Development and Validation of a Measure of Ebullient Supervision: The ES Scale

Abstract

In an increasingly service-oriented economy, the importance of understanding supervisors’ impact on employees working in a fun work environment continues to grow. While researchers have learned much about the negative outcomes of abusive supervisors, relatively little is known about ebullient supervisors and the beneficial outcomes caused by leaders who create fun work environments for employees. The purpose of this article is to define the concept of an ebullient supervisor and develop and validate a scale to measure the construct of ebullient supervision (ES). A brief overview of the construct is presented along with its anticipated theoretical and empirical relationships with constructs that might be closely related. Construct definitions are developed and items for an ES measure are evaluated using a three-sample validation study. A combined exploratory and confirmatory factor analysis approach was used. This was followed by an analysis of the criterion-related validity. The results show that the ES has convergent validity with other supervisor support measures. One of the distinguishing factors of an ES, however, is that through the supervisor’s actions and behaviors the ebullient supervisor helps create a fun work environment for employees that may have many beneficial outcomes for organizations. Developing a measure of ES allows future researchers to add new insights to the management literature by being able to better investigate the unique benefits of this type of supervision for both organizations and employees.

Keywords

ebullient supervision fun work environments scale development job satisfaction positive psychology work engagement

Many who teach leading cultures in an organizational behavior course include a reference to or even show a YouTube video of Tony Hsieh, president and founder of Zappos. It is impressive to listen to Tony describe his goal of creating a fun place to work. This video includes interviews with employees who describe how much fun they have working at Zappos. What makes Hsieh’s words especially striking is that these employees are performing one of the most routine jobs in the service sector—answering customer phone calls at a call center. Many conclude from this that leaders who behave and act in ways that create a fun work environment can create positive outcomes for both their employees and organizations.

There are many others who believe in having fun at work and their endorsements of a leadership style that makes work fun can be found in the writings of Walt Disney, Norm Brinker, founder of Chili’s, and in Frieberg and Frieberg’s (1998) description of Southwest Airlines. Walt Disney, for example, is quoted as saying, “You don’t work for a dollar—you work to have fun” (Disney, 1994, p. 80). Brinker stated, “If you have fun at what you do, you’ll never work a day in your life (Brinker & Phillips, 1999, p. 195). “Even Kouzes and Posner (1995) note in their classic, The Leadership Challenge, that “If you—and others—aren’t having fun doing what you’re doing, chances are people aren’t doing the best they can do (p. 59).”

William Glasser (1994, 1998) offers a theoretical explanation of the importance of fun by concluding it represents the highest level need in his model of a needs hierarchy. This model suggests people have five basic needs arranged hierarchically as follows: (1) Survival, (2) Love and Belonging, (3) Power and Recognition, (4) Freedom, and (5) Fun. His theory, known as control theory, argues that we are genetically programmed to satisfy our needs and having fun through growth and development is the highest. In mission statements, interviews with distinguished leaders, and several research studies both empirical and anecdotal evidence of the importance of employees having fun at work is provided. Clearly, a supervisor who through actions and behaviors creates a fun work environment for employees, what we have termed—an ebullient supervisor—can make an important contribution to organizational success.

While there is a vast amount of research reported on personal and organizational outcomes caused by negative work environments and the abusive supervisors who create them (e.g., Martinko, Harvey, Brees, & Mackey, 2013; Tepper, 2000, 2007; Zellars, Tepper, & Duffy, 2002), there is little research focused on the positive outcomes caused by supervisors who can create fun work settings or on identifying supervisors and supervisory behaviors that promote fun at work. As the research on emotional contagion and authentic smiling behaviors in customer services grows (Andrzejewski & Mooney, 2016; Barger & Grandey, 2006; Hennig-Thurau, Groth, Paul, & Gremler, 2006), the importance of understanding what supervisory actions create a fun work environment has also grown (Becker & Tews, 2016; Söderlund & Rosengren, 2010). Since the service sector represents a substantial and growing proportion of the economy, this importance will only increase. Smiling customer contact employees who show they are having fun performing their jobs will influence customers of most services in positive ways.

In this article, we contend that the concept of Ebullient Supervision can provide a valuable tool for future researchers seeking to identify leader behaviors and actions that create a fun work environment for their employees. Thus, the article has three purposes. First, this article provides a definition of the construct of Ebullient Supervision. Second, we describe the development and validation of a scale to measure Ebullient Supervision. And, finally, we test the construct validity of this new measurement scale using organizational and employee outcome variables with which we would expect it to correlate.

This article builds on the previous fun at work research of Ford and colleagues (Collison, Cohen, Ford, Newstrom, & McLaughlin, 2002; Ford, McLaughlin, & Newstrom, 2003) as well as that of Karl and her colleagues (Karl & Peluchette, 2006; Karl, Peluchette, & Hall, 2008; Karl, Peluchette, Hall-Indiana, & Harland, 2005; Peluchette, & Karl, 2005) and Tews and his colleagues (Tews, Michel, & Allen, 2014; Tews, Michel, & Stafford, 2013) in their related attempts to define and investigate the benefits of a fun work environment. While this body of research has led to general agreement that fun work environments have positive outcomes for both employees and organizations, there is less agreement on what theories best explain how supervisors create work environments that employees will perceive as fun. The lack of agreement has made it important to investigate and measure the supervisory behaviors and actions that do in fact lead to perceptions of a fun work environment. While seemingly similar to other constructs such as positive supervisory and organizational support (Eisenberger & Stinglhamber, 2011; Kurtessis et al., 2017), well-being (Mäkikangas, Kinnunen, Feldt, & Schaufeli, 2016), thriving (Paterson, Luthans, & Jeung, 2014; Spreitzer, Sutcliffe, Dutton, Sonenshein, & Grant, 2005), and positive affective events at work (Weiss & Cropanzano, 1996), we suggest fun is a valued outcome in itself rather than an aspect of these other constructs. If this argument is correct, it deserves the attention of researchers. The definition of a supervisor who creates a fun work environment and the creation of a measurement scale allowing researchers to investigate the antecedents and consequences of a supervisor that creates a fun work environment is a necessary tool to advance the research in this area.

Defining Ebullient Supervision

Following the recommendation of Podsakoff and colleagues (Podsakoff, MacKenzie, & Podsakoff, 2016) to clearly define a construct, we posit our definition: An ebullient supervisor is one who through words and deeds helps create a work environment that is considered by a subordinate as a fun place to work. While there is a large and growing literature on the benefits of creating positive psychological capital led by Luthans (e.g., Luthans, 2002a, 2002b) and the ways in which supervisors can help employees thrive at work (Paterson et al., 2014), there has been far less effort dedicated to understanding what a supervisor does to make working fun or the impact an ebullient supervisor can have on beneficial employee and organizational outcomes.

