Quantitative analysis of workplace morale in academic librarians and the impact of direct supervisors on workplace morale.

Many studies have attempted to measure the workplace morale of academic librarians through related variables such as job satisfaction. Other studies have taken a qualitative or phenomenological approach to understanding workplace morale. This exploratory study has two objectives: (1) to quantitatively measure the workplace morale of academic librarians to set an initial baseline measure for future research and (2) to analyze the variables that impact academic librarian morale with an emphasis on the variables that academic library supervisors directly influence. Findings suggest the workplace morale of academic librarians is predicted by several variables within a supervisor's influence including feedback, work autonomy, and supervisor qualities such as communication, transparency, and empowering subordinates. Outside of a small negative relationship to age and work experience, demographics were largely found to not be a predictor of workplace morale. This study also evaluated workplace morale as a predictor of turnover intention in academic librarians. There was a strong statistically significant negative correlation between workplace morale and the turnover intention of academic librarians.

Introduction

Expanding responsibilities, shrinking budgets, and rapidly evolving patron needs are some of the many ongoing issues impacting today's academic librarians and their workplace morale. Tasked with maintaining or increasing services while dealing with fewer resources and often declining personnel levels, academic library managers must consider the impact of these issues on the workplace morale of their librarians and work to proactively mitigate subsequent issues commonly attributed to low workplace morale such as declines in performance and increased turnover. Outside of a recent qualitative study (Kendrick, 2017) and some studies on constructs similar to workplace morale such as job satisfaction and burnout, very little research exists on the academic librarian workplace morale experience to guide managers in tackling this issue within their workforce.

This exploratory study provides a quantitative measure and analysis of the workplace morale of academic librarians. Past studies have posited workplace morale and related constructs such as job satisfaction as an issue for academic librarians but have not directly measured workplace morale levels. This study will also explore possible predictors of workplace morale including several demographic and institutional data points.

Additionally, a focus will be placed on other practical variables academic library managers influence (e.g., autonomy and feedback). This approach will provide academic library managers with information about factors that may impact workplace morale that they cannot influence and awareness of the factors that impact workplace morale that they can influence with their own behaviors. Last, the study will look at how workplace morale in academic librarians can predict relevant outcomes to library managers by analyzing the relationship between workplace morale and turnover intention.

Background and literature review

It is important to fully understand what is meant by “workplace morale” before any conclusions can be drawn about the workplace morale of academic librarians. Past library studies have measured morale qualitatively or via related constructs such as job satisfaction. This study approaches workplace morale similar to studies on organizational behavior topics. Definitions of workplace morale in the organizational research literature are varied although there is an agreement among researchers that the construct is a latent variable that must be measured through a combination of other attitudinal variables (Subramony et al., 2008). Following several other researchers in this area, this study defines workplace morale as a combination of job satisfaction and affective commitment to the organization (Borden et al., 2018; Harrison et al., 2006; Rosen et al., 2006). Workplace morale, constructed this way, was found to be a predictor of increases in performance and decreases in lateness, absenteeism, and turnover (Harrison et al., 2006). A recent phenomenological study by Kendrick (2017) revealed that academic librarians with low morale engage in several behaviors that are undesirable to the organization including lowering work input, time-wasting, and seeking other employment opportunities.

Job satisfaction, one half of our workplace morale measure, is an employee's attitude towards their job that is influenced by their “liking or disliking the work itself, the rewards (pay, promotions, [and] recognition), or the context (working conditions [and] colleagues)” (VandenBos, 2015, p.572). Job satisfaction is one of the heaviest researched constructs in the organizational psychology literature. Findings related to job satisfaction are mixed but studies have linked high job satisfaction to many positive organizational outcomes including increases in job performance (Riketta, 2008), decreased absenteeism (Farrell & Stamm, 1988), and increased display of organizational citizenship behaviors (Organ & Ryan, 1995).

A recent study on all types of libraries and librarians by Martin (2020) found that the library profession matched the United States national average for job satisfaction based on the Job Satisfaction Survey. Martin (2020) found no difference in overall job satisfaction scores of librarians based on many factors including gender, race, library type, and library position (including management). A study of academic librarians at ARL institutions by Galbraith et al. (2016) found no gender based difference in job satisfaction despite finding that female librarians with faculty status report less work-life balance and higher levels of stress than their male faculty librarian counterparts. Similarly, Becher (2019) found that non-tenure track librarians experience lower levels of job satisfaction and this is especially true if their library employs tenure-track or tenured librarians. Matteson and Kennedy (2016) showed that while an individual librarian's dispositional affect can predict work attitudes such as job satisfaction, library managers may be able to cause temporary increases or situational gains in employee work attitudes through modifications to job roles and the work environment. Job satisfaction brings many benefits to academic libraries and alone should be a topic of interest to academic library managers.

Affective commitment to the organization, the other half of our workplace morale measure, is an employee's “emotional attachment to, identification with, and involvement in, the organization” (Allen & Meyer, 1990, p.1). Studies have linked high affective commitment to several positive outcomes including lower turnover through employees wanting to remain in the organization (Allen & Meyer, 1990; Meyer et al., 2002), decreased withdrawal cognitions (Meyer et al., 2002), decreased work-family conflicts (Meyer et al., 2002), and increased innovativeness (Allen & Smith, 1987). While little research exists on the affective commitment of academic librarians, benefits such as decreased withdrawal cognitions and increased innovativeness should be of particular interest to academic library managers who are increasingly trying to meet new needs with decreased workforces.

