The Measurement of Employee Engagement

Although researchers do not agree on what employee engagement is, they have moved on to the measuring of employee engagement. Unfortunately, this process is also not without debate, and further clarity is needed on what is being measured to move the engagement construct forward (Masson, Royal, Agnew, & Fine, 2008). Additionally, although Macey & Schneider (2008) may have attempted to provide definitional clarity with their separation of engagement into three pieces, they may have inadvertently caused measurement confusion (Saks, 2008). Although Macey & Schneider (2008) refer to engagement as a state, it has been largely measured as a trait (Dalal et. al 2008). It is impossible to come to a conclusion on exactly how many measurement systems exist for engagement as the amount of vendors creating them and redesigning them is in constant flux and overall on the rise. In addition, many measurement models are considered proprietary and not published (Wefald & Downey, 2009). However, there are a number of commonly used forms that have become ingrained in the engagement literature. Next, we review some of these more established methods of measurement.

One of the most prominent measures of employee engagement, namely by academics, is the Utrecht Work Engagement Scale (Schaufeli & Bakker, 2003). The seventeen-item scale measures engagement divided into the categories of vigor (six questions), dedication (five questions), and absorption (six questions). These responses are rated on a seven-point frequency rating scale ranging from six (always) to zero (never), with higher scores indicating higher vigor, dedication or absorption. The scale has been used in numerous engagement studies (Hallberg & Schaufeli, 2006; Salanova, Agut, & Peiro, 2005; Schaufeli & Bakker, 2004; Siu et. al., 2010; Soane et. al, 2013; van Beek et. al, 2012; Ni, 2013).

Used in Schaufeli & Bakker’s (2004) definition, vigor refers to high levels of persistence, energy, and mental resilience while working, and the willingness to invest effort in one’s work. Dedication refers to being strongly involved in one’s work, and experiencing a sense of significance, enthusiasm, inspiration, pride, and challenge. Finally, absorption refers to being fully concentrated and deeply engrossed in one’s work, whereby time passes quickly and one has difficulties with detaching oneself from work. Shortened versions of the scale have also been used.

A promising measure is one drafted by Rich, Lepine, & Crawford (2010). Their job engagement scale attempts to return to Kahn’s (1990) definition, which draws on work from Brown & Leigh’s (1996) measure of work intensity, Russel & Barrett’s (1999) research on core affect, and Rothbard’s (2001) measure of engagement. By modifying these scales the authors were able to construct a measure that more properly reflected Kahn’s (1990) conceptualization of engagement developing from physical, cognitive, and emotional energy. Researchers also have utilized the Motivation and Engagement Scale developed by Martin, which comes in different forms dependent upon the interviewees (e.g., MES-W; Martin, 2006; Martin, 2007). Direct questions (Lee, 2012) and individually crafted and adapted scales have also been utilized (Saks, 2006; James, McKechnie, & Swanberg, 2011) to measure engagement in varying forms.

Lastly, there are the numerous vendor created measurement scales that are proprietary information. According to Attridge (2009), “some of these consulting organizations include Blessing White, Gallup, Hewitt, Sirota, Towers Watson, Valtera, and Watson Wyatt Worldwide.” However, a couple are predominantly mentioned, namely the Gallup Workplace Audit, also known as the GWA or Q12. The first version of the GWA first appeared in the 1990s to judge workplace attitudes. In 2003 the Q12 had been validated by 285,314 workgroups “drawn from teams and industries from around the world” (Gopal, 2003). By 2009, a published report claimed the Q12 had been administered to “more than 15 million employees in 169 different countries and 65 languages” (Harter et. al, 2009). That year marked the seventh iteration of the Gallup meta-analysis (Harter et. al, 2009).

As noted earlier, one can measure engagement using a burnout scale and there has been some empirical support to tie burnout to outcomes (e.g., Taris, 2006). Understanding that burnout is the opposite of engagement, it should stand to reason that the results can simply be reflected to demonstrate the positive rather than the negative. This conclusion is not wholeheartedly supported. Schaufeli, Salanova, Gonzalez-Roma, & Bakker (2001) and Schaufeli, Taris, & Rhenen (2008) confirm that burnout and engagement, to a large extent, are antipodes, acting as each other’s opposites. However, Demerouti, Mostert, & Bakker (2010) feel that each dimension should be addressed separately, rather than collectively, as the energy dimensions appear to represent highly related constructs.

Determinants of Employee Engagement

In order to move employee engagement beyond an intellectual exercise in measurement and definitions, most organizations have conducted some type of research to determine what predicts engagement. By understanding the things that drive engagement, the assumption is that organizations can then create interventions, raise engagement scores and then improve firm performance. In this next section, we focus on what researchers and consultants have suggested are the determinants of employee engagement. Although some of the work is empirical, most is not.