The earliest reported systematic study to offer insights into what a supervisor might do to create a fun setting—to be an ebullient supervisor—is seen in a survey of human resource (HR) managers reported by Ford and colleagues (Collison et al., 2002; Ford et al., 2003). This research was based on an extensive review of the practitioner literature to create a questionnaire that was sent to a sample of HR managers from the Society of Human Resource Managers (SHRM), a large HR manager’s professional association. They were asked to indicate the extent to which they agreed that the items most frequently mentioned in the literature characterized a fun work environment. The 10 items measured were recognition of personal milestones, social events, celebrations of professional achievements, opportunities for volunteerism, stress relief activities, games, humor, friendly competitions, opportunities for personal development, and office entertainment. Responses were factor analyzed and found to all load on the same factor. The respondents were also asked to list in an open-ended question other things that their organizations did to promote fun. The majority of things listed represented various ways of celebrating personal milestones or having a special social event with food. The authors concluded from their data that “a fun work environment intentionally encourages, initiates, and supports a variety of enjoyable and pleasurable activities that positively impact the attitude and productivity of individuals and groups” (Ford et al., 2003, p. 22). From this, we would conclude that an ebullient supervisor is one that creates a fun work environment through enabling these types of activities and actions.

Moreover, this study (Collison et al., 2002; Ford et al., 2003) of HR managers asked if there were any advantages or disadvantages to fun work environments. The responding managers indicated that there were many advantages but, more interestingly, that the disadvantages were not that important. In specific, fun work environments led to increased employee enthusiasm, group cohesion, employee satisfaction, organizational citizenship, mutual trust, and reduction of complaints about boredom. Furthermore, organizations where employees are having fun had an easier time recruiting and retaining employees. Thus, based on the response from these HR professionals, having fun at work has many upside benefits and no meaningful downsides. This includes humor (one of the components of this study’s measure of fun), which some research has shown to have dysfunctional outcomes (e.g., Tremblay & Gibson, 2016). The SHRM data showed no support for humor being seen as a negative.

Based on the definition offered earlier, we sought to identify the behaviors displayed and actions taken that characterized an ebullient supervisor. Our definition parallels Tepper’s (2007) definition of an abusive supervisor as one who engages in “the sustained display of hostile verbal and nonverbal behaviors, excluding physical contact” (p. 178) as perceived by a subordinate at a specific point in time. Thus, different subordinates could perceive the same behavior differently or the same subordinate could perceive the same behavior differently in different contexts.

We suggest that the effects of ebullient supervisors displaying behaviors and taking actions that create a fun work environment may yield a fun work environment with a positive effect on subordinates’ work-related attitudes and psychological health. Extending the logic of Petelczyc, Capezio, Wang, Restubog, and Aquino (2018), we suggest that the conservation of resources (COR) theory provides support for the importance of the supervisory style that is represented by the behaviors and actions of an ebullient leader. A core tenant of this theory is that people seek to acquire resources defined as those objects, personal characteristics, conditions, or energies that are valued in their own right or that are valued as conduits to the achievement or protection of valued resources (Hobfoll, 2001). As Glasser (1994, 1988) posits that fun is a need people seek to fulfill at work, we propose that an ebullient supervisor provides a conduit to achievement of valued resources by creating a fun work environment through structuring social, interpersonal or task activities at work.

Besides the many benefits of having fun at work identified in the Ford and colleagues (Collison et al., 2002; Ford et al., 2003) study noted above, other studies have also reported positive outcomes gained from a fun work setting. For example, Karl et al. (2005) report several investigations of the impact of experienced fun or the degree to which individuals experience the existence of fun on the job. In their 2005 study of health care employees, Peluchette and Karl (2005) found that experienced fun was positively related to job satisfaction. In a subsequent study, Karl and Peluchette (2006) found that experienced fun buffered the impact of emotional exhaustion on job dissatisfaction among health care workers. Furthermore, Karl and colleagues (Karl et al., 2008; Karl, Peluchette, & Harland, 2007) reported that experienced fun was positively related to job satisfaction, negatively related to emotional exhaustion, and negatively related to turnover intentions.

In a later study, Fluegge-Woolf (2014) developed a four-dimensional measure of fun, including socializing, celebrating, personal freedoms, and global fun. Socializing refers to friendly social interactions among coworkers; celebrating refers to marking special events and having social gatherings at work; personal freedoms refers to employees being afforded flexibility and autonomy regarding workplace attire, playing music, and playing around at work; and global fun refers to an overall evaluation of whether an organization has a fun work climate. Extending earlier work by McDowell (2004), Fluegge-Woolf (2014), found that fun has a positive impact on task performance and organizational citizenship behavior (OCB).

In another study focused on the hospitality industry, Tews, Michel, and Bartlett (2012) examined the impact of three forms of fun in the context of applicant attraction. They asked a sample of undergraduate job seekers to evaluate hypothetical recruiting scenarios that included three forms of fun: fun activities, fun coworker interactions, and fun job responsibilities. Their results demonstrated that fun was a stronger predictor of applicant attraction than compensation and opportunities for advancement. In a later study with Becker (Becker & Tews, 2016), Tews examined the impact of 12 fun activities on employee engagement, constituent attachment, and employee turnover with a sample of 205 employees from 11 hotels. The results indicated that fun activities were related to both engagement and constituent attachment but not turnover.

While there are other scholars supporting the value of fun at work (e.g., Han, Kim, & Jeong, 2016; Pryor, Singleton, Taneja, & Humphreys, 2010), some argue that more research is needed to validate the outcomes claimed (e.g., Owler, Morrison, & Plester, 2010). Additional support comes from Pryor et al. (2010) who concluded from their research review that workplace fun positively affects key performance indicators such as safety, quality, productivity, worker retention, and achieving goals. Likewise, Han et al. (2016) reported that fun work activities both create a fun work environment and enhance interpersonal trust and group cohesion. Other writers seek to expand the definition to better describe both what fun is and how it affects organizational outcomes (e.g., Plester, Cooper-Thomas, & Winquist, 2015).

This study utilizes the original definition of a fun work environment by Ford and colleagues (Collison et al., 2002; Ford et al., 2003, Ford, Newstrom, & McLaughlin, 2004) to investigate how the behaviors displayed and actions taken by an ebullient supervisor lead to the benefits of a fun work environment by being the conduit for providing the resources desired by employees through actions and deeds that help create a fun work environment. In other words, if the research shows that working in an environment that is fun has benefits for both the organization and the employees, then it is important to identify the behaviors and actions of supervisors that help create an environment that satisfies this need for fun.

At the present time, there is no existing construct or a way to measure how a supervisor through behaviors and actions can create a fun work environment. Thus, we are proposing a new construct of Ebullient Supervision and creating the Ebullient Supervision Scale to measure this new construct. To address the construct validity of this scale, it is necessary to examine the criterion related validity of the Ebullient Supervision Scale. In other words, we are determining whether the Ebullient Supervision Scale is behaving the way it should, given everything we know about the theory on which the Ebullient Supervision Scale is based.