One area academic library managers can influence to potentially get more out of their workforce is through changes to the work autonomy of their librarians. Work autonomy in this study is defined by the three individual work autonomy types established by Breaugh (1999). These autonomy types are method autonomy (employee's ability to determine the procedures they use to complete their work), scheduling autonomy (employee's ability to determine when they do their work), and criteria autonomy (employee's ability to create or modify job performance criteria). Work autonomy is important to this study because it is associated with many positive outcomes and the level to which an academic librarian can experience a type of work autonomy is directly influenced by their supervisor. Spector's (1986) meta-analysis of studies focusing on employee's perceived control over their work (i.e., their belief that they have work autonomy) found a relationship between increases in perceived control and increases in job satisfaction, commitment, motivation, and performance, as well as decreases in employee stress and withdrawal behaviors. Thompson and Prottas (2005) found employees with higher job autonomy were more likely to experience positive spillover between work and family, were less likely to feel stressed or experience a work-family conflict, were less likely to consider leaving the organization, and in general were more satisfied with their job, family, and life.

Patillo et al. (2009) found work autonomy to be an important factor in the recruitment and retention of library professionals and work autonomy may be “especially important to highly educated professional workers who desire greater flexibility and control over the content and terms of their work” (p.277). Patillo et al. (2009) also found academic librarians have more autonomy than public librarians and librarians with administrative or information technology roles experience higher levels of autonomy. Yeh and Walter (2016) proposed that the autonomy of librarians and library committees is a crucial factor towards creating innovative solutions to combat the evolving demands placed on academic libraries. Work autonomy's relationship to constructs such as job satisfaction, motivation, commitment, and decreases in stress, point to work autonomy as an important tool for academic library managers to deploy towards increasing performance and potentially workplace morale.

While work autonomy can certainly bring positives to the performance of individual librarians, it is very important for academic library managers to communicate performance merits and deficits to librarians through feedback. Feedback in the workplace context is the “dynamic communication process occurring between two individuals that convey information regarding the receiver's performance in the accomplishment of work-related tasks” (Baker et al., 2013, p. 260). This study focuses on two simple yet highly relevant details about supervisor feedback. First, does the academic librarian receive any feedback from their supervisor? Second, with what frequency is that feedback delivered? Feedback, like work autonomy, is a variable that an academic library manager has complete control over. Feedback is an important aspect of the supervisor-employee relationship.

Performance feedback is an essential tool for increasing the effectiveness of individuals and organizations (McCarthy & Garavan, 2006). The use of “high performance work practices” such as committing to well-planned performance feedback systems is associated with increases in productivity and decreases in employee turnover (Huselid, 1995). Feedback that is constructive (i.e., builds on strengths, does not focus on personal weakness, etc.) has been shown to predict increases in employee's intention to display organizational citizenship behaviors and increases in employee job satisfaction through creating feelings that the supervisor/organization respects the employee and is actively helping to advance their career (Sommer & Kulkarni, 2012).

Older studies on feedback frequency point towards increased frequency as a contributor to gains in learning and better performance on tasks (Salmoni et al., 1984). More recent research suggests that supervisors need to find an “optimum level” of feedback frequency. While increases in feedback can improve performance, excessive feedback can actually lead to declines in performance for employees who are new to a task or employees who have more experience with a task (Lam et al., 2011). Many organizations have acknowledged research findings and expanded their feedback systems past using simply an annual evaluation for delivering employee feedback, while others have taken the extra step of completely eliminating annual evaluations altogether (Goler et al., 2016).

Feedback that is delivered more frequently than an annual evaluation appears to be beneficial in academic libraries as well. An extensive review of academic library performance appraisal systems by Stilling et al. (2018) found that library leaders with feedback systems that contain an annual evaluation and additional feedback opportunities, rated the effectiveness of their feedback system's ability to create changes in the behavior of individual librarians, aid individual librarians in achieving job goals, and aid the library in achieving its overall organizational goals, at a statistically significant higher rate than library leaders with feedback systems that contain only an annual evaluation.

Although not directly influenced like changes to work autonomy or feedback, employee turnover remains a construct of high importance to academic library managers. Employee turnover and related costs for new employee recruitment are an ongoing reality for all organizations. As academic libraries find themselves in difficult budget situations, the funds to replace a librarian who has left the organization may not be available. For academic library managers who cannot replace librarians who leave their organization, preventing turnover is critical for slowing the rate of workforce attrition.

Even when a librarian can be replaced, going through the many steps of the hiring process followed by training and onboarding of a new employee is a costly endeavor (Knight, 2013; Raschke, 2003). In situations where turnover happens within one and a half years of hiring, these costs have been estimated to be a total loss for the organization (Knight, 2013). Turnover in academic libraries bring other non-monetary consequences such as talent loss, decreases in productivity, drain on remaining staff who are taking on extra duties, and potentially hurting patron relations when services lag due to vacancies (Raschke, 2003).

The role of supervisors in the retention or turnover of employees is well-known. Studies have linked the retention or decreased turnover intention of employees to many important factors to academic librarians such as supervisors following ethical practices (DeConinck, 2015), supervisors being good listeners (Lloyd et al., 2015), and employee's increased feelings of organizational support through their supervisor valuing work contributions and caring about their personal well-being (Eisenberger et al., 2002).

Understanding the role workplace morale has on influencing turnover intention will assist academic library managers in modifying their own behaviors, developing stronger retention practices, and working to avoid this costly disruption.

Methodology

Participants were recruited for this online study in late 2018 by email invitation distributed via seven academic library-related listservs (ACRL-frm, ALAO, ALCTScentral, COLLIB, RUSA-l, SCHOLCOMM, and ULS). Complete data was collected from a total of 559 participants. Before completing the survey, participants were made aware of their rights and provided pertinent IRB information. All 559 participants confirmed that they were currently employed professional academic librarians who do not supervise other professional librarians.

