Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Psychology. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 02 March 2021

The Psychology of Work Engagementfree

  • Michael P. LeiterMichael P. LeiterDeakin University, Department of Psychology


Engagement has continued to develop as a positive construct in organizational psychology. Initially defined as employees’ identification with their work, work engagement became understood as a configuration of vigor, dedication, and absorption that motivates exceptional work performance. Although generally viewed as a positive construct, engagement may have a dark side in giving work excessive importance in employees’ lives. There has been some debate regarding the specific qualities that define engagement and the extent to which engagement is an enduring trait in contrast to a varying response to situational constraints and opportunities. The concerns are reflected in the measures of engagement, the most widely used is the Utrecht Work Engagement Scale (UWES). The Job Demands/Resources Model has structured much of the research work on engagement in recent years, leading to initiatives to enhance engagement by improving the quality and variety of resources available to employees at work. Within this domain, job crafting appears to provide a means through which individuals or groups may broaden their opportunities to participate in engaging activities while reducing the range of drudgery inherent in their work.


The Truth about Burnout (Maslach & Leiter, 1997) proposed engagement with work as the opposite of burnout. This idea developed from observations that surveys across organizations or professions did not divide neatly into a burnout profile of exhaustion, cynicism, and inefficacy in contrast to all the normal people with nondescript psychological connections with their work life. Instead, the three subscales of the Maslach Burnout Inventory (MBI; Maslach, Jackson, & Leiter, 2017) reflected a smooth distribution, skewed in the positive direction. Samples included people who never or very rarely experienced exhaustion or cynicism and who felt efficacious nearly every day. This combination of scores reflected something more than the absence of burnout. It went beyond suggesting an adequate psychological connection with to work to describe an active state of energy, involvement, and efficacy. Maslach and Leiter (1997) deemed a strongly positive profile of MBI scores as worthy of a distinct label—engagement with work—that they proposed as a counterpart to the burnout profile.

Meta-analyses have questioned the extent to which burnout and engagement occupy distinct conceptual spaces. Most directly, Cole, Walter, Bedeian, and O’Boyle (2012) question their distinctiveness by noting strong correlations of the Utrecht Work Engagement Scale (UWES) with the Maslach Burnout Scale, the two measures’ consistent relationships with organizational constructs, and the extent to which entering burnout into multiple regressions substantially reduced the power of work engagement as a predictor of organizational outcomes. Crawford, LePine, and Rich (2010) challenged the Job Demands/ Resources prediction linking demands with burnout and resources with engagement. They found both demands and resources associated with burnout, while engagement, measured predominantly by the UWES, had distinct relationships with hindrance versus challenge demands. Further, MacKay, Allen, and Landis (2017) found that employee engagement, as assessed with various measures, made only modest contributions to explaining outcome variance beyond the power provided by various attitudinal measures. Overall, these meta-analyses find limited support for engagement’s capacity to provide a unique perspective on work life.

The Concept of Work Engagement

The UWES reliably captures a psychological connection with work that occupies the positive end of a spectrum that has negative connections, such as burnout, at the opposite end. Macey and Schneider (2008) noted that introducing a new positive construct into the thickly populated territory of organizational psychology or management studies presents serious challenges. For much of the past century the field has explored ways to assess employees’ affiliation for their work and their workplaces. They essentially argue that construct redundancy presents a greater risk than construct neglect. In their overview of definitions and measures of engagement they find insufficient rigor in distinguishing new measures from established measures. They note instances in which writers simply relabeled job involvement or job satisfaction as engagement. In contrast, some studies have developed measures for specific subsets of engagement, specifically organizational engagement in contrast to job engagement (Saks, 2006).

In a comprehensive review of employee engagement, Macey and Schneider (2008) noted that research had identified a range of factors including personality qualities, management practices, and workplace social relationships that contribute to or inhibit engagement. This review argued for engagement as a complex construct having qualities of both a state and trait. They also presented engagement as a complex construct having qualities of both a state and a trait, and further presented engagement as a strategic orientation, in that employees assessed their work situations to evaluate the extent to which engagement behaviors would further their workplace aspirations. That is, engagement does not just happen to employees; some seek out opportunities for engagement as part of their career strategy.

