographic Factors for Burnout among University Faculty across Disciplines

Work stress and burnout experience may vary according to demographic characteristics, though findings in literature are mixed. Studies of university faculty across various disciplines to date have examined the interaction of demographic factors and burnout in the emotional exhaustion, depersonalization, and personal accomplishment dimensions. 

Gender

Several researchers have found faculty who are women experience more burnout compared to men (Byrne, 1991; Ghorpade et al., 2011; Lackritz, 2004, Tümkaya, 2007; Watts & Robertson, 2011), but a systematic review by Sabagh et al. (2018) noted that relationships between gender and burnout are inconsistent and contradictory, with most of the studies in their review demonstrating no significant relationships. However, evidence does suggest that workplace experiences vary by gender (Kinman & Wray, 2013; Chronicle of Higher Education, 2020), with women and gender nonconforming faculty reporting poorer well-being, more issues with work-life balance, higher levels of stress, and greater expectations placed upon them with regard to committee service and mentoring students in comparison to male faculty.

Age and Experience

Findings regarding the impact of age on burnout are similarly mixed. Some researchers have found that younger faculty across all university disciplines have higher levels of emotional exhaustion (Byrne, 1991; Fernet et al., 2004; Ghorpade et al., 2011; Lackritz, 2004; Rothman & Barkhuizen, 2008; Sabagh et al., 2018; Tümkaya, 2007; Watts & Robertson, 2011) or observed that increasing age is protective against burnout (Evers et al., 2002; Kinman & Wray, 2013; Reevy & Deason, 2014). However, other studies have found no significant relationship between age and emotional exhaustion (Bilge, 2006; Gonzalez & Bernard, 2006; Li et al., 2013; McClenahan et al., 2007). Findings regarding the relationship between age and depersonalization have been inconsistent (Sabagh et al., 2018), though older university faculty members have reported higher levels of personal accomplishment in several studies (Byrne, 1991; Li, Li, and Sun, 2013; Rothmann & Barkhuizen, 2008).

Along these same lines, studies have demonstrated that faculty with fewer years of experience had higher emotional exhaustion while those with more experience had lower rates of emotional exhaustion (Blix et al., 1994; Gonzalez & Bernard, 1996; Kinman & Wray, 2013). However, other studies (e.g., Bilge, 2006; Byrne, 1991; Rothmann & Barkhuizen, 2008) found no correlation between years of experience and any burnout dimensions.

Faculty Rank, Employment Status, and Institution Type

Han et al. (2020a) examined the impact of institution type and professional rank on university faculty teaching efficacy, engagement, and satisfaction. While all three demographic variables demonstrated statistical significance, the effect size was small, with the researchers concluding there was no practical significance to the findings.

Impacts of employment status and academic rank on burnout are similarly unclear. Some evidence suggests that pre-tenured faculty may experience high work stress due to pressures and expectations related to tenure decisions (Azeem & Nazir, 2008; Chronicle of Higher Education, 2020; Tümkaya, 2007) compared to tenured faculty, though again findings are mixed. Full-time faculty have reported significantly more job demands, stress, and work-life conflict than part-time faculty (Kinman & Wray, 2013) and have shown burnout scores significantly more consistent with burnout compared with contingent faculty (Lackritz, 2004). Term-appointed (e.g., nontenured) or contingent (e.g., adjunct hired to teach one or more courses) faculty face stressors related to job insecurity and may perceive more job stressors and report higher levels of stress, anxiety, and depression (Reevy & Deason, 2014). Yet findings from three studies of 263 faculty in the United States found that contingent faculty without rank and with exclusive teaching duties (no research or service responsibilities) experienced lower levels of both emotional exhaustion and depersonalization, and higher levels of personal accomplishment in comparison to either pre-tenured or tenured faculty (Ghorpade et al., 2011; Lackritz, 2004). Other studies observed no significant relationship between either faculty rank or employment status and burnout (Blix et al., 1994; Fernet et al.,2004; Gonzalez & Bernard, 2006; Li et al., 2013; McClenahan et al., 2007).

Other Demographic Factors

Maslach (2003) observed that being married and having children seemed to protect against burnout, while single workers experience the highest levels of burnout. Teaching load, types of students taught, and job apportionment have also been examined in relation to burnout among faculty across disciplines. The number of students taught (Lackritz, 2004) and particularly the number of graduate students taught (Lackritz, 2004; Watts & Robertson, 2011) can also predict higher Emotional Exhaustion and Depersonalization. In addition to teaching load, Lackritz (2004) found that time grading, maintaining office hours, pursuing grant money, service time, number of service activities, and overall time spent as a faculty member all positively correlated with Emotional Exhaustion. The mixed nature of findings about how various demographic characteristics impact burnout experience suggest that demographic factors interact in complex ways. Any observations about how demographic characteristics relate to burnout are likely to vary widely among different populations.

Studies of Burnout among University Music Faculty

Few burnout studies have focused specifically on college or university music faculty, with limited observations about how demographic characteristics influence burnout experience in this population. Both Hamann et al. (1988) and Wristen (2023) compared burnout rates of university music professors to a normative group of university faculty across disciplines. University music faculty in the Hamann et al. (1988) study demonstrated slightly lower mean burnout scores in music faculty compared to police officers, nurses, counselors, physicians, and others in the helping professions. Nonetheless, the number of music faculty experiencing moderate to high burnout was significant. Participants in the Wristen (2023) study demonstrated scores more consistent with burnout compared to normative means for postsecondary faculty, but the difference was not significant. Wristen suggested that a context effect might explain these findings. Perceptions of burnout may have been temporarily and relatively lower, since in the semester during which data were collected, faculty had returned to in-person instruction following several semesters of online and hybrid instruction imposed by the pandemic. Men in the Hamann et al. (1988) study reported significantly higher burnout in the DP and PA dimensions compared with women, though there was no difference for EE by gender. In a group of music faculty at a single institution, Bernhard (2007) found a moderately strong correlation between more hours invested in class preparation and higher EE and DP, but no statistically significant differences according to tenure status.