There are numerous models and suggestions of what determines high engagement, but it is difficult to make conclusions for a few reasons. First, the definitions of engagement vary from study to study; thus, it is almost impossible to generalize across the research that has been done. Second, most of this work is not longitudinal in nature, making the relationship between the predictors and engagement a circular exercise in logic (Bakker & Demerouti, 2008). Essentially, it is not possible to know if engagement causes the predictors or if the predictors cause engagement. Third, the models of predictors require, in many cases, leaps of faith. Many are not based on detailed theory or research. Table 3 provides a sampling of models and research that focus on the determinants of employee engagement:

Predictors of Employee Engagement

Author(s), Year (sorted in chronological order)

Family roles

Rothbard (2001)

Day-level recovery

Sonnentag (2003)

Connection between work and organizational strategy,

Importance of job to organizational success,

Understanding of how to complete work projects

Corporate Leadership Council (2004)

Confident can achieve career objectives, Sense of

personal accomplishment, Confident organization will

be successful, Quality is a high priority

Mercer (2007)

Supervisor support, Positive appreciation, Collaborative

organizational climate, Innovative problem solving

Bakker, Hakanen, Demerouti, & Xanthopoulou (2007)

Discipline of free-loaders, Accountability, Fairness

Wagner & Harter (2007)

Reduced role conflict, Proper training, Personal

autonomy, Effective utilization of expert, referent and

exchange power by managers

Cozzani & Oakley (2007)

Coping and buoyancy

Parker & Martin (2009)

Effective and caring leadership, Appealing development

opportunities, Interesting work, Fulfilling tangible and

intangible rewards

Towers Watson (2009)

Professional status, Interaction, Thinking of quitting

Simpson (2009)

Communication and knowledge sharing, Opportunity to

provide input and exercise independent action,

Opportunity for growth and learning

Towers Watson (2010)

Competence, Meaning, Impact, Self-determination

Stander & Rothman (2010)

Coworker cohesion, Supervisor support, Autonomy,

Work pressure (negative), Control (negative)

Kumar & Kumar Sia (2012)

Job role, Rewards & recognition, Leadership & planning

Gujral & Jain (2013)

Hire right people, Develop employee strengths,

Enhance employee well-being, Company size, Amount

of work done off-site, Age, Gender

Gallup Consulting (2013)


As can be seen from Table 3, similar to the definitions of engagement, the antecedents vary just as widely. After reviewing all of the antecedents to engagement, the definitions, and the many types of instruments used to collect this information, we conclude that there is a strikingly obvious fact – that engagement is often a term configured to fit the needs of a particular study and not a concrete item with obviously discernible qualities.

The "So What" Question: Outcomes of Employee Engagement

Many could argue that the most important point of investigating, defining, and measuring engagement is to understand the outcomes of increasing or decreasing it amongst individuals or groups and then using that knowledge to create interventions to improve engagement. This is where the research has moved. There are a number of notable outcome studies, though predominantly in the consultant literature. For example, Towers Watson (2012) analyzed fifty global companies using engagement (with their newly defined term of sustainable engagement). Their working definition is “the intensity of employees’ connection to their organization, based on three elements” centered around “the extent of employees’ discretionary effort committed to achieving work goals (being engaged), an environment that supports productivity in multiple ways (being enabled), and a work experience that promotes well-being (feeling energized).”

They found that companies with low traditional engagement had an average operating margin just under 10%, while those with high engagement were around 14%, and those with high sustainable engagement had an operating margin just over 27%. They also found that highly engaged employees had lower absenteeism (lost productivity at work, which computes to 7.6 days per year for highly engaged compared to 14.1 for the disengaged) and absenteeism (3.2 days for the highly engaged compared to 4.2 for the disengaged), along with lower intention to leave. These data are very interesting; however, the studies are not causal in nature (not looking at changes in performance related to changes in engagement). Thus, it may be that higher performing firms happen to have higher engagement scores; the causal linkages are not yet made.

In an earlier Towers Watson study (2003), they used data collected from over 360,000 employees from forty-one companies throughout the world’s ten largest economies. They found that, over a three-year period, companies with low engagement demonstrated a -2.01 percent operating margin compared to a 3.74 percent operating margin for those with high engagement. Similarly, net profit margin showed results of -1.38 percent and 2.06 percent respectively. Gallup’s meta-analysis in 2010 of more than 152 organizations showed that companies with world-class engagement had 3.9 times the EPS growth rate than companies without (Gallup Consulting, 2010). Again, it's not clear to the reader if these studies are causal or correlational in nature.