Participants

To assess the representativeness of this survey's population to the overall librarian profession, the gender, race, and age of participants has been compared to the most recent American Library Association Office of Research and Statistics (2017) demographic report on membership. This comparison can be found in Table 1 . While the participants in this study roughly match the demographics of ALA membership, this sample population is slightly less diverse and a bit younger overall.

Table 1

Participant demographics compared to ALA (2017).

Variable	Present study	ALA (2017)
Gender
Female	79.97%	81%
Male	17.53%	19%
Prefer not to answer	2.5%	n/a
Race
African American or Black	2.86%	4.4%
American Indian or Alaskan Native	0.72%	1.2%
Asian	2.15%	3.6%
White	88.19%	86.7%
Other	6.08%	4.2%
Age
Range	23 to 72 years old	n/a
Mean age	42.8 years old	n/a
44 years old or younger	59.75%	41%

Participants had a wide range of professional librarian work experience which spanned from 1 to 42 years of experience. The mean professional librarian work experience of the sample is 11.28 years. The library department participants reported working in varied with 31.49% from Public Services, 23.26% from Instruction, 20.57% from Technical Services, 8.23% from Scholarly Communications, 3.39% from Information Technology, 0.89% from Data-Centers, and 12.17% from other or multiple departments. Participants work in an academic library at either a public (63.86%) or private (36.14%) institution of higher education. The overwhelming majority of participants work at 4 year institutions that grant doctorate degrees (65.65%), followed by 4 year institutions that grant masters degrees (18.78%), 4 year undergraduate only institutions (10.02%), and 2 year institutions (5.55%). Faculty status at those institutions was also collected with 59.75% of participants reporting they have faculty status and 40.25% reporting they did not have faculty status. Regardless of faculty status, 78.18% of participants reported they can be promoted in rank at their institution.

Measures

Participants were instructed to answer all survey questions about their current position and current direct supervisor unless otherwise directed in a specific question.

Workplace morale measure

Workplace morale has been constructed as a latent variable operationalized as an equal weights combination of affective commitment to the organization and job satisfaction.

Affective commitment to the organization was measured using the Affective Commitment Scale (ACS) (Allen & Meyer, 1990). The ACS measures an employee's feelings towards their organization. The scale features eight items (e.g., I really feel as if this organization's problems are my own.) which participants respond to using a 7-point Likert scale that ranges from 1 (strongly disagree) to 7 (strongly agree). An overall score is calculated by taking the mean of all items. Higher scores indicate a higher level of affective commitment to the organization.

Job satisfaction was measured using Michigan Organizational Assessment Questionnaire (MOAQ) (Cammann et al., 1979). The MOAQ measures an employee's self-reported level of job satisfaction in their current position. The scale features three items (e.g., All in all I am satisfied with my job.) which participants respond to using a 7-point Likert scale that ranges from 1 (strongly disagree) to 7 (strongly agree). An overall score is calculated by taking the sum of all items. Higher scores indicate a higher level of job satisfaction.

Operationalizing workplace morale as a combination of affective commitment to the organization and job satisfaction required constructing a combined score for the ACS and MOAQ. Both the ACS and MOAQ score items using the same 7-point Likert scale. The average score for the 8 items on the ACS and the average score for the 3 items on the MOAQ were calculated. These two comparable average scores were then averaged to create a new composite score for workplace morale. This new score represents the average response to all 11 workplace morale items where half of the weight is given to the 8 items on the ACS and half of the weight is given to the 3 items on the MOAQ.

Other measures

Demographic questions and other non-scale collected items (e.g., data about the availability of cost-of-living raises) can be found in Appendix A.

Work autonomy was measured with the Work Autonomy Scales (WAS) (Breaugh, 1999). The WAS utilizes three separate scales to measure different aspects of work autonomy including method autonomy (employee's ability to determine the procedures they use to complete their work), scheduling autonomy (employee's ability to determine when they do their work), and criteria autonomy (employee's ability to create or modify job performance criteria) (Breaugh, 1999). Each scale features three statements about the type of work autonomy being measured (e.g., My job is such that I can decide when to do particular work activities.) which participants respond to using a 7-point Likert scale that ranges from 1 (strongly disagree) to 7 (strongly agree). Three individual scores are calculated by taking the mean of responses to each scale. There is no overall score. Higher scores indicate a higher level of each type of work autonomy.

Turnover intention was measured with the six item version of the Turnover Intention Scale (TIS-6) (Bothma & Roodt, 2013; Roodt, 2004). The TIS-6 measures an individual's intention to leave their current employer or organization. The measure features six items related to the participant's intention to leave their current employer or organization (e.g., How often do you dream about getting another job that will better suit your personal needs?). Participants respond to each item using various 5-point Likert scales ranging from 1 (never/highly unlikely/to no extent) to 5 (always/highly likely/to a very large extent). An overall turnover intention score is calculated by taking the mean of all responses. Higher scores indicate higher turnover intention.

Additional supervisor qualities were measured via nine questions created by the authors. These nine items stand-alone but are collected together in one place for convenience. Each item is a statement related to the characteristics and quality of a supervisor (e.g., My supervisor is good at communicating with the librarians they supervise.) which participants respond to using a 7-point Likert scale that ranges from 1 (strongly disagree) to 7 (strongly agree). Each item is scored individually. Higher scores indicate higher confidence in the participant's belief that their supervisor possesses the quality. These nine items can be found in Appendix B.

Results

As the authors are unaware of any previous studies that have quantitatively measured the workplace morale of academic librarians, it is important to establish a baseline of findings observed in this study to aid the overall understanding of workplace morale in academic librarians and to inform future research on the topic. The average workplace morale score for this sample of academic librarians was 4.41 on a 1 to 7 scale that has a midpoint of 4.0. This score indicates that the workplace morale of academic librarians in this sample was fairly average with a slight lean towards higher workplace morale scores.