While not widely studied, examining burnout among music faculty is worthwhile, due to the high potential for burnout within the discipline as highlighted by Nápoles (2022). Both Bernhard (2007) and Wristen (2023) examined relatively small numbers of participants (N = 36 and N = 27, respectively) confined to one institution or small geographic region. Hamann et al., employed a more representative research sample (N =50) drawn from throughout the United States, but that study is more than 35 years old, conducted prior to the 2020 global pandemic and its ensuing stressors. It is also possible that there have been demographic shifts among university music faculty since both Hamann et al. (1988) and Bernhard (2007). The purpose of the present study was to investigate the experience of burnout among a large sample of university music faculty throughout the United States. Specifically, the research questions were:

  1. Does the rate of burnout among U.S. university music faculty differ from normative scores for U.S. cross-disciplinary faculty in each burnout dimension: Emotional Exhaustion (EE), Depersonalization (DP), and Personal Accomplishment (PA)?
  2. How did demographic characteristics relate to burnout experience in the three dimensions (EE, DP, and PA)?

Methodology

Participant Recruitment

Criteria for study participation included: being nineteen years of age or older; being currently employed as full- or part-time faculty in a department/school of music at a college or university; having no more than 10% administrative job apportionment; and being currently located in the United States of America. Invitation to participate was sent via email, using The College Music Society email list, which includes every music faculty member throughout the United States, with a follow up invitation extended two weeks prior to study closing. A total of 310 university music faculty participated in this study, consenting through agreement with the first item on the online questionnaire, consistent with the researcher’s approved institutional review board protocol. Upon completing the survey, participants were offered an optional opportunity to enter a drawing for an Amazon gift card following completion of the questionnaire. Identifying information collected for the drawing utilized a separate online survey to maintain data anonymity. 

Research Instruments

Data were collected from mid-October to mid-December 2023 via a secure Qualtrics™ survey consisting of three instruments: demographic questions, the Maslach Burnout Inventory (Maslach et al., 2018), and the University Faculty Job Perceptions Inventory (UFJPI), an inventory developed by the researcher for a separate study. Data collected via the UFJPI instrument were not examined in the present study.

Demographic items included: age; gender identity; living arrangement; dependent status; household earner status; employment status; tenure status; academic rank; total years of experience; years of experience at current institution; institution type (two-year/community college, four-year private, or four-year public); types of students taught (non-degree, associate degree, bachelor’s degree, master’s degree, doctoral degree); hours of teaching activities per week; hours of research/creative activities per week; hours of service/outreach activities per week; total number of hours spent on all college/university job activities per week; and total number of hours at all employment per week. These items addressed characteristics implicated in burnout studies of K–12 and/or university faculty as described in the literature.

The Maslach Burnout Inventory (MBI), an inventory measuring burnout in three separate dimensions—Emotional Exhaustion (EE), Depersonalization (DP), and Personal Accomplishment (PA)—was used to measure burnout due to its established validity and because it provides normative mean scores for post-secondary educators (across academic disciplines) for each burnout subscale. To avoid response bias, there was no reference to burnout in the title of the project, the consent letter, or in any of the questions composed by the researcher. 

Analytical Procedures

Following the recommendations provided in the 2018 Maslach Burnout Inventory Manual, 4th Edition, mean participant scores were calculated for the three burnout subscales of the MBI (EE, DP, and PA). These scores were then compared against the normative mean scores for post-secondary educators provided in the MBI manual using a simple one-sample z-test calculator available on Statistics Kingdom (https://www.statskingdom.com).[1] A z-test is a procedure used to determine whether the mean of a research population is significantly different from that of a reference population when the standard deviation is known and the sample size is large.

To address the second research question, two procedures were employed, both of which adopted an α= 0.05 to test for significance. To avoid potential reidentification of participants through data analysis, no demographic items with fewer than four responses were included when exploring relationships. Unfortunately, gender other than man/woman, several minority/race identifications, and some ethnic status items fell below this threshold.

To explore research question 2, the researcher first employed three separate stepwise multiple linear regression models[2] Stepwise multiple regression is a way to understand how many different independent variables (in this case, demographic characteristics) influence something else (in this case, each of the three burnout dimensions: EE, DP, and PA). Each predictor variable is added into the model individually while holding the others constant to minimize covariance and determine which characteristics are significantly influencing the outcome. to test the effect of each demographic predictor variable on each of the outcome variables: Emotional Exhaustion (EE), Depersonalization (DP), and Personal Accomplishment (PA). Stepwise multiple regression was used instead of force entry (or standard) multiple regression because there was no previous theory regarding which demographic factors most strongly predicted each burnout dimension score among university music faculty. There were 30 independent demographic variables, with a high possibility of covariance. Using stepwise multiple regression made it possible to determine which demographic variables significantly contributed to each dependent variable (EE, DP, and PA) while holding the other predictors constant.

Dummy variables were created from the demographic categorical variables to be used as predictors in the stepwise regression model. There were 22 dummy variables: gender_woman; race_asian; race_black; race_hispanic; race_biracial; single_living alone; single_living with others; married_living alone; dependents; sole earner; employed part-time; nontenured; pre-tenured; no academic rank; assistant professor; associate professor; four-year private institution; two-year community college/institution; teaching non-degree students; teaching associate degree students; teaching masters students; and teaching doctoral students. Items with the highest response frequency were adopted to comprise the reference group: gender_man; race_white; married_cohabitating; no dependents; not sole earner; employed full-time; tenured; professor; four-year public institution; and teaching bachelor’s degree-seeking students. In addition to the 22 categorical variables, there were eight continuous and ordinal variables included as predictor variables in the stepwise multiple regression models. Continuous variables were total number of years employed and years employed at current institution. Ordinal variables were age; weekly teaching hours; weekly research/creative activity hours; weekly service/outreach hours; total weekly hours at college/university; and total weekly hours at all employment.

