A survey of employee engagement
iii TABLE OF CONTENTS ACKNOWLEDGEMENTS…………………………………………………………...ii LIST OF TABLES……………………………………………………………………..v ABSTRACT…………………………………………………………………………..vi Chapter 1. INTRODUCTION…………………………………………………………… 1 Conceptual Framework Need for the Study Statement of the Problem Significance of the Study Statement of the Purpose Research Questions Hypotheses Definition of Terms Assumption, Limitations, and Delimitations Conclusion 2. LITERATURE REVIEW……………………………………………………12 Evidence on Burnout Previous Studies of Employee Engagement Predictors of Burnout and Engagement Summary 3. METHOD OF STUDY……………………………………………………..27 Participants
iv Instrument Procedure 4. RESULTS……………………………………………………………………33 Characteristics of Sample Results Hypothesis One Hypothesis Two Hypothesis Three Hypothesis Four 5. DISCUSSION………………………………………………………………..40 Summary of Key Findings Discussion and Conclusions Instrument Implications for Social Work Policy Implications for Social Work Practice Implications for Social Work Research Conclusion REFRENCES…………………………………………………………………….…..51 APPENDIX…………………………………………………………………………..56 VITA…………………………………………………………………………………59
v LIST OF TABLES Table Page 1.
Reliability Measure of Pilot Study Data……………………………...31 2.
Descriptive Statistics of Engagement Scores………………………...36 3.
Participant Total Engagement Scores………………………………...36 4.
Correlation – Years of Servic e and Engagement Score………………38
vi A SURVEY OF EMPLOYEE ENGAGEMENT KAREN WILSON Dr. Larry Kreuger, Dissertation Advisor ABSTRACT This exploratory study examined the overa ll level of engagement for employees of a public rehabilitation service agency, and the extent to which demographic and work life variables such as gender, office location, job title, and years of service contributed to their levels of engagement . Engagement was examined as a positive social work construct and was compared to the negative concept of burnout (Freeney, Y. & Tiernan, J. 2006 & Schaufli, W. B. & Bakker, A. B. 2004). The researcher utilized a cross sectional in ternet survey to survey 308 employees of a state-wide rehabilitation agency located in the Mi dwest region of the United States, using a purposive non-probability sampling strategy. She queried respondent s to collect data on their levels of engagement in aggr egate and anonymous form. Analysis was conducted using t-tests and correlations. Resu lts indicate no differences in engagement scores for males and females, for individua ls working in rural versus urban office environments, or for years of service in th e agency. Participants who supervised at least one other person scored higher than individuals who did not supervise anyone.
1 Chapter One: Introduction Staff members in human service occupatio ns work closely with the individuals they serve. They are a vital link in the chai n of services provided to our nation’s weak and vulnerable populations. However, working in the field is demanding, and employee emotional exhaustion is common (Maslach, Sc haufelli, & Leiter, 2001). Not surprising then, human service workers have been identi fied as an occupational class with above average risk for burnout (Soderfeldt, S oderfeldt, & Warg, 1995). Consequently, researchers have been studying correlates and consequences of burnout for decades. It is further understood that burnout has a negativ e impact not only upon individual workers but upon entire organizations, including other ag ency staff as well as the clients they serve (Garner, Knight, & Simpson, 2007). This study proposed that consideration should be given to the concep t of “engaging” workers in a rejuvenation effort intended to alleviate the negative and costly effects of burnout in th e field of human services. It has not always been popular to stu dy what might be considered “positive” states of human functioning (S chaufeli & Bakker, 2004). In fact , traditional research has focused on weakness, malfunctioning, and the study of negative states such as burnout. Recently, however, there has been a shift in the focus of studies away from negative conditions, towards research on human stre ngths and optimal functioning (Seligman & Csikszentmihalyi, 2000). This shift aligns with the National Asso ciation of Social Worker’s (NASW) code of ethics, which states that “A defining feat ure of social work is the profession’s focus on i ndividual wellbeing in a social context and the wellbeing of society” (NASW Code of Ethics). Becau se the profession clearly values human wellbeing and empowerment, this shift to studying positive states, such as engagement