Expected Outcomes of Ebullient Supervision

Based on a review of the extant literature, there are a number of variables and outcomes that are anticipatory of creating a fun work environment, including ebullient supervision. As part of the development of the ebullient supervision construct and a scale to measure it, we examine commonly studied organizational outcomes, including job satisfaction, work engagement, OCB, and intention to leave. These outcome variables have been shown to be related to similar constructs such as positive supervisory and organizational support (Eisenberger & Stinglhamber, 2011; Kurtessis et al., 2017), well-being (Mäkikangas et al., 2016), thriving (Paterson et.al., 2014; Spreitzer et al., 2005), and positive affective events at work (Weiss & Cropanzano, 1996), It seems plausible that these commonly studied outcome variables will behave in a similar fashion to our proposed ebullient supervision construct as they have to other similar constructs.

Job Satisfaction

In the SHRM study (Collison et al., 2002; Ford et al., 2003) of HR managers, 96% responded that they felt a fun work environment improved employee satisfaction. In their study, Karl et al. (2007) found that health care workers who experienced fun in the workplace had significantly higher job satisfaction and lower emotional exhaustion. In a related study of undergraduate students Karl and Peluchette (2006) found that experiencing fun at work had a positive impact on job satisfaction. Choi, Kwon, and Kim (2013), in another study of college students concluded that those who experienced more fun at work had a more positive attitude and affective state toward their job, which the authors equated with job satisfaction.

These studies provide support for a positive relationship between work place fun and job satisfaction. One argument for promoting supervisory behavior and deeds that are seen as ebullient is that this type of supervision leads to the opposite outcomes for employee and organization than those that would result from abusive supervision. Tepper (2000) found there was a strong negative correlation between job satisfaction and abusive supervision for those employees that perceived that they could find another acceptable job if they quit. Thus,

Hypothesis 1: Ebullient supervision is positively related to job satisfaction through the indirect effect of fun work environment.

Work Engagement

Fluegge-Woolf (2014) used the job-demands resources model to study if fun at work enhanced work engagement. Arguing that the social and interpersonal nature of fun at work may serve as a job resource and, thus, enhance work engagement, Fluegge-Woolf (2014) studied employees and supervisors and found support for her prediction that fun at work would be positively related to work engagement. In a later study investigating how fun increased engagement, Plester and Hutchison (2016) conducted 59 structured interviews of New Zealand employees and reported that fun at work increased job engagement. Ford et al.’s (2003) SHRM study found that a fun work environment decreased anxiety and stress and reduced boredom, leading to an increase in work engagement. In their study of what creates job engagement in retail employees, Roberts and Davenport (2002) identified one of three main drivers of job engagement as “the company offering a positive, fun work environment” (p. 25). Thus,

Hypothesis 2: Ebullient supervision is positively related to work engagement through the indirect effect of fun work environment.

Organizational Citizenship Behavior

Lamm and Meeks (2009) in their study of the generational differences related to workplace fun found that fun had a significant positive relationship with OCB across the three generations they studied. Fluegge-Woolf (2014) tested a hypothesis that fun at work builds trust with one’s coworkers and, consequently, OCBs. Her findings suggest that fun at work is positively related to OCBs. Choi et al. (2013) also found a significant positive relationship between workplace fun and OCBs. Thus,

Hypothesis 3: Ebullient supervision is positively related to OCB through the indirect effect of fun work environment.

Intention to Leave

Tews et al. (2013) argue that fun activities and the supervisor’s support for fun activities create an environment that facilitates the development of friendships across the employees in the organization. These friendships will reduce the likelihood of turnover in the organization. Furthermore, these researchers (Tews et al., 2013) suggest based on their study of hospitality workers that because people spend large amounts of time on the job, they are less likely to leave if their work experiences are enjoyable and fun. Perrin (1998) had earlier argued that an important benefit of having fun at work is the organization has lower turnover. HR managers agreed in the SHRM survey (Collison et al., 2002, Ford et al., 2003) that an advantage of a fun work environment is less employee turnover. Likewise, McDowell (2004) based on his fun climate study found that working in a fun climate is significantly and negatively related to intention to leave. Thus,

Hypothesis 4: Ebullient supervision is negatively related to intention to leave through the indirect effect of fun work environment.

Thus, through creating a fun work environment the ebullient supervisor is able to benefit both employee and organization in positive ways. As stated earlier, the purpose of this article is to first define, develop, and test a measure of ebullient supervision and, second, to empirically investigate the benefits of ebullient supervision on organizational and employee outcomes. Below, we first describe how we developed and tested a measure of ebullient supervision and then we present tests of our hypotheses that demonstrate the benefits of ebullient supervision on organizational and employee outcomes. The conceptual model describing these relationships is presented in Figure 1.

Figure 1.

Conceptual model.

Method

A three-study process was used in the development of a scale to measure the new construct of ebullient supervision. The process was designed to create item generation, scale development and scale evaluation according to the steps enumerated by Hinkin (1995).

Study 1: Scale Development and Item Generation

PhD students at a large research university located in the southeast United States participating in a research methods seminar course acted as an expert panel generating the first round of items. They were tasked with reviewing articles on constructs similar to or opposite from the proposed ebullient supervision construct, including abusive supervision and perceived supervisor support. An initial definition of ebullient supervision was generated from this review of the literature. Based on the literature and newly formed definition, the panel of experts generated more than 100 items that tapped into the initial definition of ebullient supervision. These items were then reviewed by scholars at several large research universities in the United States. This review and the graduate students (guided by their professor as a class exercise) pared the items down to a 15-item scale that both captured and measured the essence of the new construct of ebullient supervisor. Study 1 was conducted to test the reliability and dimensionality of the Ebullient Supervision Scale.

In Study 1, 99 MBA. students at a large Southeastern research university were recruited to complete the new Ebullient Supervision Scale. All participants were at least 18 years old, had worked a minimum of 2 years, and employed by their current supervisor for at least 1 month. All respondents were asked to respond to the questions with respect to their current supervisor. Fifty-nine of the respondents were male and 40 were female. The average age of the respondents was 32 years and average tenure with the current employer was 3.5 years. Extra credit totaling 1% of the total course grade was provided to each participant as compensation.

An exploratory factor analysis using principal components with Varimax rotation of all 15 ES items revealed a stable structure with one dimension (Table 1). Only one factor was extracted with eigenvalues higher than one. Items with factor loadings equal to or greater than .40 were retained. All communalities were higher than .5, except for item ES1 (.47) and item ES 6 (.44). However, because they were close enough to the threshold, loaded well on the factor, and were theoretically sound, we decided to keep those items. All factor loadings were equal or higher than .68. The factor explained 64.11% of total variance. The Cronbach’s alpha for this measure was .96, indicating a high level of internal consistency.

Table 1.

Factor Analysis Results of Ebullient Supervision Scale Items.