As a secondary procedure, the researcher developed ANOVA[3] ANOVA (analysis of variance) is a statistical test specifically used to examine means between multiple groups. The F statistic expresses a ratio and determines whether there are significant relationships between all groups tested. models to further explore differences between groups suggested by findings from the stepwise multiple regression procedure. Additional regression models were developed to check for demographic factors that moderated ANOVA findings.[4] Post-hoc tests are used following the initial ANOVA test to determine which groups are significantly different from one another. When examining individual pairwise relationships within a multiple group model, as was done in this case, the t statistic is the value used to calculate significance. The larger the t value, the larger the difference between groups. A regression model can be used to predict the value of a dependent variable based on the value of one or more independent variables. Regression models also describe the strength and character of such relationships. Positive correlation values indicate that as one variable increases, so does the other. Negative correlations show an inverse relationship, where one variable increases as the other decreases.

Results

Participant Demographics

Most participants (n = 265) identified as White, with the remainder identifying as Asian/Asian American (n = 12); Hispanic or Latino/a/x (n = 9); Black/African American (n = 5); Bi-racial/multi-racial (n = 5); Other (n = 3); or preferring not to answer (n = 11). Gender representation between men (n = 154) and women (n = 149) was almost even, with three participants identifying as non-binary and four declining to answer. Aside from a low number (1.7%) of participants aged 20–29—which was anticipated due to the typical requirement of a terminal degree for college/university teaching—representation was also spread relatively evenly between participants aged 30–39 (21.8%), 40–49 (32.4%), 50–59 (23.2%), and 60–69 (20.8%). Regarding marital status and living situation, most (n = 222) participants reported being married/partnered and cohabiting; 13 were married/partnered but living alone; 56 were single and living alone; 14 were single but living with others; and five participants indicated other living arrangements. Most participants (63.5%) did not have any dependents in their households. Roughly a third (n = 103) of participants were the sole income earner for their households.

Most participants (79.4%) were employed full-time at their college/university, with the remainder reporting part-time university employment. Most participants taught at four-year private (n = 126) or four-year public (n = 165) institutions, with only sixteen participants teaching at two-year or community colleges. A wide range of total hours spent per week on college/university job activities was reported: 0–10 hours (n = 19); 11–20 hours (n = 19); 21–30 hours (n = 25); 31–40 hours (n = 66); 41–50 hours (n = 92); 51–60 hours (n = 59); 61–70 hours (n = 19); and 71 or more hours (n = 11). A similar distribution was observed regarding total numbers of hours per week across all employment (including additional hours worked outside the college/university), with participants reporting 0–10 hours (n = 2); 11–20 hours (n = 5); 21–30 hours (n = 17); 31–40 hours (n = 54); 41–50 hours (n = 99); 51–60 hours (n = 80); 61–70 hours (n = 31); and 71 or more hours (n = 22). There were 116 nontenured participants, 51 pre-tenured, and 143 tenured faculty, with academic rank reported as No Rank (n = 79); assistant professor (n = 72); associate professor (n = 69); and professor (n = 90). Of those faculty reporting No Rank status, 67.1% were part-time employees and 32.9% were full-time faculty.

Burnout Scores (Research Question One)

Mean sums for the burnout subscales for participants were Emotional Exhaustion (EE) = 26.75, Standard Deviation (SD) = 13.47; Depersonalization (DP) = 6.44, SD = 5.99; and Personal Accomplishment (PA) = 37.55, SD 6.58. Z-test comparisons of these scores with those of 635 university faculty across disciplines published in the Maslach Burnout Inventory Manual, 4th edition (2018) showed that the university music faculty in the present study had scores significantly more consistent with burnout across all three subscales compared to the population reference group: EE (z = 12.032, p < .001; Cohens d = 0.68); DP (z = 2.407, p = .021; Cohens d = 0.13); and PA (z = -3.596, p < .001; Cohens d = 0.20). While the effect sizes for both PA and DP were small, EE demonstrated a moderate effect size. Pearson correlations were examined between the three burnout subscales among the music faculty, with strongly significant relationships apparent. EE and DP subscales demonstrated a strong positive relationship (p = <.001; r = .600). As expected, the PA subscale was inversely related to both EE (p = <.001, r = -.310) and DP (p = <.001, r = -.380). The direction of these relationships demonstrates the configuration most consistent with burnout experience as described in the Maslach Burnout Inventory manual: high EE and DP with low PA.

Relationships between Demographic Characteristics and Burnout Subscale Scores (Research Question Two)

Stepwise Multiple Regression: Emotional Exhaustion

A stepwise multiple regression analysis was performed to identify significant demographic predictors of Emotional Exhaustion (EE) for university music faculty members. The regression model was statistically significant, F (6, 255) = 11.243, p < .001, which explained approximately 21% of the variance in EE (adjusted R² = .191). Significant predictors included teaching associate degree-seeking students (B = -11.756, β = -.236, t = -4.111, p < .001), four-year private institution (B = -4.205, β = -.155, t = -2.732, p < .01), assistant professor rank (B = -5.915, β = -.195, t = -3.115, p < .01), age (B = -2.107, β = -.195, t = -3.089, p < .01), weekly teaching hours (B = 1.878, β = .223, t = 3.776, p < .001), and total weekly hours all employment (including college/university hours and any additional external hours (B = 1.256, β = .121, t = 2.054, p < .05). Results suggest that while both higher teaching hours and higher total hours are associated with greater emotional exhaustion, being older, working at a four-year private institution, teaching associate students, or being an assistant professor are associated with reduced emotional exhaustion burnout symptoms. 

More specifically, for every five-hour (one unit of teaching activity per week) increase in teaching activity, EE was predicted to increase by 1.878 units, holding the rest of all other significant predictors (teaching associate degree students, four-year private institution, assistant professor, age, and total hours all employment) constant. Similarly, for each ten-hour increase in the total weekly hours of employment, EE was predicted to increase by 1.256 units. The other significant predictors demonstrated a negative relationship with EE. Faculty teaching associate degree students were predicted to have 11.756 units less of EE than those who taught bachelor’s degree-seeking students, holding the rest of all other predictors constant. Participants working in four-year private institutions were likely to experience 4.205 units less of EE than those who worked in four-year public institutions, and assistant professors experienced 5.915 less units of EE than the “full” professors, holding the rest of all other predictors constant. For each ten-year increase in participant’s age, EE was predicted to decrease by 2.107 units, holding the rest of all other predictors constant. 