2 with one’s work, is long overdue for social work researchers. Research into such predictors of positive states as work engagement could foster the development of new practice strategies designed not to correct a de ficit, but rather to fo ster an individual’s capacity to maximize his/her own functioning. Although employee engagement is a new term in social work, it has been familiar to those in the management commun ity for almost two decades (Hobel, 2006). Engagement is more than simple job satisfa ction and high retention rates. Fully engaged workers are those who are physically energized, emoti onally connected, mentally focused, and feel aligned with the purpose of the agency (Loehr & Schwartz, 2003). Engaged employees have a bond with the organization. These individuals feel empowered and in control of their fate at work. They identify with the agency mission and are willing to commit the necessary emo tional and personal energies necessary to excel in their work. In short, engaged indi viduals willingly help achieve agency goals and are emotionally involved in the tasks of their organization (Buhler, 2006). Having an engaged workforce in the human se rvices field is vitally important because research shows that engaged workers help orga nizations reap benefits such as increased efficiency, higher levels of customer satisfaction, higher productivity, and lower turnover rates (Buhler, 2006). However, a lthough there is a growing body of business- oriented literature that describes how e ngaged employees contri bute to the overall success of an organization, little academic and empirical research in the human services field has been conducted on the topic. More re search is needed to determine levels of engagement for workers in the human servic es field, to describe the benefits of engagement, and to identify what factors may predict it.
3 Conceptual Framework W. D. Kahn (1990) is credited with c onceptualizing the major components of employee engagement. His model proposes th at engagement differs from basic job involvement, in that it focuses not on work er skills but, rather, on how one commits him/herself during the performance of the job. Engagement entails the active use of emotions in addition to the simple use of cognition while completing work tasks (May, Gilson, & Harter, 2004). The major propositions of the model are that people express themselves cognitively, physically, and emoti onally while performing their work roles. The model proposes that, in order for individu als to fully engage with their job, three psychological conditions must be met in the work environment: meaningfulness (workers feeling that their j ob tasks are worthwhile), safety (feeling as though the work environment is one of trust and supportiven ess), and availability (workers having the physical, emotional, and psychological means to engage in their j ob tasks at any given moment) (Kahn, 1990). Another major proposition of the engage ment model is that these three key psychological conditions are, to some degree, within the control of agency management. Employee engagement is also something that is changeable, and can vary widely from one workplace to another (Cof fman & Gonzalez-Molina, 20 02). Studies indicate that workers are, to some extent, a reflection of the administrators of an agency. Low or conversely high engagement scores have been traced back to the organization’s leadership, from top to bottom (Townsend & Gebhardt, 2007). Theref ore, the results of engagement studies should have considerable applicability to the social work field. For instance, leaders in human service agencies could utilize data from engagement studies
4 to create and implement strategies that would increase staff engagement, thereby decreasing the potential fo r burnout and maximizing successful outcomes for the agency and for the clients they serve. Need for the study Highly engaged employees make a substant ive contribution to their agency and may predict organizational success (Saks, 2006). But the reverse holds true as well. Disengaged employees can be a serious liabi lity. Ayers (2006) compares disengagement to a cancer that can slowly erode an agen cy. Customer satisfaction, employee retention, and productivity are all at risk unless burnout and disengagement can be controlled. Unfortunately, some studies show that wo rkers in general are not engaged with their jobs. Frauenhiem’s (2006) review of a recent Sibson Consulting Firm survey found that satisfaction scores with all majo r categories of work in the U.S. have dropped, and just over half of the respondents in the study rated themselves as engaged, or highly engaged. This lack of engagement affects large and small organizations all over the world, causing them to incur excess costs, to under perform on crucial tasks, and to create widespread customer dissa tisfaction (Rampersad, 2006). Disengagement can affect the financial solidarity of an agency as well. Ayers (2006) explains the potential monetary impact by estimating that if an organization has employees who are only 30% to 50% percent engage d then 50% to 70% of the payroll is an ineffective expenditure of agency res ources. And not only are these disengaged staff members taking up resources in pay and be nefits, they also work agains t the best interests of the agency and can actually turn committed em ployees against the organization (Ayers, 2006).