Ebullient supervision items	Factor loading	Communality
ES1: Greets employees in passing	.688	.473
ES2: Goes out of his/her way to brighten the day at work	.821	.673
ES3: Uses a cheerful tone when speaking with subordinates	.811	.658
ES4: Attempts to make people laugh	.821	.675
ES5: Finds uses for appropriate humor at work	.844	.712
ES6: Will gladly take on routine task responsibilities (e.g., schedules, meetings)	.665	.442
ES7: Encourages social activities after working hours	.804	.646
ES8: Helps people feel enthusiastic about their jobs	.910	.828
ES9: Tries to put people at ease	.889	.791
ES10: Smiles when someone enters his/her office for any meeting or discussion	.777	.603
ES11: Lets people know that it’s OK to be playful at work	.714	.510
ES12: Initiates fun activities outside of work	.801	.641
ES13: Praises individual wins	.812	.659
ES14: Compliments employees in front of others	.888	.788
ES15: Finds reasons to celebrate (e.g., birthdays, group or personal milestones).	.719	.517
KMO measure of sampling adequacy	.920
Bartlett’s test of sphericity	1224.553 (df = 105)***

Note. ES = Ebullient supervision.

***

p < .001.

Study 2: Scale Evaluation

Study 2 was conducted to confirm the structure and dimensionality of the new measure utilizing a new sample, and to assess expected versus actual relationships between Ebullient Supervision and related constructs. Thus, we could test the convergent and discriminant validity of the newly developed Ebullient Supervision measure. Specifically, we explored how Ebullient Supervision relates to Abusive Supervision and Perceived Supervisor Support.

Two hundred and nineteen adults were recruited from Amazon Mechanical Turk (MTurk). Because organizational life, and more specifically the supervisor and employee relationship is omnipresent across many industries and culturally diverse backgrounds, using a sample that reflects this diversity was important for generalizability. For example, the modal household income for MTurk participants in the United States has been reported to be between $40,000 and $60,000. “While 78% have at least a bachelor’s degree, our experience has also revealed that workers come from a wide range of industries and work backgrounds (Barger, Behrend, Sharek, & Sinar, 2011).” In addition to providing a diverse sample, MTurk is quick way to obtain responses. MTurk has been identified as a reasonably representative sampling frame (Buhrmester, Kwang, & Gosling, 2011; Line, Hanks, & Zhang, 2016; Mason & Suri, 2012). Research has also shown that Internet-based studies generalize across presentation formats and are consistent with findings from traditional methods (Gosling, Vazire, Srivastava, & John, 2004; McGraw, Tew, & Williams, 2000).

Participants in our study had to be at least 18 years old, employed full-time and living in the United States. The respondents were given the website of the survey housed on the Qualtrics website. Respondents were asked to respond to the questions with their current supervisor in mind. The survey was pilot tested twice to ensure the questions were understandable. Reverse-coded questions were evaluated to provide evidence that the respondents were carefully reading the questions. Participants were paid $.80 for their anonymous participation. Extreme outliers and respondents that completed the survey in less than one minute and 30 seconds were excluded from the sample. After deleting these outliers, our final sample panel consisted of 213 out of 219 participants (20-62 years old, full-time employment, resident of the United States, 55% male and 45% female).

Measures

As previously described, the 15 items generated in the first study comprised the Ebullient Supervision Scale. Items were measured on a 7-point Likert-type anchored scale (e.g., 1 = Never, 7 = Always). Cronbach’s α was .96.

Abusive Supervision was measured with the 15-item scale developed by Tepper (2000). Sample items from the scale include, “My boss ridicules me” and “My boss tells me my thoughts or feelings are stupid.” Items were measured on a 5-point Likert-type scale (e.g., 1 = I cannot remember him/her ever using this behavior with me, 5 = He/she uses this behavior very often with me). Cronbach’s α was .90.

Perceived Supervisor Support was measured with eight items adopted from the Perceived Supervisor Support Scale developed by Eisenberger, Huntington, Hutchison, and Sowa (1986). Sample items from the scale include, “My supervisor strongly considers my goals and values” and “Help is available from my supervisor when I have a problem.” Items were measured on a 7-point Likert-type anchored scale (e.g., 1 = strongly agree, 7 = strongly disagree). Cronbach’s α was .97.

Results

Confirmatory Factor Analysis (CFA)

In order to assess convergent and discriminant validity between Ebullient Supervision (ES) and Perceived Supervisor Support (PSS), a CFA was conducted. Abusive Supervision (AS) was also included into the model in order to test for the difference between ES and AS. Item 7 error term (“Encourages social activities after working hours”) and Item 12 error term (“Initiates fun activities outside of work”) were allowed to covary since they have similar theoretical justification. After correlating them the model fit improved, but there were still two items in the scale with lower loadings (ES7 = .558 and ES12 = .571) and lower variance (ES7 = .312 and ES12 = .326). Since they were higher than .5 we elected to retain them for this study but decided to eliminate them in any following studies. Most of the fit indices fell outside of the generally accepted standards (Hair, Black, Babin, & Anderson, 2010; χ² = 1682.03, df = 661, χ²/df = 2.54, Tucker-Lewis index [TLI] = .83, comparative fit index [CFI] = .84, root mean square error of approximation [RMSEA] = .08). Since this is not a hypothesized model, but instead, a model built to test discriminant validity between constructs, the model fit was not an impediment to further testing of the ES scale.

All factor loadings were equal or greater than .53 and significant (p < .001). Average variance extracted (AVE) scores for ES and PSS were above .5, and AS had an AVE score of .47. Since the variables of interest were above the threshold of .5 and AS was nearly at the threshold, we believe an argument can be made that these results confirm convergent validity (Hair et al., 2010). We expected that ES and PSS, as related constructs would be correlated, but not so highly correlated as to indicate they were measuring the same construct.

In all cases, the square root of AVE was found to be higher than the intercorrelations between two constructs of interest (Fornell & Larcker, 1981). The correlations between all other constructs were low, providing further evidence for discriminant validity. In addition, the correlation between ES and AS was negative (r = −.46, p < .001), as predicted, since they represent opposite constructs. The composite construct reliability exceeded the recommended .70 threshold for all variables (Fornell & Larcker, 1981).

In order to further test for discriminant validity between ES and PSS, we compared the hypothesized two-factor model (ES and PSS) to a nested model (one-factor model). Results of CFA indicated that the two factor-model (χ² = 602.79, df = 228, χ²/df = 2.64, TLI = .903, CFI = .913, RMSEA = .08) fits these data better than the one factor nested model (χ² = 1073.52, df = 229, χ²/df = 4.69, TLI = .783, CFI = .803, RMSEA = .132). Furthermore, the chi-square difference test showed that the two-factor model fits these data significantly better than the one nested model tested (Δχ² = 470.73, p < .001). In sum, these results support our belief that this new construct is distinctly different from PSS and opposite to AS.

Study 3: Test of Criterion Related Validity

In this final phase, the criterion-related validity of the ES construct using a broad sample was assessed. To do this, we tested whether the ES construct predicted well-established work-outcomes. The hypotheses stated above were tested in Study 3 using 599 adults recruited from MTurk. Extreme outliers and respondents that completed the survey in less than 3 minutes were removed. After deleting these 20, our final sample consisted of 579 participants (20-72 years old, 47% male, and 53% female). The qualifications for the respondents and process were the same as that used in Study 2. This survey was longer, resulting in the respondents being paid $1.65. Most respondents (66%) had worked full-time for over 10 years, 19.1% worked from 5 to less than 10 years, 12.8% worked full-time from one to less than 5 years, and 2.1% worked full-time less than 1 year. More than two thirds had worked for their current supervisor for 2 years or more. Most respondents (64.6%) reported having had more than five supervisors during their careers. Less than 1% did not have a high school diploma, 9% had a high school diploma, 30.2% had some college or an associate’s degree, 40.6% had a bachelor’s degree, and 19.9% had a graduate degree.