Stepwise Multiple Regression: Depersonalization

Depersonalization (DP) was similarly explored via a stepwise multiple regression analysis. The regression model was statistically significant, F (5, 256) = 9.724, p < .001, which explained approximately 16% of the variance in DP (adjusted R² = .143). Significant predictors included age (B= -1.404, β = -.289, t = -4.515, p < .001), weekly teaching hours (B = .54, β = .143, t = 2.356, p < .05), total weekly hours all employment (B = .825, β = .178, t = 2.947, p < .01), married living alone (B = 4.374, β = .148, t = 2.583, p < .05), and assistant professor rank (B = -2.128, β = -.157, t = -2.445, p < .05).

Both increased weekly teaching hours and weekly hours spent at all employment predicted higher DP. For each five-hour increase in teaching activity per week, DP was predicted to increase by .54 unit, holding rest of all other predictors constant. For each ten-hour increase in the total weekly hours spent at all employment, DP was predicted to increase by .825 unit, holding the rest of all other predictors constant. Married living alone participants were likely to experience 4.374 units more of DP than married cohabitating participants, holding the rest of all other predictors constant. Assistant professors were likely to experience 2.128 unit of less DP than the “full” professors, holding rest of all other predictors constant. For every ten-year increase in participant age, DP was predicted to decrease by 1.404 units, holding the rest of the predictors constant.  

Stepwise Multiple Regression: Personal Accomplishment

A stepwise multiple regression analysis was also performed to identify significant demographic predictors of Personal Accomplishment (PA). The regression model was statistically significant, F (3, 258) = 6.307, p < .001, which explained approximately 6.80% of the variance in EE (adjusted R² = .057). Significant predictors included age (B = 1.277, β = .239, t = 3.651, p < .001), single living with others (B = 4.993, β = .153, t = 2.531, p = .012), and pre-tenured (B = 2.299, β = .139, t = 2.120, p = .035). For each ten-year increase in participant’s age, PA increased by 1.277 units, holding other predictors constant. Single music faculty members who lived with others experienced 4.993 units higher PA than married cohabitating participants, and pre-tenured faculty were likely to experience 2.299 units higher PA than tenured professors, holding all other predictors constant.

ANOVAs Testing Differences between Demographic Groups

The stepwise multiple regression for each burnout dimension—EE, DP, and PA—suggested that differences might exist between groups in terms of academic rank and institution type. Since assistant professors had lower EE and DP, ANOVAs were performed to check for group differences in EE, DP, and according to academic rank. None of these models were significant. However, because ANOVA tests raw group differences without controlling other variables, it is possible that the effect of academic rank might have been masked or confounded by other variables. Accordingly, additional ANOVAs were performed to examine academic rank and age, academic rank and teaching hours, and academic rank and total university hours to further explore possible reasons for the finding that assistant professors had lower EE and DP compared with “full” professors. Results demonstrated that “full” professors were significantly older than both assistant and associate professors, F (3, 306) = 50.392, p < .001. ANOVAs for academic rank and teaching hours and academic rank and total university hours were not significant.

An ANOVA examining EE by institution type (four-year public, four-year private, or two-year/community college) was significant, F (2, 304) = 10.88 , p < .001. Pairwise t-tests showed significant differences in EE burnout scores according to institution type, with participants teaching in four-year public institutions showing scores most consistent with burnout. The largest effect size was observed between four-year public institutions and two-year/community colleges, t(304) = 3.97, p <.001, although music faculty at four-year public institutions also had significantly higher EE than those teaching at four-year private schools, t(304) = 3.18, p = .004. Music faculty employed at four-year private institutions also had higher EE than music faculty teaching at two-year/community colleges, t(304) = 2.50, p =.035.

In light of these differences in Emotional Exhaustion between faculty at the various types of institutions, and since institution type has not been previously well-explored as a burnout factor, new linear regression models were tested to probe whether the main effect of EE by institution type was moderated by other demographic characteristics, including age; total number of years of experience; living arrangements; dependents; total weekly hours worked at college/university job; total weekly hours worked at all employment; hours spent on teaching activities (including preparation, delivery, and assessment); hours spent on research/creative activities; and hours spent on service/outreach. All moderation models for institution type used two-year/community colleges as the reference group. The only significant moderators emerging in these models were number teaching hours per week and total number of hours worked weekly at all employment. These two moderators influenced EE among four-year public music institution faculty in comparison to other groups. None of the demographic characteristics examined moderated the difference between the four-year private and two-year institutions; instead, accounting for these demographics lessened the difference in EE between these groups, in some cases to a null difference.

Hours spent on teaching activities per week moderated the finding for Emotional Exhaustion by institution type, F(5, 301) = 11.58, p <.001. Pairwise comparisons showed that more hours spent on teaching activities significantly predicted greater emotional exhaustion for music faculty members employed at four-year public colleges/universities compared to those teaching at two-year institutions, t(301) = 2.11, p = .036. In other words, more hours spent on teaching activities amplified the difference in EE between faculty at four-year public and two-year institutions. A separate regression model demonstrated that total weekly hours worked at all employment similarly moderated EE scores among faculty at different types of institutions, F(5, 301) = 7.69, p <.001, with more hours worked at all employment amplifying the difference in EE between four-year public institution music faculty and faculty at two-year/community college institutions, t(301) = 2.02, p = .044.