Predictors of workplace morale

Many demographic and position-specific data points were collected in an attempt to better understand the workplace morale of academic librarians in this exploratory study. While no specific hypotheses were formed, it was generally assumed that the various data points collected in the study would have an impact on workplace morale. An emphasis was placed on collecting information that would generally inform academic library managers about their workforce and which areas a manager could reasonably impact that could play a role in predicting workplace morale (e.g., feedback frequency).

Characteristics of the individual

Gender

An independent samples t-test was conducted to compare the workplace morale of participants who identified as male or female. There was not a statistically significant difference in workplace morale between males (M = 4.51, SD = 1.45) and females (M = 4.40, SD = 1.45); t(543) = 0.683, p = .495.

Race

A one-way between groups ANOVA was conducted to compare the effect of race on the workplace morale for participants identifying as White, African American or Black, Asian, American Indian or Alaskan Native, and Other. There was not a significant effect of race on workplace morale between these five race groups; F(4, 554) = 0.603, p = .660. It should be noted that several of the race groups may be too small (e.g., African American or Black (n = 17)) to draw a valid statistical conclusion from this data.

Age

A Pearson product-moment correlation was calculated to assess the relationship between workplace morale and age. There was a statistically significant weak negative correlation between workplace morale and age; r = −0.127, N = 559, p = .003. The negative relationship indicates that age and workplace morale move in separate directions. Workplace morale will decrease when age increases and workplace morale will increase when age decreases.

Work experience

A Pearson product-moment correlation was calculated to assess the relationship between workplace morale and work experience as a professional academic librarian. There was a statistically significant weak negative correlation between workplace morale and work experience; r = −0.128, N = 559, p = .002. The negative relationship indicates that the variables move in separate directions. Workplace morale will decrease when work experience increases and workplace morale will increase when work experience decreases.

Characteristics of the institution

Institution type

An independent sample t-test was conducted to compare workplace morale based on the type of institution academic librarians are employed in. There was not a statistically significant difference in workplace morale between librarians working at a public institution (M = 4.35, SD = 1.47) and those working at a private institution (M = 4.52, SD = 1.43); t(557) = −1.307, p = .192.

Degree granting status

A one-way between groups ANOVA was conducted to compare the effect of the institution's degree granting status on workplace morale for participants from institutions that are a 2-year school, 4-year school, 4-year school that offers graduate degrees, and 4-year school that offers doctorate degrees. There was not a significant effect of degree granting status on workplace morale between the four degree granting status groups; F(3, 555) = 1.887, p = .131.

Faculty rank

An independent samples t-test was conducted to compare workplace morale based on whether or not academic librarians held faculty rank at their institution. There was not a statistically significant difference in workplace morale between librarians who are faculty at their institution (M = 4.47, SD = 1.41) and those who are not faculty (M = 4.33, SD = 1.51); t(557) = 1.147, p = .252.

Promotion or rank increase availability

An independent samples t-test was conducted to compare workplace morale based on whether or not academic librarians could be promoted in rank at their institution. There was not a statistically significant difference in workplace morale between academic librarians who could be promoted in rank at their institution (M = 4.41, SD = 1.41) and those who could not be promoted in rank (M = 4.42, SD = 1.61); t(557) = −0.116, p = .908.

Characteristics of the library

Library department

A one-way between groups ANOVA was conducted to compare the effect of library department on workplace morale for participants working in public services, technical services, information technology, instruction, scholarly communications, data centers, and other departments. There was not a significant effect of library department on workplace morale between the seven department groups; F(6, 552) = 0.693, p = .656.

Professional development funds availability

An independent samples t-test was conducted to compare workplace morale based on whether or not professional development funds were available to participants at their library. There was a statistically significant difference in workplace morale between librarians who had professional development funds available to them (M = 4.47, SD = 1.42) and those that did not have professional development funds available to them (M = 3.82, SD = 1.61); t(557) = 3.149, p = .002.

Librarian staffing levels

A one-way between groups ANOVA was conducted to compare the effect of librarian staffing levels on workplace morale for participants who indicated librarian staffing at their institution over the past 3 years has increased, decreased, or stayed the same. There was a statistically significant effect of librarian staffing levels on workplace morale between the three staffing conditions; F(2, 556) = 13.541, p < .001. Post hoc comparisons using Fisher's LSD test indicated the mean score for the increased number of librarians condition (M = 4.80, SD = 1.33) was significantly different than the decreased number of librarians condition (M = 4.04, SD = 1.46); p < .001. The mean score for the same number of librarians condition (M = 4.55, SD = 1.44) was significantly different than the decreased number of librarians condition (M = 4.04, SD = 1.46); p < .001. There was not a significant difference between the increased librarian condition and same number of librarians condition; p = .102.

Duties spread

The 217 participants who indicated librarian staffing levels at their institution have decreased over the past 3 years were given the following additional question: “Have the duties of the librarians who have not been replaced been assigned to the remaining librarians at your institution?”

An independent samples t-test was conducted to compare workplace morale based on whether or not participants indicated that the job duties of non-replaced librarians have been reassigned to the remaining librarians at their institution. There was not a statistically significant difference in workplace morale between librarians whose institution reassigns the work of librarians rather than replacing librarians (M = 4.02, SD = 1.47) and those whose institution is not reassigning the work of non-replaced librarians to the remaining librarians (M = 4.21, SD = 1.38); t(215) = −0.559, p = .577. It should be noted that the subgroup sample size for participants indicating job duties are not reassigned (n = 21) may be too small to draw a valid statistical conclusion from this data.