Measures

Fun work environment (FWE) was measured using the 10 FWE items developed by Ford et al. (2003). This scale had been tested in a subsequent study, citing a Cronbach α of .87 (Ford, McLaughlin, & Newstrom, 2005). The scale asks respondents to indicate their perception of the frequency their organizations do things that create a FWE. These items include the following: recognize personal milestones (e.g., birthdays, birth of a child, hiring anniversaries); sponsor social events (e.g., picnics, parties, social gatherings); and use of humor (e.g., cartoons, jokes in newsletters, emails). The items were measured on a 5-point scale from not at all to extensively.

Job satisfaction (JS) was measured using the Cammann, Fichman, Jenkins, and Klesh’s (1983) three-item scale used by Tepper (2000) with a coefficient α of .93. Examples of items included “All in all, I am satisfied with my job” and “In general, I don’t like my job” (reverse scored). The items were measured on a 7-point scale from strongly disagree to strongly agree.

Intention to leave (ITL) was measured using a three-item scale developed by Meyer, Allen, and Smith (1993). The items are as follows: “I think about leaving my current employer.” “How likely is it that you will search for a job in another organization?” “How likely is it that you will actually leave your organization within the next year?” The items were measured on a 7-point scale with the first item scaled from strongly agree to strongly disagree and the other two anchored from extremely likely to extremely unlikely. The scale had a coefficient α of .71.

Organizational citizenship behavior (OCB) was measured using the OCB Short Form (Spector, Bauer, & Fox, 2010). The scale consisted of 10 items and had a coefficient α of .80. Items are directed toward coworkers and organization. Examples of the items are as follows: “Took time to advise, coach, or mentor a co-worker.” “Lent a compassionate ear when someone had a work problem.” “Volunteered for extra work assignments.” “Gave up meal and other breaks to complete work.” The items were measured on a 5-point anchored scale from never to more than twice a week.

Work engagement (WE) was measured using a 9-item scale derived from the 17-item Utrecht Work Engagement Scale originally developed by Schaufeli, Bakker, and Salanova (2006). Based on an analysis of the original Utrecht Work Engagement Scale, Seppälä et al. (2009) concluded that a 9-item scale (α varied from .75 to .87) was more efficient yet remained reliable and valid. Examples of questions include the following: “At my work, I feel that I am bursting with energy,” “I find the work that I do full of meaning and purpose” and “Time flies when I’m working.” All items were measured on a 5-point anchored scale from strongly agree to strongly disagree.

Ebullient supervision (ES) was measured with 13 items (α = .94) from the previously developed Ebullient Supervision Scale. Since Items 7 and 12 presented low variance and low-factor loadings in our previous study (Study 2) they were dropped (see the appendix for final scale). Items were measured on a 7-point scale from never to always.

Results

A CFA for each model using AMOS Version 5 was conducted, allowing us to analyze convergent and discriminant validity of constructs. Table 2 shows the results of the CFA for each model. In total four CFAs were run: one for each dependent variable. The measurement model fit of each model was close to satisfactory,¹ considering that ES is a new scale being tested and that the FWE scale has been used in different formats.² RMSEA for all models ranged from .08 to .07, CFIs ranging from .88 to .90, and TLI ranging from .87 to .89.

Table 2.

CFA Results.

Model	Factor loadings	CR	AVE	df	χ²	TLI	CFI	RMSEA
ES → FWE → JS	From .60 to .94	.95	.87	296	1292.648	.89	.90	.08
ES → FWE → WE	From .51 to .91	.92	.56	461	1755.997	.88	.89	.07
ES → FWE → OCB	From .50 to .85	.87	.46	431	1635.851	.87	.88	.07
ES → FWE → ITL	From .60 to .91	.91	.78	296	1285.951	.89	.90	.08
ES		.94	.57
FWE		.89	.46

Note. TLI = Tucker-Lewis index; CFI = comparative fit index; RMSEA = root mean square error of approximation; ES = ebullient supervision; FEW = fun work environment; JS = job satisfaction; ITL = intention to leave; WE = work engagement; CR = construct reliability. CR, AVE present on the models are referring to the outcome variable.

All factor loadings were equal or greater than .51 and significant (p < .001), except for OCB, which had two items loading under .5. In all models, AVE scores for ES, ITL, JS, and WE were above .5. FWE and OCB had relatively lower AVE scores (.46 for FWE and .41 for OCB). Since FWE was close enough to the threshold and had factor loadings higher than .5, we argue that these results confirmed convergent validity (Hair et al., 2010). On the other hand, OCB besides having a lower AVE also had two items with lower loadings. Thus, we decided to delete those two items (“Worked weekends or other days off to complete a project or task” and “Volunteered to attend meetings or work on committees on own time”) in order to achieve convergent validity. After these deletions, OCB’s AVE is .46 with all factor loadings higher than .5. The results shown in Table 2 are related to the model where OCB has eight items. Considering the aforementioned aspects regarding convergent validity, future research should further refine both the FWE and the OCB scales.

In all cases, the square root of AVE was found to be higher than the intercorrelations between the two constructs of interest (Fornell & Larcker, 1981). The correlations between all other constructs were lower than .8, providing further evidence for discriminant validity. The composite construct reliability exceeded the recommended .70 threshold for all variables in every model (Fornell & Larcker, 1981). This provides support that the variables used in this research are measuring different constructs, are valid and reliable.

In order to verify mediation effects of FWE by ES on the dependent variables (JS, ITL, job engagement, and OCB), PROCESS version 3 on SPSS was used. Analysis of histograms and Q-Q plots showed that the shapes of the distributions for most variables approached that of a normal curve. Since some variables presented kurtosis and skewness outside the normal threshold range (±1.96; Abu-Bader, 2010), in order to test the mediation models, percentile bootstrap confidence intervals with 5,000 resamples was used to analyze the hypothesized indirect effects.

Table 3 shows means, standard deviations, and correlations of the study variables. All variables were significantly correlated. FWE (r = .54), WE (r = .44), and JS (r = .44) had the higher correlations with ES scale.

Table 3.

Means, Standard Deviations, and Correlations.

Constructs	M	SD	1	2	3	4	5
1. ES	4.51	1.24	1
2. FWE	2.32	0.81	.54**	1
3. ITL	3.24	1.95	−.33**	−.32**
4. JS	5.33	1.54	.44**	.44**	−.74**
5. WE	3.60	0.89	.44**	.44**	−.63**	.77**
6. OCB	3.29	0.79	.24**	.21**	−.08	.14**	.31**

Note. ES = ebullient supervisor; FWE = fun work environment; ITL = intention to leave; JS = job satisfaction; WE = work engagement; OCB = organizational citizenship behavior.

p < .001. *p < .005.