Given these findings, the researcher was interested in whether faculty at different types of institutions were simply working a significantly different number of hours, either at their university/college job or at all employment. Post-hoc t- tests between groups showed that music faculty at four-year public institutions were working more hours weekly at their university/college compared to either two-year (t(304) = 4.49, p < .001) or four-year private institution music faculty (t(304) = 5.54, p < .001). There was no significant difference between hours worked per week at the university/college job between four-year private and two-year/community college faculty. However, regarding all employment, both within and external to the university, there were significant differences between all three groups. Faculty at four-year private institutions worked more hours (all employment) weekly than those at two-year/community college institutions, t(304) = 2.87, p = .0124. However, music faculty at four-year public institutions worked more hours than those at either four-year private institutions (t(304)= 2.91, p = .0107) or two-year/community colleges (t(304) = 4.22, p < .001). Music faculty teaching at four-year public institutions were indeed working more hours weekly, both at their college/university job, and at all employment.

Discussion

In contrast to the observations of previous studies of burnout among music faculty (Hamann et al., 1988; Wristen, 2023), burnout scores in each dimension were significantly higher among university music professors than the normed scores of professors across various disciplines. This is unsurprising given the numerous job stressors that music professors face. These include not only daily stressors—including long and often irregular hours; diminishing resources, including financial support; overwork; close mentorship relationships with students; and, often, managing the demands of one’s own schedule of performances—but also more existential concerns such as continual challenges with university funding, erosion of faculty lines, and other systemic issues exacerbated by the pandemic. The moderate effect size for EE compared with other post-secondary educators indicates that this dimension of burnout may be of particular concern for university music educators.

As discussed in the introduction, findings regarding relationships between full- versus part-time employment and burnout are mixed. There were no significant relationships between full-time or part-time employment status and EE identified in the present study. Similarly, having no academic rank (i.e., adjunct or contingent faculty) did not predict higher burnout scores in any burnout dimension. As was the case in most of the studies of faculty across disciplines reviewed in Sabagh et al. (2018), but in contrast to Hamann et al.’s (1988) study of music faculty, gender identity did not emerge as a significant predictor for any dimension of burnout. There were also no significant relationships observed between any burnout dimension and race or ethnic status, having dependents at home, income earner status, total years of experience, or years employed at current institution.

Living arrangements impacted DP and PA, but in different ways. Those married but living alone reported significantly higher DP. This finding may indicate that those who are married but separated from their spouses have a higher degree of cynicism/depersonalization in their personal lives that influences their regard for students or colleagues in the workplace. The finding of higher PA on the part of music faculty who are single but living with others is interesting. While it is logical to assume that being able to share significant accomplishments with someone else in the household might give rise to a more awareness and celebration of personal accomplishment, no significant relationship was observed for those who were married and cohabitating with their spouse and PA, suggested there may be additional factors that influence this relationship.

While working a greater number of hours weekly at the college/university job did not emerge as predictive of any burnout dimension, working more hours at all employment—a separate item which presumably included hours worked beyond the college/university job—predicted both higher EE and DP. Thus, working more hours per week overall was a driving factor for burnout among participants. This finding amplifies Demerouti (2014) regarding the risks of having insufficient recovery time between work hours and underscores the risk of burnout within a professional culture that expects and often even rewards overwork, to the detriment of both individual music faculty and, in the longer term, to music institutions.

The type of work that constitutes those hours per week is also important. Consistent with Bernard (2007), hours spent teaching—including all preparation, delivery, and assessment—clearly and strongly predicted both EE and DP across all college/university music faculty. The same was not true for hours spent on research/creative activities, hours spent on service activities, or even on the total number of weekly hours worked at the university/college, which suggests that teaching allocation is a particularly ripe area for institutional intervention.

While all employees in “helping” professions involving close human interactions are more vulnerable to burnout over time (Byrne, 1999; DuBois & Mistretta, 2020; Johnson et al., 2005), the student-mentor relationships that characterize collegiate music study demand substantial and sustained time investment and emotional energy. Music professors are highly invested in and involved with their students, typically spending many hours supporting students’ participation in supplemental educational/professional activities (festivals, masterclasses, auditions, grant applications etc.,), and often serving as first responders to direct students in crisis to appropriate support services (health, counseling, financial, academic, etc.). Academic music programs face numerous financial stressors and are currently confronting increasing reductions in resources including both faculty and support staff positions. When faculty leave or retire, or faculty lines are eliminated, their work responsibilities may be assumed by remaining faculty. Additionally, music institutions often require duties not captured in the faculty member’s formal work allocation, such as advising, supervising graduate student research or degree committees, attending and grading recitals, participating in auditions or other institutional events, or teaching independent studies. Lack of sufficient support staff often places additional logistical pressures on faculty, such as troubleshooting technology, addressing university facility issues, and arranging teaching, performing, or recording facilities. Since university music programs are widely understood to require significant financial resources to operate (equipment, facilities, low student-teacher ratios, etc.), there may be a tendency to underrepresent the true staffing needs of a music department/school to higher administrations to avoid further cuts to programming or budgets. However, sustained demands for music faculty members to undertake invisible teaching are likely to further increase EE and DP, leading to more faculty burnout and thus risking more faculty attrition.

Consistent with several previous studies of university faculty across academic disciplines (Evers et al., 2002; Kinman & Wray, 2013; Reevy & Deason, 2014), increasing age was inversely related to both EE and DP and positively correlated with higher PA, a configuration suggesting being older might protect against burnout in each dimension among music faculty. This finding may reflect that the older music faculty in this study had developed mechanisms over time to cope with job stressors and thereby avoid burning out, or may simply represent a survivor bias in which the older faculty represented in this study were those who had not burned out. Since total years of employment--either overall or at the current institution--did not significantly predict burnout in any dimension, experience alone does not explain this finding. There are also additional nuances to this finding as it intersects with academic rank.

The stepwise multiple regressions indicated that assistant professors experienced less EE and DP than the reference group of “full” professors, and that those identifying as pre-tenured (presumably mostly assistant professors) experienced more PA compared to the reference group of tenured faculty. These observations contrast with findings from previous studies, most of which have not identified a relationship between academic rank and burnout as discussed in the introduction. While some studies (Azeem & Nazir, 2008; Chronicle of Higher Education, 2020; Tümkaya, 2007) have implicated the pre-tenure period as particularly stressful for university faculty, it is possible that higher stress may not lead to increased burnout. Given that assistant professors were significantly younger than “full” professors, the observation that assistant professors had less EE and DP burnout than “full” professors seems at odds with the finding regarding increasing age protecting against burnout in every dimension. There are several possible explanations for this finding.