Macey and Schneider (2008) criticize some measures as assessing the conditions for engagement, such as work-based friendships or supervisory attention, rather than assessing engagement per se. They singled out the UWES as having the distinct quality of a foundation of rigorous development as a measure. Analyses have supported the contention that the UWES defines work engagement as distinct from other positive work-related constructs (Christian, Garza, & Slaughter, 2011). However, others have found the UWES to overlap excessively with measures of burnout as well as organizational commitment and job performance. Byrne, Peters, and Weston (2016) concluded that the UWES overlapped considerably with other positive work-related constructs. They also noted that the UWES did not stand in direct opposition to burnout, although it did contrast with burnout in important ways. Although the UWES enjoys the status of the most widely used measure of work engagement, its use has its detractors.

Measuring Engagement

Although Kahn had introduced engagement as a construct in work psychology earlier (Kahn, 1990) research on work engagement proceeded at a modest pace until the introduction of the Utrecht Work Engagement Scale (Schaufeli, Salanova, González-Romá, & Bakker, 2002). In developing the work engagement scale Schaufeli and colleagues sought to define the construct with reference to the active presence of vigor and dedication rather than deducing engagement from the absence of exhaustion and cynicism, as with the MBI. The vigor subscale contrasts with exhaustion, while the dedication scale contrasts with cynicism. Schaufeli and colleagues introduced absorption as a feature of engagement because their theoretical model included a quality of flow as a part of engagement, and employees had noted the importance of absorption when interviewed about their experiences with work engagement. The developers of the UWES expressed reservations about the validity of efficacy as an element of burnout. They did not see fit to include efficacy as an aspect of engagement although they noted efficacy’s relevance to the engagement construct. However, developing the scale as a direct contrast to the Maslach Burnout Inventory-General Scale (MBI-GS) required Schaufeli and colleagues (2002) to address a question presented by the professional efficacy scale. To capture burnout as a complex syndrome, the MBI uses negatively worded items to assess exhaustion and cynicism while using positively worded items to assess efficacy. The burnout syndrome includes high scores on exhaustion and cynicism and with low scores on efficacy. The MBI has received criticism from the view that the relatively low correlations of the efficacy subscale with the other two subscales suggested that efficacy did not operate as part of a syndrome (Salanova, Grau, Cifre & Llorens, 2000). In contrast, the consistently strong correlation of exhaustion with cynicism indicated that these two constructs defined the core burnout construct. Alternatively, the relatively independent status of efficacy leaves room for employees to experience exhaustion and cynicism with a positive sense of efficacy as a state distinct from the full burnout syndrome, which includes a negative sense of efficacy. They may attribute their exhaustion and cynicism to excessive demands aggravated by poor management practices while maintaining a strong sense of their own capabilities. This psychological connection with work could exist distinct from burnout. In that way, low efficacy may function as active part of the burnout syndrome without moving in lockstep with the other two aspects of the syndrome. This framework follows the pattern found in medical syndromes. Some syndromes comprise highly correlated elements but in other instances the congruence of rarely combined elements define syndromes (Melnick, Powsner & Shanafelt, 2017). The burnout syndrome includes both features.

In their analysis Schaufeli and colleagues (2002) contrast a single-factor model (engagement and burnout as opposite ends on a continuum) and a two-factor model of the three MBI subscales as one factor in contrast to the three UWES subscales as a second factor. Neither model met the criteria for a well-fitting confirmatory factor analysis (e.g., CFI of .83 and .85 respectively), nor did their model fit statistics differ markedly from one another. A modification that improved the fit adequately (CFI = .92) occurred through including the MBI efficacy subscale with the UWES subscales as one factor in contrast to a latent factor of exhaustion with cynicism. The resulting latent factors correlated at −.47 and −.62 in their two samples.

This analysis successfully established a positively constructed measure of engagement distinct from positive scores on the MBI; however, the results included caveats. First, the latent engagement variable in the final model that contrasted with burnout included a subscale of the MBI. The analysis did not delete the efficacy subscale, but included it with the UWES subscales in the final model. This approach did not align clearly with the authors’ reservations about efficacy as an aspect of either burnout or engagement. Second, the −.62 correlation between the engagement and burnout factors for the employee sample did not entirely counter the argument that engagement and burnout function as opposite psychological connections with work. Although this correlation falls short of a perfect −1.00 that would indicate a complete opposite, r=−.62 for a sample of over 900 people does not argue convincingly that engagement and burnout function independently of one another. In many contexts, researchers would interpret a correlation of such magnitude as reflecting a strong relationship between two constructs.