Duties spread impact

The 196 participants who indicated librarian staffing levels at their institution have decreased over the past 3 years and the job duties of librarians who have not been replaced have been reassigned to the remaining librarians at the institution, were given the following additional question: “Do you believe your department has been more heavily impacted by the absorbing of duties than other departments in the library?”

An independent samples t-test was conducted to compare workplace morale based on whether or not participants believed their department was more heavily impacted than other departments by absorbing job duties from non-replaced librarian positions. There was a statistically significant difference in workplace morale between librarians who indicated their department was more heavily impacted (M = 3.81, SD = 1.44) and those who did not feel their department was more heavily impacted (M = 4.30, SD = 1.46); t(194) = −2.352, p = .02.

Compensation related variables

Cost-of-living or base salary/wage increase

An independent samples t-test was conducted to compare workplace morale based on whether or not librarians had received a cost-of-living or base increase to their salary/wages for the most recent year. There was a statistically significant difference in workplace morale between librarians who received a cost of living or base increase to their salary/wages in the most recent year (M = 4.71, SD = 1.35) and those that did not receive a cost-of-living or base increase to their salary/wages in the most recent year (M = 4.12, SD = 1.49); t(557) = 4.939, p < .001.

Merit-based raise availability

An independent samples t-test was conducted to compare workplace morale based on whether or not academic librarians were eligible to receive a merit-based increase to their salary/wages in the most recent year. There was not a statistically significant difference in workplace morale between librarians who were eligible to receive a merit based raise in the most recent year (M = 4.42, SD = 1.38) and those that were not eligible (M = 4.40, SD = 1.52); t(557) = 0.214, p = .831.

Budget related variables

Institution-wide budget

A one-way between groups ANOVA was conducted to compare the effect of institution- wide budget on workplace morale for participants who indicated their institution-wide budget over the past 3 years was increasing, decreasing, stagnant, or unknown to them. There was a significant effect of institution-wide budget on workplace morale between the four institution-wide budget conditions; F(3, 555) = 3.389, p = .018. Post hoc comparisons using Fisher's LSD test indicated the mean score for the increasing budget condition (M = 4.61, SD = 1.4) was significantly different than the decreasing budget condition (M = 4.16, SD = 1.53); p = .01. The mean score for the decreasing budget condition (M = 4.16, SD = 1.53) was significantly different than the budget unknown condition (M = 4.57, SD = 1.41); p = .008. All other interactions between institution-wide budget conditions, including the stagnant condition (M = 4.29, SD = 1.40), were not significant.

Library-wide budget

A one-way between groups ANOVA was conducted to compare the effect of library- wide budget on workplace morale for participants who indicated their library-wide budget over the past 3 years was increasing, decreasing, stagnant, or unknown to them. There was a significant effect of library-wide budget on workplace morale between the four library-wide budget conditions; F(3, 555) = 3.03, p = .029. Table 2 contains the results of post hoc comparisons using Fisher's LSD test to indicate significant mean differences in workplace morale between library-wide budget conditions.

Table 2

Significance of workplace morale mean differences based on library-wide budget conditions.

	IncreasingM = 4.66SD = 1.48	DecreasingM = 4.26SD = 1.52	StagnantM = 4.32SD = 1.39	UnknownM = 4.70SD = 1.35
IncreasingM = 4.66SD = 1.48	X	0.035⁎	0.078	0.858
DecreasingM = 4.26SD = 1.52	0.035⁎	X	0.647	0.015⁎
StagnantM = 4.32SD = 1.39	0.078	0.647	X	0.038⁎
UnknownM = 4.70SD = 1.35	0.858	0.015⁎	0.038⁎	X

The mean difference between conditions is significant at p = .05.

Collection development budget

A one-way between groups ANOVA was conducted to compare the effect of collection development budget on workplace morale for participants who indicated their collection development budget over the past 3 years was increasing, decreasing, stagnant, or unknown to them. There was not a significant effect of collection development budget on workplace morale between the four collections budget conditions; F(3, 555) = 1.572, p = .195.

Feedback

Feedback received

An independent samples t-test was conducted to compare workplace morale based on whether or not academic librarians receive any feedback on their work performance from their immediate supervisor. There was a statistically significant difference in workplace morale between librarians who indicated they receive feedback from their immediate supervisor (M = 4.51, SD = 1.42) and those that indicated they do not receive feedback (M = 3.80, SD = 1.49); t(557) = 4.001, p < .001.

Feedback frequency

The 483 participants who indicated they receive feedback from their immediate supervisor were given the following additional question: “What is the frequency of feedback you receive from your immediate supervisor?”

A one-way between groups ANOVA was conducted to compare the effect of feedback frequency (i.e., once per year/during an annual evaluation, twice a year, monthly, weekly, and daily) on workplace morale. There was a significant effect of feedback frequency on workplace morale between the five feedback frequency conditions; F(4, 478) = 3.948, p = .004. Table 3 contains the results of post hoc comparisons using Fisher's LSD test to indicate significant mean differences in workplace morale between feedback frequency conditions. It should be noted that the subgroup sample size for participants indicating they receive daily feedback (n = 9) may be too small to draw any valid statistical conclusions about that group having no significance.

Table 3

Significance of workplace morale mean differences based on feedback frequency conditions.