The secondary ANOVA examining differences in burnout between faculty ranks was not significant. However, it is possible that differences between groups may have existed. Stepwise multiple regression holds other variables constant, allowing for examination of individual factors while ANOVA tests raw group differences without controlling other variables. On the ANOVA, the relationship between academic rank and EE and DP may have thus been masked or confounded by other variables. While ANOVAs examining weekly teaching hours and total university hours per week by academic rank were not significant, the “full” professors were generally teaching fewer hours per week than assistant or associate professors. Perhaps “full” professors were simply teaching less, working fewer hours overall, and were older—all factors that protect against burnout—such that the ANOVA did not show significant group differences for EE, DP, or PA. Alternatively, perhaps there were uncaptured factors at work. There may be unique features of assistant professor rank that may protect against burnout. For example, sometimes institutions limit assistant professors’ service responsibilities so they have time to establish their teaching and their research/creative agendas. Perhaps assistant professors have other qualities not measured in this study, such as higher idealism or stamina, which in turn protects them against Emotional Exhaustion and Depersonalization despite their younger age. While subsequent ANOVAs were not useful in explaining the finding, assistant professors clearly had significantly less EE and DP than “full” professors. Both age and academic rank clearly influenced burnout, but each of these variables functioned independently in the model. It is likely that age and academic rank are confounded by other factors in their contribution to burnout.

Differences regarding burnout among music faculty teaching at different institution types are particularly intriguing, especially since institution type has not been previously well-studied or differentiated in relation to burnout in any academic discipline, much less among music faculty specifically. Faculty teaching at a four-year private music institutions had less EE compared with those teaching in a four-year public colleges or universities, a finding amplified by the ANOVA exploring differences between faculty at all types of institutions. This observation was bolstered by the finding that teaching associate degree-seeking students (degrees typically conferred by two-year institutions) significantly and strongly predicted lower EE scores. Post-hoc t-tests showed that four-year public institution faculty were working more hours in general—both at their college/university jobs and at additional employment—compared to both four-year private and two-year institution faculty. Notably, it was the number of hours worked per week at all employment that moderated the higher EE among four-year public university music faculty; the number of hours worked per week at the university job did not significantly moderate this finding. This suggests not only that teaching at a four-year public college/university may put music educators at higher risk for EE burnout, but that this risk increases even more as total number of work hours per week increases. Music faculty teaching at four-year public institutions were undertaking more employment outside their universities compared to faculty at other types of institutions. It is possible that faculty at four-year public universities were working more external hours to seek validation/recognition or increase their sense of career achievement. It is also possible that this group of music faculty had concerns about salary that caused them to seek more outside employment, possibilities that will be explored in a future study.

Caution must be exercised when interpreting findings from observational studies of this type, particularly cross-sectional studies (i.e., where data collected represent a single point in time). It is possible, even likely, that burnout experiences in various dimensions may fluctuate over time. Normative data provided by the Maslach Burnout Inventory manual for the reference group of university professors across disciplines was published in 2018, while the present data were collected in Fall 2023. It is possible that burnout scores among professors across all disciplines have increased in the interim. Additional studies using data collected from professors across all disciplines would allow a more robust comparison of how the experience of burnout varies among professors according to academic discipline. While the researcher made every effort to avoid self-reporting bias by carefully avoiding the word burnout on the research instruments and referring to the research as a study of work attitudes and conditions, it is also possible that participants may have over- or under-represented their feelings as measured by the Maslach Burnout Inventory. While the entire questionnaire required only 15–20 minutes to complete, it is also possible that participants may not have read each item carefully, though the valence of wording was mixed unpredictably from item to item to minimize this possibility.

Since there was a small number (n = 16) of two-year/community college music faculty in this study, it is possible their experiences were not representative of this group. Additionally, there were insufficient numbers of participants identifying in any gender category other than man or woman to examine burnout among these groups. Similarly, given that 85.5% of the participants identified as White, a robust examination of differences in burnout experience between various racial or ethnic groups was not possible in the present study. Exploring such differences in future studies would add needed nuance in understanding how burnout experiences differ among music faculty.

Finally, the relatively low R-squared[5] R-squared is a measure of how well all variables combine together in the model to explain the outcome variable. statistics observed for each of the three stepwise regression models indicate there are additional factors beyond the demographic factors examined that predict high burnout scores. While stepwise multiple regression identifies the optimal set of predictors for a model, it does not control for covariance and in fact can inflate the problem. It is possible that some demographic variables examined in this study were removed from the three stepwise regression model that were important, but whose statistical significance may have been impacted by covariance with other study variables. It is likely that there are additional factors not directly examined in this study—for example, differences in job descriptions/responsibilities, resources/support, work climate, or individual physical health status, stamina, or disposition—that influence the development of burnout among music faculty. A low adjusted R-squared is only a concern when developing a model to comprehensively account for all the variables that predict an outcome. It does not impact validity when using stepwise regression as a means of exploring whether significant relationships between predictor and outcome variables exist as was the case in the present study.

Even acknowledging these limitations, this study demonstrated that burnout in every dimension was significantly higher among music faculty drawn from throughout the United States in comparison with normative scores for post-secondary faculty across disciplines, with Emotional Exhaustion standing out as a burnout dimension of particular concern for this group. Additionally, significant relationships between various demographic characteristics and burnout dimensions emerged. Considered together, these results illuminate several areas for potential institutional consideration and targeting. In particular, a higher number of weekly teaching hours predicted higher burnout in the EE and DP dimensions, pointing to the need for music administrators to reconsider and rebalance music faculty teaching loads. The total number of all hours worked per week also strongly predicted EE and DP. These factors may be particularly crucial for music faculty teaching at four-year public institutions, who demonstrated higher Emotional Exhaustion compared to those teaching at other types of institutions. Music faculty at four-year institutions are teaching more hours per week and working more hours at all employment, and these factors further amplified EE among this group. To illuminate factors beyond work hours and demographic characteristics, the researcher will next examine how these music faculty members’ perceptions about their job conditions predict their scores in the three burnout dimensions. It would also be useful to study music faculty burnout compared to that of academic faculty in other disciplines. Continued studies along these lines will help identify and elucidate internal and external factors that most strongly predict burnout among music faculty.