The UWES has become a widely used measure of work engagement. The original 17-item scale produces the three subscales of vigor, dedication, and absorption. In contrast, researches often calculate a single engagement score from the 9-item short form (Schaufeli, Bakker, & Salanova, 2006) that includes items from each of the three subscales. Examinations of the 9-item UWES across many translations of the measure have produced mixed results, with some analyses confirming a three-factor structure and other analyses producing a single-factor solution (e.g., De Bruin & Henn, 2013; Schaufeli & De Witte, 2017). In light of these findings, researchers have at times reported the UWES as a single-factor engagement score. Schaufeli, Shimazu, Hakanen, Salanova, and De Witte (2017) have validated a three-item UWES that includes one item from each of the three subscales to produce a single-factor engagement score (Knight, Patterson, & Dawson, 2017). These three formats maintain consistency with the conceptual framework of work engagement comprising vigor, dedication, and absorption while accommodating a variety of research agendas. The full 17-item version supports delving into the distinct qualities of each subscale; the ultra-short version provides a metric for work engagement while placing the least burden on survey respondents. The UWES can provide this range of usage because work engagement functions as a unitary construct. The subscales differentiate clearly on factor analyses of the full scale but their high intercorrelations allow the combining of the items into a unitary measure of work engagement.

Psychological Connections with Work

In addition to engagement’s contrast with burnout, the field has also considered the distinction between engagement and workaholism. Taris, Schaufeli, and Shimazu (2010) found these two experiences to share qualities, including an attraction to work and willingness to devote extensive time to it. However, engagement corresponded with measures of health and well-being while workaholism did not. Salanova, Del Líbano, Llorens, and Schaufeli (2014) examined this issue through a cluster analysis that distinguished four profiles: relaxed, engaged, workaholic, and burned out. The workaholic profile shared the engagement profile’s qualities of energy, challenge, skills, and identification but differed markedly on pleasure, with workaholics finding less pleasure in their work. In contrast, the burnout profile shared the workaholic profile’s lack of pleasure but scored negatively on all of the other qualities relative to the other profiles. These findings aligned well with the results of Taris and colleagues (2010).

Leiter and Maslach (2016) also used a person-oriented approach to identify distinct profiles of employees’ psychological connections with work. Through latent profile analysis of two large samples of health care providers they identified five profiles based upon scores on the MBI-GS. In this approach they argued that energy, involvement, and efficacy—the constructs that underlie the three MBI subscales—relate to fundamental dimensions of employees’ relationships to work. As such, distinct patterns of these three dimensions would identify fundamental ways of connecting or disconnecting with work.

Leiter and Maslach’s (2016) analysis identified five profiles. Burnout (negative on all three subscales) and Engagement (positive on all three subscales) defined the two poles of the conceptual space. In addition, the analysis identified three additional profiles: Ineffective (negative only on efficacy), Overextended (negative only on energy), and Disengaged (negative only on cynicism). In relating the profiles to organizational constructs, they found that the Overextended had a negative view of workload demands but expressed indifference towards resources issues, such as community, control, and reward. In contrast, the Disengaged profile expressed indifference towards workload demands while expressing concerns about resource mismatches. The Burnout profile views both demands and resources negatively while the Engaged profile viewed demands and resources positively. The Ineffective profile occupied an intermediate position in that it expressed neutral view on both demands and resources that distinguished it from Engagement while stopping short of the distress associated with Overextended, Disengaged, and Burnout. This profile makes a definitive contribution to the conceptual space in characterizing a psychological connection with work that has a neutral quality. It defines a space for people who lack the enthusiasm of engagement without experiencing explicit distress regarding their work life.

A person-centered analysis (Mäkikangas, Schaufeli, Tolvanen, & Feldt, 2013) allows one to identify within a larger population subgroups with distinct and meaningful profiles that fall outside of the range expected on the basis of overall correlations alone. For example, the MBI exhaustion and cynicism scales have consistently large correlations but they do not occur in alignment in every instance. In contrast, the efficacy subscale does not correlated strongly with the other two MBI subscales, but the measure provides important information when they do all align. As Mäkikangas and colleagues (2013) demonstrated in their analysis, focusing on these distinct configurations of scores identifies considerable change and development within a population that appears highly stable on the basis of overall correlations.