	Once a yearM = 4.34SD = 1.42	Twice a yearM = 4.19SD = 1.44	MonthlyM = 4.78SD = 1.36	WeeklyM = 4.89SD = 1.37	DailyM = 4.79SD = 1.66
Once a yearM = 4.34SD = 1.42	X	0.472	0.004⁎	0.01⁎	0.345
Twice a yearM = 4.19SD = 1.44	0.472	X	0.006⁎	0.008⁎	0.236
MonthlyM = 4.78SD = 1.36	0.004⁎	0.006⁎	X	0.644	0.988
WeeklyM = 4.89SD = 1.37	0.01⁎	0.008⁎	0.644	X	0.847
DailyM = 4.79SD = 1.66	0.345	0.236	0.988	0.847	X

The mean difference between conditions is significant at p = .05.

Work autonomy

The following are the results for the three types of work autonomy measured via the WAS.

Method autonomy

A Pearson product-moment correlation was calculated to assess the relationship between workplace morale and method autonomy. There was a strong statistically significant positive correlation between workplace morale and method autonomy; r = 0.566, N = 559, p < .001. This positive relationship indicates that the variables move in the same direction. Workplace morale will increase when method autonomy increases and workplace morale will decrease when method autonomy decreases.

Scheduling autonomy

A Pearson product-moment correlation was calculated to assess the relationship between workplace morale and scheduling autonomy. There was a moderate statistically significant positive correlation between workplace morale and scheduling autonomy; r = 0.427, N = 559, p < .001. The positive relationship indicates that the variables move in the same direction. Workplace morale will increase when scheduling autonomy increases and workplace morale will decrease when scheduling autonomy decreases.

Criteria autonomy

A Pearson product-moment correlation was calculated to assess the relationship between workplace morale and criteria autonomy. There was a moderate statistically significant positive correlation between workplace morale and criteria autonomy; r = 0.483, N = 559, p < 0.001. The positive relationship indicates that the variables move in the same direction. Workplace morale will increase when criteria autonomy increases and workplace morale will decrease when criteria autonomy decreases.

Supervisor related variables

Supervisor qualities

A Pearson product-moment correlation was calculated for each of the nine supervisor quality questions to assess the relationship between workplace morale and participant's indication of the degree to which they believe their supervisor possesses each quality or characteristic. Each supervisor quality question returned a statistically significant positive correlation which indicates workplace morale will increase as participants agree that their supervisor possesses the quality or characteristic and workplace morale will decrease as participants disagree that their supervisor possesses the quality or characteristic. These results are reflected in Table 4 .

Table 4

Supervisor qualities as predictors of workplace morale (N = 559).

Quality question	r	p
My supervisor recognizes and acknowledges when I do an exceptional job.	0.568	< 0.001
My supervisor is good at communicating with the librarians they supervise.	0.577	< 0.001
My supervisor is an advocate for the department and the librarians they supervise.	0.590	< 0.001
My supervisor is transparent (i.e., provides relevant updates, does not unnecessarily withhold information, limits the amount of time employees are surprised by new work or directives, etc.).	0.584	< 0.001
My supervisor has the appropriate level of competence to effectively do their job.	0.570	< 0.001
My supervisor treats everyone equally.	0.532	< 0.001
My supervisor empowers me to bring about positive change at my library.	0.672	< 0.001
My supervisor fosters a culture of collegiality and positive direction for our unit.	0.620	< 0.001
My supervisor is willing to mentor other librarians in the department.	0.462	< 0.001

Long-lasting supervisor incident

An independent samples t-test was conducted to compare workplace morale based on whether or not librarians indicated that a single incident with a supervisor in the past still impacts their overall workplace morale to the current day. There was a statistically significant difference in workplace morale between librarians who indicated there was a past supervisor incident that still impacts them (M = 4.20, SD = 1.45) and those who were not impacted by a past incident (M = 4.80, SD = 1.38); t(557) = −4.674, p < .001.

A total of 65.3% (n = 365) of participants indicated that a past supervisor incident is still impacting their morale to the current day. A follow-up question was asked to these participants to clarify if the incident was positive (n = 59) or negative (n = 306). An independent samples t-test was conducted to compare workplace morale based on whether librarians had a positive or negative long-lasting supervisor incident. There was a statistically significant difference in workplace morale between librarians who indicated the long-lasting supervisor incident was positive (M = 5.14, SD = 1.35) and those who indicated it was negative (M = 4.02, SD = 1.40); t(363) = 5.618, p < .001.

Turnover intention

A Pearson product-moment correlation was calculated to assess the relationship between workplace morale and turnover intention. There was a strong statistically significant negative correlation between workplace morale and turnover intention; r = −0.837, N = 559, p < .001. The negative relationship indicates that the variables move in opposite directions. Turnover intention will increase when workplace morale decreases and turnover intention will decrease when workplace morale increases.

Discussion and recommendations

This study aimed to quantitatively measure workplace morale in academic librarians to expand the understanding of this issue which has primarily been measured in other studies via adjacent constructs or qualitative methods. The present findings suggest the overall workplace morale of academic librarians is relatively average with a full sample mean of 4.41 on a 1–7 scale with a midpoint of 4. Certainly, the workplace morale of individual academic librarians can and will differ from this study's full sample mean but the findings suggest that there is not an overall systemic issue of low workplace morale in academic librarians. Understanding the factors that predict those individual differences in workplace morale is important to library managers looking to take steps to raise overall workplace morale and get the most out of their workforce.

This study found no relationship between race or gender and workplace morale. These non-significant findings are worth noting considering the library profession is predominantly white and female. As previously noted, the sample sizes for the minority race groups may be too small to make any definitive conclusions about the impact of race on workplace morale. Kendrick and Damasco (2019) should be consulted for more current information on the workplace morale experience of minority academic librarians. The age and work experience of an individual were found to have a weak significant relationship to workplace morale. Age (r = −0.127, p = .003) and work experience (r = −0.128, p = .002) behaved almost identically in their relationship to workplace morale. The findings establish a weak relationship where an increase in age or experience is associated with a decrease in workplace morale and vice-versa. Academic library managers should be aware that older and experienced academic librarians may have some reduction in workplace morale compared to their younger and less-experienced colleagues.