Table 4 shows the results of all regression models. The positive direct effect of ES on JS controlling for FWE (b = .34, p < .001), and the positive total effect (b =.54, p < .001) were both significant. The positive indirect effect of ES on JS through FWE (b = .20 confidence interval [CI] = [.140, .260]) was also significant, as the 95% percentile bootstrap confidence interval did not include zero. Thus, Hypothesis 1 was confirmed.

Table 4.

Results of Direct, Indirect, and Total Effects.

Relationship	Direct effect	Indirect effect	Lower limit CI 95%	Upper limit CI 95%	Total effect
H1: EB → FWE → JS	.34**	.20	.140	.260	.54**
H2: EB → FWE → WE	.21**	.11	.076	.145	.32**
H3: EB → FWE →OCB	.11**	.04	.008	.072	.15**
H4: EB → FWE → ITL	−.36**	−.16	−.248	−.085	−.53**

Note. CI = confidence interval; ES = ebullient supervision; FWE = fun work environment; ITL = intention to leave; JS = job satisfaction; WE = work engagement; OCB = organizational citizenship behavior.

p < .005. **p < .001.

The positive direct effect of ES on WE controlling for FWE (b = .21, p < .001), and the positive total effect (b = .32, p < .001) were all significant. The positive indirect effect of ES on WE through FWE (b = .11 CI [.076, .145]) was also significant, as the 95% percentile bootstrap confidence interval did not include zero. Thus, Hypothesis 2 was also confirmed.

OCB was also tested as a dependent variable. The positive direct effect of ES on OCB controlling for FWE (b = .11, p < .001) and the total positive effect (b = .15, p < .001) were also significant. The positive indirect effect of ES on OCB through FWE (b = .04 CI [.008, .072]) was also significant, as the 95% percentile bootstrap confidence interval did not include zero. Thus, Hypothesis 3 was supported.

Finally, the negative direct effect of ES on ITL controlling for FWE (b = −.36, p < .001) was significant and the negative total effect (b = −.53, p < .001) were significant. The negative indirect effect of ES on ITL through FWE (b = −.16 CI [−.248, −.085]) was also significant, as the 95% percentile bootstrap confidence interval did not include zero. Thus, Hypothesis 4 was also confirmed.

Discussion

This article makes several important contributions to the management literature and also offers suggestions for future research. First, it creates a scale to measure ebullient supervision that is different from any other scale in assessing the extent to which a supervisor makes work fun. Since the research demonstrates that fun work environments have positive outcomes for both employees and organizations, it is important to develop a measure that can investigate how the actions and behaviors of a supervisor can help create a fun work environment. We contend that the concept of ebullient supervision fills this gap in the literature. Second, this article provides evidence that the construct of ebullient supervision as measured by this new scale does in fact substantially impact a number of important dependent variables that are beneficial to the individual and organization. Third, it shows that an ebullient supervisor affects these beneficial variables through employee perceptions as to what constitutes a fun work environment. Fourth, our findings affirm that the supervisor, while influential in determining employee behavior, is only part of the total work environment and that other mechanisms beyond an ebullient supervisor’s effects on a fun work environment may influence employee and organizational outcomes. Finally, this article reinforces the value of conducting additional research to further investigate the beneficial effects of a fun work environment spurred on by a supervisor’s behaviors and actions.

The support for our hypotheses shows that our new construct of ebullient supervision is strongly correlated to a number of variables that are frequently researched in management studies. Our data show ebullient supervision is strongly associated with job satisfaction (r = .44), intention to leave (r = −0.33), work engagement (r = .44), and organizational citizenship behaviors (r = .24). These four dependent variables have been shown in prior research to be important outcomes associated with efficient and effective management of organizations and their supervisors.

Our findings also show the importance that the indirect effects of ebullient supervision have on these important four dependent variables as they work through a fun work environment. We find that, depending on the dependent variable, the indirect effect accounts for anywhere from 27% (for OCB) to 37% (for job satisfaction) of the total effect of ebullient supervision on the dependent variables. These findings mean (depending on the dependent variable) between 63% and 73% of the effect of ebullient supervision on the dependent variables is not explained through the mechanism of fun work environment. This supports the claim that although ES captures the role of the supervisor in the creation of a fun work environment, that other mechanisms are at play that affect the influence of ES on beneficial employee and organizational outcomes.

Finally, this article illuminates the “positive side” of leading. Our findings show that leadership behaviors that lead to fun work environments affect important attitudinal and behavioral outcomes. These findings show how positive supervision can have beneficial effects for individuals and the organization. These findings have important implications for theory and practice as they provide new emphasis on researching the benefits of supervisory practices and behaviors that promote employees having fun at work.

Limitations and Future Research

Like all research, our study has a number of limitations. First, our samples are convenience samples. Thus, there is naturally some question regarding the generalizability of the findings. However, we believe that the breadth and diversity of experiences of our samples (particularly the MTurk samples) help mitigate this concern. Nevertheless, future research should use samples that are more specifically focused on different industrial classifications and demographic categories that would both refine and extend our results.

Second, our variables are all measured by self-reports and thus, may suffer from self-report biases. Although, we believe that our measures are accurate, the possibility of spurious correlations exists. However, our exploratory factor analysis (EFA) and CFA findings show distinct factors indicating clear differences among the variables, indicating that self-report bias is not overly influencing the relationships among the variables. Future research should use third parties to determine the dependent variables such as OCBs, to help mitigate this possible bias.

Third, although we hypothesized that ebullient supervision predicts fun work environment and subsequently important dependent variables, our analysis cannot show causality due to its cross-sectional design. Although it is theoretically unlikely that the dependent variables, such as OCBs cause ebullient supervision, the possibility exists. Future research can test causality of the ES scale in work-related outcomes by conducting longitudinal studies. Furthermore, an experimental design would help show the internal validity of these effects as we hypothesize them.

Finally, Plester (2009) found that organizational culture had an effect on fun in the workplace. Further research is needed on how organizational culture affects or is affected by ES. Moreover, the FWE scale used in this study (eight items) needs to be further tested. The two items excluded during the EFA were used by Tews et al. (2013) study, where they kept 5 items from the original 10 item scale developed by Ford et al. (2003). Our studies did not confirm the full scale nor did they confirm Tews et al. (2013) scale for a fun work environment. Future studies analyzing fun work environment should reevaluate and refine this scale.