The Job Demands/Resources Model

The Job Demands/ Resources (J D/R) model (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001) has guided a large proportion of the research on work engagement. Its first iteration focused on burnout, using the Oldenburg Burnout Inventory (OLBI) that measures exhaustion and disengagement Demerouti, Bakker, De Jonge, Janssen, & Schaufeli, 2001. The model demonstrated a central concept that excessive demands contributed to exhaustion while shortages of resources contributed to disengagement. The model’s use expanded with the introduction of the UWES as a positive measure of engagement, replacing the focus on the negatively worded disengagement subscale of the OLBI (Demerouti, Bakker, De Jonge et al., 2001). Over the subsequent years, a variety of studies have confirmed the core elements of the model. First, tests of the model have confirmed stronger relationships between demands and exhaustion than between demands and engagement. In parallel, results have confirmed stronger relationships between resources and engagement than between resources and exhaustion. Examinations of the model have also demonstrated that these patterns do not operate exclusively. They have shown weaker paths from resources to exhaustion and from demands to engagement as well as indirect effects (Bakker, Demerouti, & Sanz-Vergel, 2014).

Second, research has confirmed the J D/R model’s contribution to predicting variety of organizational and personal outcomes (Bakker & Demerouti, 2014; Bakker, Schaufeli, Leiter, & Taris, 2008) with a focus on job performance. Bakker, Demerouti, and Schaufeli (2003) demonstrated the model’s value in showing two distinct paths to important outcomes related to withdrawal. First, a path from exhaustion predicted illness that in turn predicted absences. Second, a path from resources predicted dedication that in turn predicted turnover intention. Nahrgang, Morgeson, and Hofmann (2011) provided parallel support to the two-pathway quality of the model in demonstrating a pathway from demands through exhaustion led to accidents while a path from resources through engagement led to compliance with safety culture. The inclusion of two distinct paths with related but different outcomes (turnover and absences) gives the model the potential to point towards a greater variety of actions to improve the quality of work life.

The J D/R model does not keep demands and resources entirely separate but includes interactions between demands and resources. J D/R accommodates the buffering functions such as the moderation by control of the relationship between demands and exhaustion (e.g., Karasek & Theorell, 1990). Parallel buffering effects occur through other resources, such as social support and performance feedback (Bakker, Demerouti, & Euwema, 2005; Xanthopoulou, Bakker, Demerouti, & Schaufeli, 2007).

The J D/R model has expanded to include personal as well as job resources as contributors to work engagement. The model depicts personal and job resources as mutually enhancing. Personal resources of consequence include resilience and self-management. These resources not only contribute to employees’ potential for experiencing engagement, they also improve their potential for making use of available job resources. Xanthopoulou, Bakker, Demerouti, and Schaufeli (2007) reported that personal resources played a more active role in the relevance of job resources to engagement than they did in the relationship of job demands with exhaustion.

As Bakker, Demerouti, and Sanz-Vergel (2014) noted, the J D/R model draws its strength to a large part from its flexibility. The model’s focus on resources and demands in the general sense accommodates a variety of strains and assets within an organization. This feature encourages comparisons across organizations in different sectors by accommodating the demands and resources that have distinct prevalence in those organizations. For example, although the actual demands and resources pertaining to the work of a hospital-based nurse differ in important ways form those pertaining to the life of a retail sales person, the general notion of balancing demands with resources has relevance for both.

Interventions to Improve Work Engagement

Having identified engagement with work as a meaningful contributor to both well-being and performance, and having concluded that many workplaces maintain insufficient levels of work engagement among their employees, the question arises as to how to enhance engagement. As with many organizational issues, practitioners confront the intervention conundrum (Leiter & Cooper, 2017). While problems and shortcomings develop in organizations without expending any planning, resources, or effort, interventions to address such problems require impressive and sometimes forbidding quantities of planning, resources, and effort. However, intervention research brings the substantial benefit that the process of deliberately changing a phenomenon for the better demonstrates and deepens understanding. Work to improve engagement has benefitted from the field’s positive focus. Employees need not acknowledge the presence of distress or discord among their employees. They can present efforts to improve engagement as a benefit to everyone involved.