5 To further emphasize how widespread this problem is and how critical it can be, consider a recent Gallup study (Coffman & Gonzalez-Molina, 2002) in which hundreds of companies were surveyed. Results from their surveys showed that 54% of workers were not engaged and 17% were actively di sengaged. In other words, the companies surveyed were operating on only a fraction of the resources that s hould be available to them.. In this study, the most engaged wo rk groups were noted to be the most productive and the rest were shown to be mediocre or, in some cases, destructive (Coffman & Gonzal ez-Molina, 2002). It is not only the management and busi ness literature that note a decrease in worker engagement. Recent research also suggests concerns about rehabilitation professionals’ job satisfaction rates. A 2004 national study conducted by the Rehabilitation Services Admini stration (RSA) indicates that the average salary across the nation for counselors is four to six thous and dollars below salaries in other human service professions (Chan, 2004). The national study also notes that 26% of Vocational Rehabilitation (VR) Counselors are considering leaving thei r agency to find better pay (Chan, 2004). In addition to pay disparity, ot her factors in counsel or satisfaction or dissatisfaction are equally disconcerting. For in stance, research indi cates that students who graduate with a master’s degree in Rehabilitation Counseling perceive VR counselors to be overworked, underpaid, and overloaded with paperwork (Chan, 2004). Published studies then provide clear indi cations that disengagement is a major problem for many types of organizations. This study was designed to help social service administrators understand and foster the pos itive state of staff engagement in their agencies. An engaged workforce may provide a buffer against the costly effects of
6 disengagement and burnout, and may prove to be a critical element in achieving successful outcomes for agencies as we ll as for their individual clients. Statement of the Problem Despite evidence of how destructive employee burnout or disengagement can be, studies from the human services fiel d on the opposite condition, engagement, are limited. Surprisingly little academic and empirical research has been conducted overall, and a large portion of it co mes from the business manage ment community (Saks, 2006). Additionally, studies do not differentiate human services staff from workers in other industries. To address this problem, more research that focuses specifically on the engagement levels of workers in human se rvices occupations is necessary. Empirical data are needed so professionals can be tter understand employee engagement and use what they learn about it to de velop managerial interventions and alternative strategies that foster engagement for human services workers. Significance of the Study The results of this study will provide in sight and information for administrators, practitioners, and researchers about employee engagement in the human services field. Administrators from the agency that wa s evaluated may benefit from the survey feedback, and could implement strategies for change that address participant
Administrators in other agencies can also benefit by understanding how critical engagement is, and that as a positive construct, it can be measured easily in a variety of settings. Once the level of employee engage ment is measured, administrators can develop and implement change strategies th at would actually impr ove staff engagement
7 in their organization, thereby potentially increasing the overall effectiveness of the agency, and possibly decreas ing levels of burnout. Practitioners may benefit by understanding engagement and coming to realize that they will be more successful in serving clients and, at the same time, at less risk for occupational burnout, if they are working at a job in which they can fully engage. Additionally, practitioners who function in supervisor y positions may benefit by understanding that staff usually adopt the characteri stics and attitudes of their leaders, making it difficult for staff to be engaged unless the managers are (Kerfoot, 2008). Supervisors should, therefore, work toward creating a work environment that lends itself to engagement from themselves and fr om the staff. Finally, researchers can use the information to conduct similar studies that will contribute to the knowledge base about staff engagement in human services workers. Researchers should attempt to further determine what similarities in engagement levels may exist across occupations in the field, what factors may predict enga gement, and what the specific benefits of engagement are for human services agencies. Statement of Purpose The purpose of this study was to exam ine factors that predict levels of engagement for employees in a state-wide pub lic human services agency located in the Midwestern United States. Demographic and work life variables were examined to determine if they impacted scores on the employee engagement scale. The exploratory research questions and hypotheses were deve loped following a review of the literature and the completion of a pilot study. As will be explained in Chapter Two, work life variables are thought to be related to le vels of employee engagement. However,
8 empirical studies on work engagement are limit ed and the literature is unclear as to which variables are the stronge st predictors. No identified studies have examined workers specifically in the human service fields such as social work, psychology, or rehabilitation. Therefore, vari ables for this study were c hosen by reviewing the limited data that are available regarding work engagement, followed by examining factors related to burnout. The assumption of this study is that if a factor predicts burnout, that same factor may have an oppos ite relationship to engageme nt. The factors which were explored for this study include office loca tion, employee’s years of service to the agency, gender, and whether or not the em ployee’s job function includes supervising other staff. Research Questions The research questions deemed most important for this study were narrowed down to the following four: 1.