In addition to the research suggestions provided above, the purpose of this article is to define the concept of an ebullient supervisor and develop and validate a scale to measure the construct of ES. The scale that was validated in the article provides researchers with a tool to further investigate the construct of ebullient supervision, across both goods producing industries and service industries. We used fun work environment, job satisfaction, work engagement, organizational citizenship behavior, and intention to leave to study some of the well-established consequences of ES. Future studies should focus on identifying additional consequences of ES, both positive and negative. If fun is an important need as Glasser (1998) argues and an ebullient supervisor is an important conduit to achieving valuable resources as Fluegge-Woolf (2014) suggests, then further work is needed to better understand what supervisor’s deeds and behaviors can help create a fun work environment for employees. Everyone has either seen or worked in organizations where employees are clearly having fun doing their jobs and saw confirmation of the research showing a positive relationship between fun at work and employee and organizational outcomes. Indeed, because of the dramatic growth in service sector employment where employee attitudes and behaviors are often the critical determinant of customer satisfaction, it is becoming increasingly important to understand what supervisory actions, behaviors and attitudes help create fun work environments. If, as the service profit chain suggests, customers who repatronize any organization and also recommends it to others do so in part because of their satisfaction with positive employee encounters, then all organizations need to better understand how they can identify and develop their ebullient supervisors in order to create a fun work environment.

Footnotes

Appendix

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Robert C. Ford is professor emeritus of management at the University of Central Florida. His research interests are focused on the management of service and hospitality organizations.

Renata Guzzo is a PhD candidate in the Conrad N. Hilton College of Hotel and Restaurant Management at University of Houston. Her research interests include organizational behavior and human resources, corporate social responsibility, and lodging management and operations.

JeAnna L. Abbott holds the Spec’s Charitable Foundation Endowed Professorship in Social Responsibility at the University of Houston. She holds a joint appointment with the Conrad N. Hilton College and the C.T. Bauer College of Business. Her research interests include positive organizational culture, positive leadership, likability, compassion, and negotiation in the workplace. She also has expertise in big data analytics and applied statistics.

John T. Bowen is a professor in the Conrad N. Hilton College at the University of Houston. His research interests include customer loyalty, employee and customer involvement in the service delivery process and generational differences in consumer behavior.

References

Abu-Bader

(2010). Advanced and multivariate statistical methods for social science research. Oxford, England: Oxford University Press.

Andrzejewski

S. A.

Mooney

E. C.

(2016). Service with a smile: Does the type of smile matter? Journal of Retailing and Consumer Services, 29, 135-141.

Barger

P. B.

Behrend

T. S.

Sharek

D. J.

Sinar

E. F.

(2011). I-O and the crowd: Frequently asked questions about using Mechanical Turk for research. TIP, 11.

Barger

P. B.

Grandey

A. A.

(2006). Service with a smile and encounter satisfaction: Emotional contagion and appraisal mechanisms. Academy of Management Journal, 49, 1229-1238.

Becker

F. W.

Tews

M. J.

(2016). Fun activities at work: Do they matter to hospitality employees? Journal of Human Resources in Hospitality & Tourism, 15, 279-296.

Brinker

Phillips

D. T.

(1999). On the brink: The life and leadership of Norman Brinker. Arlington, TX: Summit.

Buhrmester

Kwang

Gosling

S. D.

(2011). Amazon’s Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6(1), 3-5.

Cammann

Fichman

Jenkins

G. D.

Jr. Klesh

J. R.

(1983). Assessing the attitudes and perceptions of organizational members. In Seashore

S. E.

Lawler III

E. E.

Mirvis

P. H.

Cammann

(Eds.), Assessing organizational change: A guide to methods, measures, and practices (pp. 71-138). New York, NY: Wiley.

Choi

Kwon

Kim

(2013). Effects of attitudes vs experience of workplace fun on employee behaviors: Focused on Generation Y in the hospitality industry. International Journal of Contemporary Hospitality Management, 25, 410-427.

10.

Collison

Cohen

Ford

R. C.

Newstrom

McLaughlin

F. S.

(2002). SHRM Fun Work Environment Survey. Alexandria, VA: SHRM Research.

11.

Disney

(1994). Walt Disney famous quotes (Disney’s Kingdom ed.). Lake Buena Vista, FL: Walt Disney.

12.

Eisenberger

Huntington

Hutchison

Sowa

(1986). Perceived organizational support. Journal of Applied Psychology, 71, 500-507.

13.

Eisenberger

Stinglhamber

(2011). Perceived organizational support: Fostering enthusiastic and productive employees. Washington, DC: American Psychological.

14.

Fluegge-Woolf

E. R.

(2014). Play hard, work hard: Fun at work and job performance. Management Research Review, 37, 682-705.

15.

Ford

R. C.

McLaughlin

F. S.

Newstrom

J. W.

(2003). Questions and answers about fun at work. Human Resource Planning, 26(4), 18-33.

16.

Ford

R. C.

McLaughlin

F. S.

Newstrom

J. W.

(2005). Creating and sustaining fun work environments in hospitality and service organizations. Journal of Human Resources in Hospitality & Tourism, 4(1), 11-30.

17.

Ford

R. C.

Newstrom

J. W.

McLaughlin

F. S.

(2004). Making workplace fun more functional. Industrial and Commercial Training, 36(3), 117-120.

18.

Fornell

Larcker

D. F.

(1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 48(1), 39-50.

19.

Frieberg

(1998). Nuts! Arlington, TX: Bard Press.

20.

Glasser

(1994). The control theory manager. New York, NY: Harper Business.

21.

Glasser

(1998). Choice theory. A new psychology of personal freedom. New York, NY: Harper Collins.

22.

Gosling

S. D.

Vazire

Srivastava

John

O. P.

(2004). Should we trust web-based studies? A comparative analysis of six preconceptions about internet questionnaires. American Psychologist, 59, 93-104.

23.

Hair

J. F.

Black

W. C.

Babin

B. J.

Anderson

R. C.

(2010). Multivariate data analysis. (7th ed.). Upper Saddle River, NJ: Pearson Prentice Hall.

24.

Han

Kim

Jeong

(2016). Workplace fun for better team performance: Focus on frontline hotel employees. International Journal of Contemporary Hospitality Management, 28, 1391-1416.

25.

Hennig-Thurau

Groth

Paul

Gremler

D. D.

(2006). Are all smiles created equal? How emotional contagion and emotional labor affect service relationships. Journal of Marketing, 70(3), 58-73.

26.

Hinkin

T. R.

(1995). A review of scale development practices in the study of organizations. Journal of Management, 21, 967-988.

27.

Hobfoll

S. R.

(2001). The influence of culture, community, and the nested-self in the stress process: Advancing conservation of resources theory. Applied Psychology, 50, 337-421.

28.

Jackson

D. L.

Gillaspy

J. A.

Jr. Purc-Stephenson

(2009). Reporting practices in confirmatory factor analysis: An overview and some recommendations. Psychological Methods, 14(1), 6-23.

29.

Karl

Peluchette

(2006). How does workplace fun impact employee perceptions of customer service quality? Journal of Leadership & Organizational Studies, 13(2), 2-13.

30.

Karl

K. A.

Peluchette

J. V.

Hall

L. M.

(2008). Give them something to smile about: A marketing strategy for recruiting and retaining volunteers. Journal of Nonprofit & Public Sector Marketing, 20(1), 71-96.

31.

Karl

Peluchette

Hall-Indiana

Harland

(2005). Attitudes toward workplace fun: A three sector comparison. Journal of Leadership & Organizational Studies, 12(2), 1-17.