In addition to characteristics unique to the individual, this study analyzed institutional characteristics to determine if they have any impact on workplace morale. This study found no relationship between institution type (i.e., public or private) and workplace morale. While differences certainly exist based on institution type (e.g., type of student, religious affiliations, enrollment, etc.), there was not a significant difference in workplace morale for academic librarians based on institution. Along the same lines, there was no relationship between an institution's degree granting status and workplace morale.

There was also no relationship between workplace morale and having faculty status or the ability to be promoted in rank. It seems intuitive that having faculty status and/or the ability to be promoted in rank would contribute to workplace morale but these results do not support those assumptions.

Library-specific predictors of workplace morale were also analyzed. There was no relationship between library department and workplace morale. While job satisfaction only makes up half of this study's workplace morale measure, these findings somewhat differ from previous studies that have found a difference in job satisfaction of library workers based on library department (Lynch & Verdin, 1987). Academic librarians who have access to professional development funds had a statistically significant higher morale than those who do not have access to funding (p = .002). Academic library managers interested in the workplace morale of their librarians should look to preserve professional development funding as much as possible when working with budget cuts.

Librarian staffing levels, compared to three years ago, were found to be a predictor of workplace morale (p < .001). Participants who reported more librarians on staff (p <.001) and participants who reported no change in librarian staffing levels (p < .001) had a higher morale than those that reported less librarians on staff. Interestingly, there was no difference in workplace morale between the more librarians group and same number of librarians group. These findings suggest that academic librarian workplace morale is increased not by gaining additional colleagues but rather by not losing existing colleagues. Academic library managers facing workforce attrition issues should be cognizant of how the loss of colleagues may impact the workplace morale of their remaining librarians. Similarly, the handling of job duties from non-replaced librarians also plays a role in impacting workplace morale. The workplace morale of academic librarians who reported their department was more heavily impacted than other departments by absorbing the job duties of non-replaced librarians had a lower workplace morale than those indicating their department had not been particularly impacted (p = .02). While academic library managers may not be able to prevent a specific department from facing higher attrition than other departments, they should be aware of how the absorbing of job duties impacts the remaining librarians in those shrinking departments. To reduce this impact, managers can take steps such as providing more time to meet project deadlines and understanding extra-role behaviors may be much less possible by librarians who have absorbed additional job duties. Engaging in activities such as creating detailed procedure manuals and other succession planning activities may also help ease the burden on librarians who must absorb new duties.

Compensation related variables were analyzed for their impact on workplace morale. Academic librarians who reported receiving a base increase to their pay in the most recent year, sometimes called a cost-of-living adjustment, had higher workplace morale than those who did not receive an increase (p < .001). While an academic library manager probably cannot control the availability of these raises, it is worth noting that the absence of these yearly increases can be a contributing factor to lower workplace morale. There was no relationship between workplace morale and an individual's ability to earn a merit-based raise.

It is important to understand how various budgets impact workplace morale for academic librarians as many higher education institutions face budget challenges. This study analyzed how changes (i.e., increasing, decreasing, stagnant, or budget situation unknown to participant) over the past three years in the institution-wide budget, library- wide budget, and collection development budget, predict workplace morale. For the institution-wide budget, the only predictors of increased morale were for the increasing condition over the decreasing condition (p = .01) and the budget unknown condition over the decreasing condition (p = .008). These findings suggest a decreasing institution-wide budget weighs on the workplace morale of academic librarians and it is better for workplace morale to be unaware of the budget situation rather than to know if the budget is decreasing. It is worth noting that there was not a difference in workplace morale between the stagnant and decreasing institution-wide budget conditions.

For the library-wide budget, the only predictors of increased morale were for the increasing budget condition over the decreasing budget condition (p = .035), the budget unknown condition over the decreasing condition (p = .015), and the budget unknown condition over the stagnant condition (p = .038). It is no surprise that increasing library- wide budgets have a positive impact on workplace morale in comparison to decreasing library-wide budgets. These findings also suggest that not knowing the library-wide budget situation leads to an increase in workplace morale in comparison to knowing the budget is either decreasing or stagnant. Interestingly, the results show no relationship between workplace morale and any changes in collection development budgets. Overall, these findings suggest that academic library managers should be aware that academic librarians who work with budgets may be more susceptible to impacts on their workplace morale, especially as budgets decrease.

Feedback is one of the areas analyzed in this study that academic library managers can directly influence by simply changing their own behaviors. Academic librarians reporting they receive performance feedback had higher workplace morale than those indicating they receive no performance feedback (p < .001). These results were expected as there is vast research on the benefits of feedback. This study also found that the frequency with which feedback is delivered is important to academic librarians. In all interactions, findings show that academic librarians have higher workplace morale in the monthly or weekly feedback condition when compared to the once per year or twice per year feedback conditions. These results strengthen both research and business practices which find that feedback is ineffective when delivered simply as part of annual performance appraisal procedures. Academic library managers should be mindful to provide feedback to their librarians on a more consistent schedule even if their organization is utilizing an annual performance appraisal.

Work autonomy is an additional area analyzed that academic library managers can directly influence by simply changing their own behaviors. All three types of work autonomy measured in this study were found to have a positive relationship to workplace morale. Academic library managers may be able to increase the workplace morale of their librarians by increasing their level of method autonomy (ability to determine the procedures they use to complete their work), scheduling autonomy (ability to determine when they do their work), and/or criteria autonomy (ability to create or modify job performance criteria) (Breaugh, 1999). While certain aspects of an academic librarian's job may be inflexible (e.g., tasks that must be completed in a certain way, working rotating night/weekend shifts, etc.), academic library managers can take steps to increase other areas of work autonomy to help librarians achieve better control over their work and increased levels of work-life balance. These autonomy related changes bring their own benefits in addition to increases in workplace morale.