In their meta-analysis of research in work engagement intervention, Knight, Patterson, and Dawson (2017) differentiated programs as building either personal or job resources, training leaders, or building engagement as an aspect of building workplace health. They found that programs designed to increase job resources and programs integrated into organizational well-being initiatives exhibited small but significant effects on improving engagement. They identified a temporal pattern regarding the UWES subcomponents in that the impact on vigor occurred promptly after intervention while the impact on dedication and absorption occurred later. However, follow-up assessments produced mixed results, with some gains diminishing over time while results in other studies sustained. The analysis did not find systematic variation in effectiveness across the four types of intervention (increasing personal resources, increasing job resources, leadership, or health promotion). The small number of studies (20) and the wide variation of specific programs within these four types prevented the authors from making definitive conclusions about the relative effectiveness of these approaches. The analysis did produce support for approaching work engagement intervention as a group project in contrast to individual training or coaching. The authors proposed that group formats might benefit from their greater potential to activate important job resources, such as social support.

Knight, Patterson, and Dawson (2017) noted the impact of logistic problems that plague organizational intervention research. Major events, including mergers, acquisitions, and redundancies occur during intervention implementation and follow-up periods disrupting programs and potentially affecting engagement. Individuals’ choice to participate or drop out appear non-random, but associated with demographic features. The studies included in their meta-analysis varied considerably in the amount of available information about the thoroughness or adeptness of implementation. A weak effect may reflect an inappropriate procedure or an appropriate procedure improperly implemented. They encourage more thorough tracking of process to identify mechanisms of change.

The field remains far from a conclusive judgment on how to increase work engagement due to three inherent challenges. First, Knight, Patterson, and Dawson (2017) decry the paucity of intervention research. Despite work engagement enjoying a surge of research interest that generate publications only 20 studies met their straightforward criteria for inclusion in the meta-analysis. Second, even within their four types of intervention approach, studies varied greatly in specific procedures. The innovation apparent in the variety of intervention approaches came at the cost of comparability among the studies. They identified few points of reference among the diverse approaches. Third, the studies shed only modest light on the mechanisms through which their modest improvements in work engagement occurred.

Job Crafting

Enhancing job resources on the level of workgroups or organizational divisions requires more authority than individual employees can access. Increasing expertise, recruiting support staff, or purchasing equipment costs money. Although organizations regularly enhance resources in these ways, these occasions rarely provide opportunities for research on the impact on employees’ engagement. Instead, research has focused on initiatives that enhance job resources on a more modest scale. Job crafting (Wrzesniewski & Dutton, 2001) occupies the intersection of individual and organizational intervention, in that it includes employee-initiated strategies to access resources and modify their work activities in ways they find personally meaningful. Job crafting works as an individual intervention in that the process coaches individuals on ways to enhance their personal interactions with work. Job crafting works as an organizational intervention in that rather than focusing on personal resources, skills, or attitudes, the process focuses on the people, activities, demands, and resources at work. As such, job crafting produces micro-level organizational change.

Demerouti, Bakker, and Gevers (2015) found that job crafting initiatives directed towards enhancing job resources increased work engagement. In turn, increased work engagement mediated a positive relationship of job crafting with creativity and with extra-role behavior. Both the improved access to resources and the associated positive experience of work increased employees’ level of organizational citizenship.

Petrou, Demerouti, and Schaufeli (2018) noted variations in employees’ use of job crafting in line with the elements of regulatory focus theory (Higgins, 1997) that distinguishes employees as promotion-focused (i.e., concerned with development) versus prevention-focused (concerned with duties). For promotion employees, effective communications about organizational change initiatives increased their job crafting behavior and their associated work engagement. In contrast, for prevention employees, inadequate communication about organizational change initiatives increased their job crafting behavior and their associated work engagement. Their job crafting activity appeared to function as a means of buffering the implications of communication gaps. This analysis indicated ways in which job crafting can operate in line with organizational agendas or operate as a means of assisting individuals to reduce the uncertainty introduced through organizational change.