32.

Karl

K. A.

Peluchette

J. V.

Harland

(2007). Is fun for everyone? Personality differences in healthcare providers’ attitudes toward fun. Journal of Health and Human Services Administration, 29, 409-447.

33.

Kouzes

Posner

B. Z.

(1995). The leadership challenge, San Francisco, CA: Jossey-Bass.

34.

Kurtessis

J. N.

Eisenberger

Ford

M. T.

Buffardi

L. C.

Stewart

K. A.

Adis

C. S.

(2017). Perceived organizational support a meta-analytic evaluation of organizational support theory. Journal of Management, 43, 1854-1884.

35.

Lamm

Meeks

M. D.

(2009). Workplace fun: The moderating effects of generational differences. Employee Relations, 31, 613-631.

36.

Line

N. D.

Hanks

Zhang

(2016). Sustainability communication: The effect of message construals on consumers’ attitudes towards green restaurants. International Journal of Hospitality Management, 57, 143-151.

37.

Luthans

(2002a). Positive organizational behavior: Developing and managing psychological strengths. Academy of Manage-ment Executive, 16(1), 57-72.

38.

Luthans

(2002b). The need for and meaning of positive organizational behavior. Journal of Organizational Behavior, 23, 695-706.

39.

Mäkikangas

Kinnunen

Feldt

Schaufeli

(2016). The longitudinal development of employee well-being: A systematic review. Work & Stress, 30(1), 46-70.

40.

Marsh

H. W.

Hau

K.-T.

Wen

(2004). In search of golden rules: Comment on hypothesis-testing approaches to setting cutoff values for fit indexes and dangers in overgeneralizing Hu and Bentler’s (1999) findings. Structural Equation Modeling, 11, 320-341.

41.

Martinko

M. J.

Harvey

Brees

J. R.

Mackey

(2013). A review of abusive supervision research. Journal of Organizational Behavior, 34(Suppl. 1), S120-S137.

42.

Mason

Suri

(2012). Conducting behavioral research on Amazon’s Mechanical Turk. Behavior Research Methods, 44(1), 1-23.

43.

McDowell

(2004). Fun at work: Scale development, confirmatory factor analysis, and links to organizational outcomes. Dissertation Abstracts International. Available from ProQuest Dissertations and Theses database. (UMI No. Vol. 3155834)

44.

McGraw

K. O.

Tew

M. D.

Williams

J. E.

(2000). The integrity of web-delivered experiments: Can you trust the data? Psychological Science, 11, 502-506.

45.

Meyer

J. P.

Allen

N. J.

Smith

C. A.

(1993). Commitment to organizations and occupations: Extension and test of a three-component conceptualization. Journal of Applied Psychology, 78, 538-551.

46.

Owler

Morrison

Plester

(2010). Does fun work? The complexity of promoting fun at work. Journal of Management & Organization, 16, 338-352.

47.

Paterson

T. A.

Luthans

Jeung

(2014). Thriving at work: Impact of psychological capital and supervisor support. Journal of Organizational Behavior, 35, 434-446.

48.

Peluchette

Karl

K. A.

(2005). Attitudes toward incorporating fun into the health care workplace. Health Care Manager, 24, 268-275.

49.

Perrin

(1998). A serious business. Accountancy, 121(1254), 40-42.

50.

Petelczyc

C. A.

Capezio

Wang

Restubog

Aquino

(2018). Play at work: An integrative review and agenda for future research. Journal of Management, 44, 161-190.

51.

Plester

(2009). Crossing the line: Boundaries of workplace humour and fun. Employee Relations, 31, 584-599.

52.

Plester

Cooper-Thomas

Winquist

(2015). The fun paradox. Employee Relations, 37, 380-398.

53.

Plester

Hutchison

(2016). Fun times: The relationship between fun and workplace engagement. Employee Relations, 38, 332-350.

54.

Podsakoff

P. M.

MacKenzie

S. B.

Podsakoff

N. P.

(2016). Recommendations for creating better concept definitions in the organizational, behavioral, and social sciences. Organizational Research Methods, 19, 159-203.

55.

Pryor

M. G.

Singleton

L. P.

Taneja

Humphreys

J. H.

(2010). Workplace fun and its correlates: A conceptual inquiry. International Journal of Management, 27, 294-302.

56.

Roberts

D. R.

Davenport

T. O.

(2002). Job engagement: Why it’s important and how to improve it. Employment Relations Today, 29(3), 21-29.

57.

Schaufeli

W. B.

Bakker

A. B.

Salanova

(2006). The measurement of work engagement with a short questionnaire: A cross-national study. Educational and Psychological Measurement, 66, 701-716.

58.

Seppälä

Mauno

Feldt

Hakanen

Kinnunen

Tolvanen

Schaufeli

(2009). The construct validity of the Utrecht Work Engagement Scale: Multisample and longitudinal evidence. Journal of Happiness Studies, 10, 459-481.

59.

Söderlund

Rosengren

(2010). The happy versus unhappy service worker in the service encounter: Assessing the impact on customer satisfaction. Journal of Retailing and Consumer Services, 17, 161-169.

60.

Spector

P. E.

Bauer

J. A.

Fox

(2010). Measurement artifacts in the assessment of counterproductive work behavior and organizational citizenship behavior: Do we know what we think we know? Journal of Applied Psychology, 95, 781-790.

61.

Spreitzer

Sutcliffe

Dutton

Sonenshein

Grant

A. M.

(2005). A socially embedded model of thriving at work. Organization Science, 16, 537-549.

62.

Tepper

B. J.

(2000). Consequences of abusive supervision. Academy of Management Journal, 43, 178-190.

63.

Tepper

B. J.

(2007). Abusive supervision in work organizations: Review, synthesis, and research agenda. Journal of Management, 33, 261-289.

64.

Tews

M. J.

Michel

J. W.

Allen

D. G.

(2014). Fun and friends: The impact of workplace fun and constituent attachment on turnover in a hospitality context. Human Relations, 67, 923-946.

65.

Tews

M. J.

Michel

J. W.

Bartlett

(2012). The fundamental role of workplace fun in applicant attraction. Journal of Leadership & Organizational Studies, 19, 105-114.

66.

Tews

M. J.

Michel

J. W.

Stafford

(2013). Does fun pay? The impact of workplace fun on employee turnover and performance. Cornell Hospitality Quarterly, 54, 370-382.

67.

Tremblay

Gibson

(2016). The role of humor in the relationship between transactional leadership behavior, perceived supervisor support, and citizenship behavior. Journal of Leadership & Organizational Studies, 23, 39-54.

68.

Weiss

H. M.

Cropanzano

(1996). Affective events theory: A theoretical discussion of the structure, causes and consequences of affective experiences at work. Research in Organizational Behavior, 18, 1-74.

69.

Zellars

K. L.

Tepper

B. J.

Duffy

M. K.

(2002). Abusive supervision and subordinates’ organizational citizenship behavior. Journal of Applied Psychology, 87, 1068-1076.