In the “State of the American Workplace: Employee Engagement Insights for U.S. Business Leaders” report done by Gallup (2013), Gallup reemphasized its findings from its research done since 2000. In their research for the years 2010- 2012, the group came to the conclusion that engagement showed positive and relevant findings through meta-analysis in nine performance outcomes: customer ratings, profitability, productivity, turnover, safety incidents, shrinkage (theft), absenteeism, patient safety incidences, and quality (defects). In addition, the study found that “teams in the top 25% versus the bottom 25% incur far less in healthcare costs” (Gallup, 2013). Naturally, this research was performed using Gallup’s Q12 using 263 research studies across 192 organizations in 49 industries and 34 countries (Gallup, 2013).

Departing from some of the research done by consultants, Sonnentag (2003) found that individuals who received ample day-level recovery time were more likely to experience a high level of work engagement during the subsequent day. That level of work engagement ultimately led to those individuals taking more initiative and further pursuing their learning goals. Medlin & Faulk (2011) found that engagement predicted optimism amongst university students, which in turn predicted performance and turnover intention, as well as perceived individual performance. Soane et. al (2013) found that meaningful work led to higher levels of engagement, further strengthened by well-being, which transferred to lower levels of absenteeism. Rich, LePine, & Crawford (2010) surveyed firefighters to discover that engagement created customer helpfulness/courtesy, as well as a stronger dedication to being involved in organizational matters.

Table 4 on the next page summarizes some of the reported outcomes of increasing engagement among eighteen studies (assume all beneficial, i.e. absenteeism = lower absenteeism amongst engaged employees). As shown in Table 4, the outcomes associated with employee engagement are numerous and varied. This is because: 1) the various different measurement methods used and 2) the numerous definitions that are used to create those measurement systems. There likely is significant appeal to the term because it theoretically addresses the problems of motivation and performance (Little & Little, 2006). At the same time, the confusion in terms and studies has made many people skeptical about the true value of the engagement concept and corresponding research largely due to these two reasons as well as a lack of designs that test causality (Bakker & Demerouti, 2008; Taris, 2006).

Outcomes of Engagement

Author(s), Year (Sorted Chronological Order)

Customer satisfaction, Productivity, Profit,

Employee retention, Employee safety

Harter, Schmidt, & Hayes (2002)

Operating Margin, Net Profit Margin

Towers Watson (2003)

Pursuit of learning, Personal initiative

Sonnentag (2003)

Lower Turnover

Corporate Leadership Council (2004)

Work life benefits physical/mental health and

stress level at home

Crabtree (2005)

Safe patient care

Laschinger & Leiter (2006)

Job satisfaction, Organizational commitment,

Intention to quit, Organizational citizenship


Saks (2006)

Customer loyalty, Profitability, Productivity,

Turnover, Safety incidents, Absenteeism,

Shrinkage, Patient safety outcomes, Quality


Harter et. al (2009)

Absenteeism, Turnover, Shrinkage, Safety

incidents, Patient safety incidents, Quality

(defects), Productivity, Profitability, Earnings

per share (EPS), Hope, Diagnosis of

depression, Diagnosis of anxiety, Involvement

in wellness programs

Gallup, State of the American Workplace

Report (2010)

Customer helpfulness/courtesy, Involved in

organizational matters

Rich, LePine, & Crawford (2010)

Burnout, Workaholics

Van Beek, Taris, & Schaufeli (2011)

Optimism, Turnover intention, Perceived

individual performance

Medlin & Faulk (2011)

Lower absenteeism

Soane et. al (2013)

Operating Margin, Lower absenteeism (lost

productivity), Lower absenteeism, Lower

intention to quit

Towers Watson (2012)

Lower Turnover, Fewer sick days/missed days,

Higher company commitment, Less poor

behaviors in non-work roles

Gallup Consulting (2013)

Organizational citizenship behavior, Employee

commitment, Employee satisfaction

Gujral & Jain (2013)

Job performance

Ali, Hussain & Azim (2013)

Service climate

Ellinger, Musgrove, & Ellinger (2013)


Some of the more passionate conversations and case studies to support engagement come from service-focused companies. The link between engagement and customer satisfaction (or patient satisfaction in health care) is supported by the companies doing this work. Also, these within-company case studies appear to have higher quality research behind them (causal research, inclusion of control variables in the data analysis). The studies that link engagement with firm performance are weaker in that they are often not causal but correlational, and they do not include adequate control variables in the analysis. Given the broad range of findings, numerous types of analysis conducted, lack of common definitions, excitement about engagement linking to outcomes and ongoing research, we will take a step to propose the following:

There is not enough high quality data to confirm that engagement alone drives high organizational performance, but we are suggesting there is enough data to show that engagement is better than disengagement. We will propose in the next sections of the paper how to move from the baseline of engagement as the first step to high performance to linking engagement to roles as a way to move those engagement infinitives forward.