Several supervisor qualities were examined for their impact on workplace morale. There was a positive relationship between workplace morale and an academic librarian's belief that their supervisor possessed each quality (see Table 4). Each of these supervisor qualities are easily influenced by academic library managers through their own efforts and thoughtfulness. Academic library managers may be able to increase the workplace morale of their librarians through adopting several of these common sense qualities such as recognizing exceptional work, increasing transparency, mentoring librarians, and empowering librarians. These findings follow with Martin (2018) who found academic librarians reflecting on past positive leaders indicated those leaders were emotionally intelligent, empowering, visionary, a strong communicator, competent, skilled, trustworthy, and a catalyst for change.

Findings also show that the actions of a supervisor can have a long-term impact on the workplace morale of academic librarians. Almost two-thirds (63.5%) of participants indicated that there was a single incident in the past with a supervisor that still impacts their workplace morale to this day. The large majority (83.8%) of these incidents were reported as negative. Findings show that academic librarians who reported being impacted by a previous supervisor incident had lower workplace morale than those academic librarians who did not report a past supervisor incident (p < .001). This prolonged low workplace morale experience was also documented by Kendrick (2017) who found long-lasting impact on the workplace morale of academic librarians who were micromanaged or were a subordinate to an authoritarian supervisor. Academic library managers should be aware that past negative interactions with supervisors can continue to impact their librarians several years later. It is also important to keep in mind that an incident with a supervisor who is no longer with the organization could be causing workplace morale issues of academic librarians who remain with the organization. Being cognizant of these issues may be especially noteworthy to academic library managers starting a new position and/or joining a new organization.

Understanding how workplace morale predicts turnover intention in academic librarians is very important in an era of position attrition and decreasing budgets. This study found a very strong negative relationship (r = −0.837, p < .001) between workplace morale and turnover intention. Although the observed relationship is a correlation and direct causation cannot be stated, this strong interaction is very notable. These findings show that workplace morale and turnover intention move very drastically in opposite directions. Academic library managers may be able to reduce an individual librarian's turnover intention through taking one of the many steps identified in this study to increase that individual's workplace morale.

Table 5 provides a summary of factors academic library managers can influence to promote increased workplace morale in their librarians.

Table 5

Factors academic library managers can influence to promote workplace morale in academic librarians.

Making available funding for professional development.

Providing performance feedback.

Providing performance feedback with an increased frequency such as on a weekly or monthly basis.

Increasing all types of work autonomy (e.g., how to complete work tasks).

Adjusting work expectations for librarians absorbing extra job duties due to attrition in departments.

Adopting supervisor qualities such as recognizing exceptional work, transparency, mentoring, and empowering librarians.

It should be noted that the data for this study was collect prior to the COVID-19 pandemic. The pandemic has caused dramatic changes to academic library services, working conditions for librarians, and library budgets. It is not unreasonable to assume the workplace morale of academic librarians has also been impacted with sudden changes to typical day-to-day life and a need to manage emerging life-work integration. Academic library managers need to be more vigilant than ever in providing performance feedback especially as work culture and job expectations evolve. Managers should consider allowing for increased levels of work autonomy as individuals manage new schedules, childcare, and their physical and mental health. Supervisor qualities such as communication, transparency, and empowering direct supports should continue to be key attributes for academic library managers concerned with the well-being of their direct reports. Future studies should analyze the workplace morale of academic librarians as higher education efforts to get back to normal or establish a new normal.

Conclusion

The underlying goal of this research is to help organizations, academic library managers, and academic librarians improve workplace morale as the construct is important to the success of all three. To that end, this exploratory research quantitatively measured workplace morale in academic librarians and conducted a thorough examination of workplace morale predictors with an emphasis on the impact academic library managers have on the librarians that directly report to them.

The workplace morale experience of the academic librarians in this sample was fairly average. Findings show that demographics, with the exception of age and experience which share a small negative relationship to workplace morale, and characteristics of the institution play no role in predicting workplace morale in academic librarians. This study found many predictors of workplace morale within an academic library manager's direct influence including feedback, feedback frequency, work autonomy, workload expectation of departments with heavy librarian attrition, and several supervisor qualities with themes such as communication, transparency, mentoring, and empowerment. Workplace morale and the turnover intention of academic librarians were found to move strongly in opposite directions.

The results and issues raised by this study are one step towards better understanding workplace morale in academic librarians. Future investigations and research can enhance and expand on these results in several ways. While this study used the same demographic categories as ALA, several participants indicated the options for gender and race could have been more inclusive towards them. Collecting better detailed demographic information would also help future research take a more granular look at personal characteristics and how they interact with workplace morale. In particular, a better look at workplace morale in terms of gender identity and a more robust sample with stronger minority representation, so that statistical conclusions can be made about the role of race on academic librarian workplace morale, are warranted. This study did not ask about union status which is a variable worth investigating. Future research can also look to gather data on workplace morale from academic library middle managers who have subordinates and supervisors.

The focus of this research was on the relationship between academic librarians and the academic library managers who directly supervise them. However, some participants gave feedback about issues impacting their workplace morale that dealt with higher- level managers in the library or even administrators at the overall institution level. A common theme to these comments was poor communication. These additional contributing factors should be explored. Studies that follow-up on the turnover intention of academic librarians should consider workplace morale and additional factors that play a role in turnover decisions such as family dynamics, finances, and retirement plans.

Declaration of competing interest

None.