To what extent do employees in rural and urban office locations differ in their report of engagement? 2.
To what extent does an employee’s year s of service effect reported level of engagement? 3.
To what extent do male and female employees differ in their report of engagement? 4.
To what extent does performance of supervisory duties impact employee’s reported level of engagement?
9 Research Hypotheses This study tested the fo llowing four hypotheses: 1.
Respondents from rural office locations will report lower engagement scores than those from urban office locations. 2.
Respondents with more years of servi ce will report lower engagement scores. 3.
Male respondents will report higher le vels of engagement than female respondents. 4.
Respondents with supervisory job titles will report higher engagement scores. Variables The Dependent variable in this study was the employee’s total score on the eight item employee engagement scale. The Inde pendent variables were the employee’s number of years of service in the agency, the location of the empl oyee’s office, his or her gender, and supervisory job duties. Type I cutoff was set to .05. Definition of Terms For the purposes of this study, the follo wing words and phrases are defined as follows: Employee Engagement . Engagement is more than simple job satisfaction. It can best be described as a harn essing of one’s self to his or her roles at work. In engagement, people express themselves c ognitively, physically, and emotionally while performing their work roles (Kahn, 1990). Disengagement . In disengagement, people withdraw and defend themselves physically, cognitively, or emotionally while performing their work roles (Kahn, 1990).
10 Burnout . Burnout is a psychological syndrom e that results from dissatisfaction with one’s job and involves exhaustion, c ynicism, detachment from the job, and ineffectiveness (Maslach, 2003). Human services workers . Individuals who are empl oyed in a wide variety of fields, including social wor k, psychology, or rehabilitation. Th ey provide direct client services or supervise and administ er client services programs. Office location . For this study, office location was defined by the participating agency as (a) Urban: Those offices in the Ka nsas City, Missouri, or St. Louis, Missouri, region, and (b) Rural: Those offices outside of the Kansas City or St. Louis regions. Assumptions, Limitations and Delimitations A major assumption of this study was that all participants would answer the survey questions truthfully. Anecdotal comm ents from participants in the pilot study noted that the instrument was simple and took only minutes to complete. It therefore was assumed that a significant number of those surveyed would respond. Additionally, it was assumed that the various offices of th is agency operate in a similar manner with respect to policies and practices despite the fact that they are located in geographical areas that span the state. There are notable limitations to this study. It is understood that biases may be inherent in self-reported information, for in stance. Additionally, the perceptions of the employees examined in this survey are speci fic to the rehabilitati on field and may differ from the opinions of staff working in othe r disciplines; therefore, caution is urged regarding external validity. Because the survey was conducted using an internet-based
11 program, there is the possibility that particip ants will be concerned about the privacy or confidentiality of their re sponses, causing fewer to participate fully (Couper, 2000). One typical limitation associated with internet surveys is th e obvious factor that not all households have internet access (Norusis, 2006). However, in this study, the survey was sent to the employee’s work site and, in this setting, every employee has access to a computer with internet service, thus maki ng this one of the delimitations of the study design. An additional delimitation is noted in that participants were given permission by the administration of the agency to answer the survey during working hours. Also, the survey was in a simple, easy to follow fo rmat and took only minutes to complete. Conclusion This chapter focused on the overall purpose and direction for this study. Chapter Two will highlight the key research in this study ar ea with a review of th e pertinent literature. The following Chapters Three, Four, and Five will provide a detailed description of the study, its findings and a discussion of its imp lications for policy, pr actice, and research in the field of human services.