 

[1] A z-test is a procedure used to determine whether the mean of a research population is significantly different from that of a reference population when the standard deviation is known and the sample size is large.

[2] Stepwise multiple regression is a way to understand how many different independent variables (in this case, demographic characteristics) influence something else (in this case, each of the three burnout dimensions: EE, DP, and PA). Each predictor variable is added into the model individually while holding the others constant to minimize covariance and determine which characteristics are significantly influencing the outcome.

[3] ANOVA (analysis of variance) is a statistical test specifically used to examine means between multiple groups. The F statistic expresses a ratio and determines whether there are significant relationships between all groups tested.

[4] Post-hoc tests are used following the initial ANOVA test to determine which groups are significantly different from one another. When examining individual pairwise relationships within a multiple group model, as was done in this case, the t statistic is the value used to calculate significance. The larger the t value, the larger the difference between groups. A regression model can be used to predict the value of a dependent variable based on the value of one or more independent variables. Regression models also describe the strength and character of such relationships. Positive correlation values indicate that as one variable increases, so does the other. Negative correlations show an inverse relationship, where one variable increases as the other decreases.

[5] R-squared is a measure of how well all variables combine together in the model to explain the outcome variable.

 

References

  1. Azeem, S. M. & Nazir, N. A. (2008). A study of job burnout among university teachers. Psychology and Developing Societies, 20 (1), 51–64. 10.1177/097133360702000103
  2. Bakker, A. B. & Demerouti, E. (2017). Job demands-resources theory: Taking stock and looking forward. Journal of Occupational Health Psychology, 22 (3), 273–285. 10.1037/ocp0000056
  3. Bakker, A. B. & Schaufeli, W. B. (2000). Burnout contagion processes among teachers. Journal of Applied Social Psychology, 30, 1–21. 10.1111/j.1559-1816.2000.tb02437.x
  4. Bakker, A. B., Schaufeli, W. B., Sixma, H., & Bosveld, W. (2001). Burnout contagion among general practitioners. Journal of Social and Clinical Psychology, 20, 82–98. 10.1521/jscp.20.1.82.22251
  5. Bernhard, H. C. (2007). A survey of burnout among university music faculty. College Music Symposium, 47, 117–126. 40374508
  6. Bilge, F. (2006). Examining the Burnout of Academics in Relation to Job Satisfaction and Other Factors. Social Behavior and Personality: An International Journal, 34 (9), 1151–1160. 10.2224/sbp.2006.34.9.1151
  7. Blix, A. G., Cruise, R. J., Mitchell, B. M., & Blix, G. G. (1994). Occupational stress among university teachers. Educational Research, 36 (2), 157–169. 10.1080/0013188940360205
  8. Byrne, B. M. (1991). Burnout: Investigating the impact of background variables for elementary, intermediate, secondary, and university educators. Teaching and Teacher Education, 7 (2), 197–209. 10.1016/0742-051X(91)90027-M
  9. Byrne, B. M. (1999). The nomological network of teacher burnout: A literature review and empirically validated model. In R. Vandeberthe, & A. Michael Huberman (Eds.), Understanding and preventing teacher burnout: A sourcebook of international research and practice (pp. 15–37). Cambridge, UK: Cambridge University Press.
  10. Chronicle of Higher Education with support from Fidelity Investments (2020). On the verge of burnout: Covid-19’s impact of faculty well-being and career plans. Retrieved February 3, 2021 from https://connect.chronicle.com/rs/931-EKA%20218/images/Covid%26FacultyCareerPaths_Fidelity_ResearchBrief_v3%20%281%29.pdf
  11. Demerouti, E. (2014). Individual strategies to prevent burnout. In Leiter, M.P., Bakker, A.B., and Maslach, C. (Eds.), Burnout at work (pp. 32–55). New York: Psychology Press.
  12. Dubois, A. L. & Mistretta, M. A. (2020). Overcoming burnout and compassion fatigue in schools: A guide for counselors, administrators, and educators. New York: Routledge.
  13. Evers, W., Brouwers, J. E., & Tomic, W. (2002). Burnout and self-efficacy: A study on teachers’ beliefs when implementing an innovative educational system in the Netherlands. British Journal of Educational Psychology, 72, 227–243. 10.1348/000709902158865
  14. Faculty wellness and careers (2020). Retrieved Jan. 28, 2021 from https://www.coursehero.com/blog/faculty-wellness-
  15. Fernet, C., Guay, F., & Senécal, C. (2004). Adjusting to job demands: The role of work self-determination and job control in predicting burnout. Journal of Vocational Behavior, 65 (1), 39–56. 10.1016/S0001-8791(03)00098-8
  16. Ghorpade, J., Lackritz, J., & Singh, G. (2011). Personality as a moderator of the relationship between role conflict, role ambiguity, and burnout. Journal of Applied Social Psychology, 41 (6), 1275–1298. 10.1111/j.1559-1816.2011.00763.x
  17. Gillespie, N. A., Walsh, M., Winefield, A. H., Dua, J., & Stough, C. (2001). Occupational stress in universities: Staff perceptions of the causes, consequences and moderators of stress. Work & Stress, 15 (1), 53–72. 10.1080/02678370117944
  18. Golkar, A., Johansson, E., Kasahara, M., Osika, W. Perski, A., & Savic, I. (2014). The influence of work-related chronic stress on the regulation of emotion and on functional connectivity in the brain. PLoS ONE, 9(9), 1–11. 10.1371/journal.pone.0104550