Mäkikangas, Bakker, and Schaufeli (2017) found evidence that job crafting may require a high level of functioning prior to the intervention. Their analysis found positive effects for job crafting for individuals with preexisting positive affect, including self-efficacy. Further, job crafting worked more effectively in work units with an existing clear sense of mission, a positive bias towards innovation, and good working relationships with organizational leadership. So, thriving individuals in well-functioning work units gain more benefits from job crafting. This pattern supports the notion that initiatives that work well to address problems, such as burnout, may differ from initiatives that work well to enhance further the positive qualities of work life.

Liao, Wayne, and Rousseau (2016) examined a parallel approach, idiosyncratic deals (i-deals). I-deals have a more explicit status than job crafting action sin that employees negotiate idiosyncratic deals with their employers. Job crafting actions may occur more subtly. Job crafting employees may establish a new work pattern or access to new resources through a frank conversation with their supervisors. Alternatively, job crafting employees may operate on a less official scale, devising work initiatives with members of their work groups. On the most subtle scale, job crafting employees may simply nudge their work patterns in new directions without mentioning their plans to anyone. In contrast, employees negotiate i-deals and employers concur with a unique way of working for those who have successfully negotiated i-deals. Hornung and colleagues (2010) demonstrated that in addition to making positive contributions to employee recruitment and retention, i-deals contribute to increased work engagement.

Idiosyncratic deals present employers challenges in managing perceptions of fairness. Employees without i-deals may feel disadvantaged in their calculations of the balance of demands with rewards. Because i-deals form an official psychological contract, they may endure longer than the informal gains from job crafting.

These studies and others (e.g., Attridge, 2009) support the conclusion that organizational initiatives can improve work engagement through programs to improve workplace health, supervisory communication, organizational culture, and leadership.

The Dark Side of Engagement

Work engagement has its dark side (Bakker, Albrecht, & Leiter, 2011). George (2011) noted that increased engagement need not invariably result in a win–win situation. Ideally, greater engagement leads to greater productivity and well-being, but productivity may at times come at the expense of well-being. As noted, the distinction between workaholism and work engagement rests on the extent to which employees find the work enjoyable. However, activities that employees have enjoyed at one time they may not enjoy indefinitely. If engagement motivates employees to devote excessive time at work, the employer’s gain in productivity may come at the cost of employees’ health. Second, George (2011) pointed out that one employee crafting a job to reduce time on unpleasant or tedious tasks may dump these tasks upon their coworkers producing a win–lose situation among employees.

The concept of engagement always held a potential for a dark side. When Kahn (1990) first used the term, engagement, to describe employees’ psychological connection with work, he presaged its complex implications for well-being by describing engagement as “harnessing” employees to their “work roles” (Kahn, 1990, p. 694). That phrasing casts some doubt upon work engagement as a win–win situation. The employer appears to win at the cost of employees’ autonomy, at the least with potential adverse consequences for their health and well-being.

Sonnentag pointed towards mechanisms through which work engagement could affect well-being negatively. As found in Sonnentag, Binnewies, and Mojza (2010), work engagement predicted increases in future job demands in a time-lagged analysis. Additionally, increased demands predicted subsequent decreases in work engagement. From this perspective the enthusiasm inherent in work engagement could motivate employees to over-commit to work responsibilities. Halbesleben, Harvey, and Bolino (2009) reported similar concerns with increasing work demands. This mechanism could function to undermine efforts to sustain engagement in the long run.

Sonnentag argued for examining moderators to the relationship of work engagement with positive outcomes. She pointed out that Sonnentag, Mojza, Binnewies, and Scholl (2008) differentiated between engaged employees who could detach from work concerns during non-work times from engaged employees who did not detach in this way. The former group reported more positive affective states at the end of the work week.

These considerations confirm that work engagement does not function as an absolute good. From the perspective of employees, initiatives to enhance work engagement need to anticipate the implications for employees’ potential to fulfill their personal aspirations and to maintain their health while pursuing greater productivity. Initiatives need to anticipate the potential for overwhelming employees with increasing demands or to disrupt balances of work time with non-work involvements. As with so much of organizational psychology, simplicity must also occur in moderation.


Work engagement has a future. The simplicity of the construct at first glance recedes to reveal a psychological connection with work with multiple facets and implications. Thorough examination of measurement issues have confirmed the UWES as an effective tool for assessing engagement as a positive construct. Much work remains to design and test interventions to improve work engagement and to package those processes into readily available programs.

Suggested Readings