One word covers two different measurements. In research, employee engagement means people investing themselves physically, mentally and emotionally in their work.¹ Later researchers describe it as a fairly lasting state, often measured as vigor, dedication and absorption.² It overlaps with job satisfaction, but a 2011 meta-analysis found the two are not the same thing.³

Gallup, the survey and consulting company, describes its Q12 survey as measuring something else: the working conditions that contribute to that state.⁴ An engagement score means different things depending on which of the two it captures.

Two answers to what is employee engagement

Academic researchers, from William Kahn in 1990 to Wilmar Schaufeli’s group in the 2000s, define employee engagement as a person’s state of energy and involvement in the work.¹² Gallup defines it as involvement and enthusiasm, but measures it through questions about working conditions.⁵⁴

The academic line starts with a 1990 paper in the Academy of Management Journal by William Kahn, then at Boston University. He described personal engagement as people bringing their selves into their work roles, expressing themselves physically, cognitively and emotionally while doing the job, and disengagement as withdrawing and defending the self instead. His evidence was qualitative: interviews and observation of 16 counselors at a summer camp and 16 members of an architecture firm.¹

Kahn was interested in moments: he described people bringing more or less of themselves to a task as the day ebbs and flows.¹ Schaufeli and colleagues, by contrast, describe engagement as a persistent, pervasive state rather than a momentary one.²

Kahn found three psychological conditions behind engagement: meaningfulness, a sense that investing yourself pays a return; safety, feeling able to show yourself without fear for your self-image, status or career; and availability, having the physical, emotional or psychological resources to engage at that moment.¹ The safety condition shares its name with the team-level idea described in why psychological safety matters for teams.

Gallup’s measure points somewhere else. Gallup, which says its Q12 survey has been given to nearly 64 million employees since 1998, describes the Q12 as a measure of “engagement conditions”: perceptions of role clarity, resources, feedback and feeling appreciated, which it treats as contributors to engagement.⁴ Gallup adds that the Q12 total correlates highly with direct measures of work engagement.⁴ A 2011 meta-analysis by Michael Christian and colleagues set two tests for what counts as an engagement measure: it must refer to the work actually performed, and to a physical, emotional or cognitive investment in that work. By that standard, the authors judged that an earlier Gallup measure did not fit, because its items refer to work conditions rather than the work task.³

Vigor, dedication and absorption: how researchers measure the state

The Utrecht Work Engagement Scale (UWES), developed by Wilmar Schaufeli, Arnold Bakker and colleagues, defines work engagement as a positive, fulfilling, work-related state of mind with three parts: vigor, dedication and absorption. Its nine-item short version was tested on 14,521 people in 10 countries in a 2006 study.²

In the authors’ descriptions, vigor is high energy and mental resilience while working, a willingness to put effort in and persistence when things go wrong. Dedication is strong involvement in the work, with a sense of significance, enthusiasm, inspiration, pride and challenge. Absorption is being fully concentrated and happily engrossed, so that time passes quickly and it is hard to detach from the work.²

People rate how often they feel each item, from never to every day, on a 0-6 scale. The three scores are hard to separate in practice: in the 2006 data the latent factors correlated at a median above .90 across countries, and the authors suggested researchers might, for now, use one total score. Engagement also proved fairly stable in retests in Australia and Norway, with the first measurement explaining roughly a third to a half of the variation a year later, which the authors read as a chronic rather than a passing state.²

The UWES authors see engagement as the positive opposite of burnout, with vigor facing exhaustion and dedication facing cynicism.² That view is contested, as the section on critics below shows.

How Gallup defines and counts engaged employees

Gallup defines employee engagement as the involvement and enthusiasm of employees in their work and workplace, and sorts people into engaged, not engaged and actively disengaged by applying a proprietary formula to their answers on 12 Q12 statements, in its State of the Global Workplace 2026 report.⁵

The statements ask about the conditions of the job: whether people know what is expected of them, have the materials they need, get recognition, feel someone at work cares about them and have chances to learn and grow. Gallup’s standard survey opens with an overall satisfaction question before the 12 statements, and the company restricts reproduction of the items.⁴ In Gallup’s categories, engaged employees are highly involved in and enthusiastic about their work, not engaged employees are psychologically unattached and give time but not energy, and actively disengaged employees are resentful that their needs are not being met.⁵

In 2025, 20 percent of employees worldwide were engaged by this measure, down from a peak of 23 percent in 2022; 64 percent were not engaged and 16 percent actively disengaged. The figures come from Gallup’s World Poll of nationally representative samples of employed adults, are self-reported and have not been peer-reviewed. Regional figures, based on three years of polling, ranged from 31 percent in the United States and Canada to 12 percent in Europe.⁵

Myth

Any survey's percent-engaged figure can be compared with Gallup's global figure.

Fact

Gallup's figure comes from its own proprietary scoring of the Q12, which it calls a higher bar than satisfaction or percent-favorable indexes, so scores from different surveys are not on the same scale.