12 Chapter Two: Literature Review Although there is abundant so cial services research on burnout, empirical data on employee engagement are limited (Freeney &Tiernan, 2006). Most of what has been written has come from consulting firms and t hose practicing in the field of management (Saks, 2006). The following review of the literature will discuss the evidence on burnout, including its costly ne gative effects and how it is now being compared to engagement. Highlights from previous studi es on employee engagement will also be reviewed. Additionally, studies which examine predictive factors for burnout and engagement will be explored. Evidence on Burnout The relationship that people have with their work and the difficulties that can occur if that relationship goes awry have l ong been recognized as a significant social problem and the term “burnout” is now rou tinely used to describe this phenomenon (Maslach et al., 2001). Numerous studie s on burnout have been conducted which conclude that job burnout is a psychological syndrome that can develop in response to chronic stressors at work (Maslach et al ., 2001). It is also und erstood that burnout can negatively impact individuals a nd agencies in many ways. Maslach et al. (2001) divi ded the negative consequen ces of burnout into two categories, job performance and indivi dual health. Job performance burnout is associated with absenteeism, intention to the leave the job, and actual staff turnover. Researchers have reported that there is a negative monetary impact from staff turnover. In some organizations, the price of recrui ting and training new workers can be more than $50,000 per employee (Yoon & Kelly, 2008). When staff that experience burnout
13 choose to stay in their respective jobs, thei r productivity and eff ectiveness decreases. Additionally, people who are burned out can cau se personal conflicts on the job site and may disrupt the job tasks of their co-workers . In other words, burnout is contagious and can perpetuate itself on the job. There is al so some evidence that burnout can spillover into an employee’s home life. Finally, the health component of burnout is correlated with stress-related conditions and illness -- substance abuse, a nxiety, depression, and decreased self-esteem have a ll been associated with burnout (Maslach et al., 2001). Although burnout has been studied for decades, it continues to be a major problem for all types of organizations (Ayers , 2006). Levels of burnout do not appear to be decreasing and may be on the rise. A recent Conference Board survey in the United States cited that fewer Amer icans are satisfied with their work. In 1995, 56% of those surveyed were satisfied with their vacation policies, for instance, and ten years later only 51% were. Satisfaction with physical faci lities also showed a decline over the 10 year study, from 56% down to 52%. Moreover, age and income did not seem to affect employee satisfaction, either. The noted tr end was simply all downward (Baldwin, 2005). In addition, the Sibson Consulting Firm recently conducted a survey of 1,200 US employees. It found that, in general, satisfa ction scores with al l major categories of work have dropped, and just over half of th e respondents to their study rated themselves as engaged or highly enga ged (Frauenhiem, 2006). Coffman and Gonzalez-Molina (2002) indicat e that, in general, employees start out energized, engaged, and ready to take on the responsibility of the job. However, over the first six months of employment onl y 38% remain engaged and after ten years engagement drops to about 20%. This is a pa rticularly alarming statistic when one of
14 the most important characteristics for any human services agency is the quality and effectiveness of its personnel (Ewalt, 1991). Another recent area of concern and possibl e factor in high burnout levels is the increased demand for services with a subseque nt decrease in federal and local revenues to provide programming. As a result, admini strators and program managers are focusing more on efficiency, effectiveness of servic es, and increased acc ountability of direct service providers which, in turn, can increase stress a nd dissatisfaction among staff (Ritchie, Kirche & Rubens, 2006). Additionall y, human services workers are operating in an increasingly bureaucratic system whic h, in many settings, limits the effect a staff member can have on a client’s life. This bur eaucratization allows for better