  19. Gonzalez, S. & Bernard, H. (2006). Academic workload typologies and burnout among faculty in Seventh-day Adventist colleges and universities in North America. Journal of Research on Christian Education, 15 (1), 13–37. 10.1080/10656210609484992
  20. González-Morales, G., Pieró, J. M., Rodríguez, I., & Bliese, P. D. (2012). Perceived collective burnout: A multilevel explanation of burnout. Anxiety, Stress & Coping, 25(1), 43–61. 10.1080/10615806.2010.542808
  21. Hamann, D. L., Daugherty, E., & Sherbon, J. (1988). Burnout and the college music professor: An investigation of possible indicators of burnout among college music faculty members. Council for Research in Music Education, 98, 1–21. 40318240.
  22. Han, J., Perron, B. E., Yin, H., & Liu, Y. (2020a). Faculty stressors and their relations to teacher efficacy, engagement and teaching satisfaction. Higher Education Research & Development. 10.1080/07294360.2020.1756747.
  23. Han, J., Yin, H., Wang, J., & Zhang, J. (2020b). Job demands and resources as antecedents of university teachers’ exhaustion, engagement and job satisfaction. Educational Psychology, 40(3), 318–335. 10.1080/01443410.2019.1674249
  24. Han, J., Yin, H., Wang, J., & Bai, Y. (2020c). Challenge job demands and job resources to university teacher well-being: The mediation of teacher efficacy. Studies in Higher Education, 45(8), 1771–1785. 10.1080/03075079.2019.1594180
  25. Johnson, S., Cooper, C., Cartwright, S., Donald, I., Taylor, P., & Millet, C. (2005). The experience of work-related stress across occupations. Journal of Managerial Psychology, 20 (2), 178–187. 10.1108/02683940510579803
  26. Kinman, G. & Wray, S. (2013). University and College Union (UK). Higher stress: A survey of stress and well-being among staff in higher education. Retrieved Jan. 28, 2021 from www.ucu.org.uk.
  27. Lackritz, J. R. (2004). Exploring burnout among university faculty: Incidence, performance, and demographic issues. Teaching and Teacher Education, 20 (7), 713–729. 10.1016/j.tate.2004.07.002
  28. Li, Y., Li, J., & Sun, Y. (2013). Young faculty job perceptions in the midst of Chinese higher education reform: The case of Zhejiang University. Asia Pacific Journal of Education, 33 (3), 273–294. 10.1080/02188791.2013.787388
  29. Liston, C., McEwen, B. S., & Casey, B. J. (2009). Psychosocial stress reversibly disrupts prefrontal processing and attentional control. Proceedings of the National Academy of Sciences of the United States of America, 106 (3), 912–917. 10.1073/pnas.0807041106
  30. Lowenstein, L. (1991). Teacher stress leading to burnout—its prevention and cure. Education Today, 41(2), 12–16. 1991-34571-001
  31. Magarell, J. (1982). Decline in faculty morale laid to governance role, not salary. Chronicle of Higher Education, 25 (11), 1.
  32. Maslach, C. (2003). Burnout: The cost of caring. Los Altos, CA: Malor Books.
  33. Maslach, C., Jackson, S., & Leiter, M. (2018). Maslach Burnout Inventory Manual, 4th Edition. Mind Garden, Inc. www.mindgarden.com.
  34. Maslach, C. & Leiter, M. P. (1997). The Truth About Burnout: How Organizations Cause Personal Stress and What to Do About It. San Francisco: Jossey-Bass.
  35. McClenahan, C. A., Giles, M. L., & Mallett, J. (2007). The importance of context specificity in work stress research: A test of the demand-control-support model in academics. Work & Stress, 21 (1), 85–95. 10.1080/02678370701264552
  36. Midway, S., Robertson, M., Flinn, S., & Kaller, M. (2020). Comparing multiple comparisons: practical guidance for choosing the best multiple comparisons test. PeerJ, 8, e10387. 10.7717/peerj.10387
  37. Nápoles, J. (2022). Burnout: A review of the literature. Update: Applications of Research in Music Education, 40(2), 19–26. 10.1177/87551233211037669
  38. Özgül, N. & Polat, E. (2018). Burnout levels of academic staff: An investigation at a public university in Turkey. Sakarya University Journal of Science, 22(6), 1752–1759. 10.16984/saufenbilder.392656
  39. Pines, A. M. & Aronson, E., with Kafry, D. (1982). Burnout: From tedium to personal growth. New York: The Free Press.
  40. Reevy, G. M. & Deason, G. (2014). Predictors of depression, stress, and anxiety among non-tenure track faculty. Frontiers in Psychology, 5, 1–17. 10.3389/fpsyg.2014.00701
  41. Rothmann, S. & Barkhuizen, N. (2008). Burnout of academic staff in South African higher education institutions. South African Journal of Higher Education, 22 (2), 439–456. https://hdl.handle.net/10520/EJC37435
  42. Sabagh, Z., Hall, N. C., & Saroyna, A. (2018). Antecedents, correlates and consequences of faculty burnout. Educational Research, 60 (2), 131–156. 10.1080/00131881.2018.1461573.
  43. Savic, I. (2013). Structural changes of the brain in Relation to Occupational Stress. Cerebral Cortex, 1–11. 10.1093/cercor/bht348
  44. Tümkaya, S. (2007). Burnout and humor relationship among university lecturers. Humor, 20 (1), 73–92. 10.1515/HUMOR.2007.004
  45. Watts, J. & Roberson, N. (2011). Burnout in university teaching staff: A systematic literature review. Education Research, 53, 33–50. 10.1080/00131881.2011.552235
  46. World Health Organization. (2018). International classification of diseases for mortality and morbidity statistics (11th Revision). Retrieved May 2, 2021 from https://icd.who.int/browse11/l-m/en.
  47. Wristen, B. (2023). An exploratory study of relationships between job perceptions and university music faculty burnout. College Music Symposium, 63 (2), 1–16. 10.18177/sym.2023.63.sr.11612