control and coordination of large numbers of workers in one location, but is also increases the division of labor. As a result, worker is olation, fragmentation, and de-skilling of workers is going on (Arches, 1991). A majority of the previous research on burnout has focused on identifying its antecedents and outcomes. Only in recent year s have researchers started to pay attention to the opposite state of burnout: employee e ngagement (Gonzalez-Roma, Schaufeli, Bakker, & Loret, 2004). Now studies are be ing conducted which seek to compare the positive construct of engagement with the ne gative state of burnout. The proposition is that an understanding of engagement could yield perspectives about how to alleviate burnout (Maslach et al., 2001). It is believed that, in the process of building an environment where employees are likely to become engaged, many of the problems associated with stress and dissatisfaction in the workplace will be overcome (Freeney & Tiernan, 2006). This requires a focus not only on burnout prevention, but also on
15 developing strategies to fost er energy, involvement, and effectiveness in employees (Maslach & Leiter, 1997). Researchers in these recent studies have investigated staff engagement to determine if it is the pola r opposite of burnout (Maslach et al., 2001, Schaufeli & Bakker, 2004, Gonzalez-Roma, et al. 2004, Freen ey & Tiernan, 2006). So far, research findings support the pro position that engagement is the an tithesis to burnout (Freeney & Tiernan, 2006). Engagement is said to be characterized by vigor, dedication, and absorption, whereas the core dimensions of burnout are described as exhaustion, cynicism, and inefficacy (Gonzalez-Roma et al. 2006). Burnout and engagement are further reported to be opposite in that they have different conse quences and different predictors (Schaufli & Bakker, 2004). Kahn (1990) compared burnout with dise ngagement and said that disengaged employees are ones who withdraw from the j ob physically, emotiona lly, and cognitively which, in turn, likens it to th e state of burnout (Freeney & Tiernan, 2006). An important distinction between engagement and burnout is that burnout re lates specifically to job demands. Engagement, on the other hand, is indicated by job resources such as job control, the availability of learning o pportunities, access to necessary materials, participation in the decision-making process, positive reinforcement, and support from colleagues (Freeney & Tiernan, 2006). Maslach and Leiter (1997) conducted burnout profiles with staff in two hospital units. Employees in one unit displayed typi cal burnout profiles, scoring unfavorably across six areas of measurement; by compar ison the other unit scored favorably on factors related to engagement, includi ng workload, control, fairness, and
16 meaningfulness. One of the most interest ing findings of the researchers was that patients in the units where staff were engaged were significantly more satisfied with the level of care than those w ho were treated in units wh ere the staff was burned out (Maslach & Leiter, 1997). Recent research indicates then that the conditions of engagement and burnout are unquestionably linked, or indeed opposites of one another. It should be understood that an engaged employee will not be burned out, but an employee who is not burned out is not necessarily engaged (Freeney & Tierna n, 2006). The comparison research between burnout and engagement is an important contribution to the field and should be continued. However, it does not look specificall y at workers in the human services field and does not identify indicators and conse quences of engagement in human service occupations. Previous Studies of Employee Engagement The most comprehensive study to date of employee engagement has come from The Gallup Organization’s resear ch using the Q12 instrument. For more than 50 years, the Gallup poll has been questioning cust omers and employees on a variety of workplace topics. Their survey s attempt to find out more than simply how satisfied persons are with their jobs. It addresses th e extent to which employees needs are being met and examines the emotional ties they have to their employment. According to Coffman and Gonzalez-Molina (200 2), results of this Gallup research have shown that business units in which employees score in the top half on employee engagement have, on average, a 56% higher success rate with cu stomer loyalty, a 44% higher success rate