The four definitions below differ in what they count, which is why scores built on them are not interchangeable.

Is engagement just job satisfaction under a new name?

Engagement overlaps with older ideas such as job satisfaction and organizational commitment, and critics have asked since at least 2008 whether it adds anything.⁶ A 2011 meta-analysis of 91 studies concluded that engagement is related to those job attitudes but distinct from them.³

The sharpest statement of the problem came from William Macey and Benjamin Schneider in 2008. They wrote that the term had been heavily marketed by HR consulting firms, that people used it for psychological states, personality traits and behaviors alike, and that it risked the “jangle fallacy”: giving a new name to something already known. They also argued that the Gallup measure describes the conditions people work under, so the state of engagement itself is not assessed. Both authors listed Valtera Corporation as their affiliation, and Schneider also the University of Maryland.⁶ Valtera was a consulting firm that ran employee surveys, including engagement surveys, for client organizations, according to the author notes of a 2009 book they co-wrote.⁷

1. Conditions at work: clear expectations, the right resources, feedback, feeling appreciated; what Gallup says its Q12 measures
2. The state itself: vigor, dedication and absorption; what academic scales such as the UWES measure
3. Behavior: discretionary effort, including effort beyond the formal role; what Macey and Schneider call behavioral engagement

In a commentary in the same journal issue, Daniel Newman and David Harrison went further, arguing that engagement comprises no new conceptual content, only a blend of old content, and that the usefulness of engagement as a state depends on evidence that it can be told apart from overall job attitude.⁸ The key study below tested that question.

The study

Moderate evidence

Testing the old-wine charge: Christian, Garza and Slaughter, 2011

Michael Christian, Adela Garza and Jerel Slaughter pooled 91 studies of work engagement. Engagement correlated .53 with job satisfaction, .59 with organizational commitment and .52 with job involvement: substantial overlap, but none close to 1.0, which the authors read as evidence that engagement is a separate construct. Engagement was also linked to task performance (a correlation of .43) and, in a regression, explained variance in task performance beyond the three job attitudes combined.³

The authors concluded that engagement is more than a blend of old wines, but the caveats are theirs too: most studies were cross-sectional, which rules out firm conclusions about cause, and most used self-reports, which can inflate correlations. For performance rated by someone else, the link was .39, similar in size to published estimates for job satisfaction.³

Burnout is the unsettled boundary. A 2012 meta-analysis of 50 samples from 37 studies, led by Michael Cole, found high correlations between the dimensions of burnout and of engagement, similar patterns with other variables, and engagement effects that were substantively reduced once burnout was controlled for. The authors concluded that doubts about whether the two are really distinct cannot be dismissed as speculation.⁹ The WHO’s formal description of burnout is covered in how stress, anxiety and burnout differ.

What engagement is linked to at work

Higher engagement goes with better individual performance and better business-unit results in the research so far, but most of that evidence is correlational, and the largest datasets come from Gallup researchers analyzing responses to Gallup’s own survey.³¹⁰⁴

The peer-reviewed starting point is a 2002 meta-analysis in the Journal of Applied Psychology by James Harter of Gallup and colleagues. Across 7,939 business units in 36 companies, it found relationships of practical size between unit-level employee “satisfaction-engagement” and customer satisfaction, productivity, profit, employee turnover and accidents. This article draws on the abstract only.¹⁰

Gallup’s 2024 update, published by the company rather than in a journal, pooled 736 proprietary studies of 183,806 business and work units. It reports a corrected correlation of 0.49 between unit engagement and a composite of performance, and a median 23 percent difference in profitability between units in the top and bottom quarters of engagement. The report states that it does not directly address causality, and points to earlier Gallup studies that, it says, suggest engagement and financial performance influence each other, with engagement the stronger predictor.⁴

A review from outside Gallup reads the evidence more cautiously. A 2021 evidence review by the CIPD, the UK professional body for HR, found a positive link between engagement and mostly individual performance, but with consistently small effect sizes, and reverse causality in several studies: good performance may also raise engagement.¹¹

Gallup’s 2026 report puts the cost of low engagement at about $10 trillion in lost productivity in 2025, or 9 percent of global GDP. That figure is modeled, not measured: Gallup applies published meta-analytic estimates of the engagement-productivity link, including its own, to each country’s output per worker.⁵

Can an organization raise engagement?

Workplace programs have been followed by small gains in engagement, on thin evidence. In a 2017 meta-analysis of controlled workplace interventions by Caroline Knight, Malcolm Patterson and Jeremy Dawson, intervention groups scored slightly higher on engagement than comparison groups, in studies all rated at high risk of bias.¹²

Twenty studies met the review’s criteria, covering four kinds of intervention: building personal resources, building job resources, leadership training and health promotion. Pooling the 14 with usable data gave a standardized difference of 0.29, which the authors describe as small. Group-based programs showed larger effects in a subgroup analysis, results varied widely, and the review accepted both randomized and non-randomized controlled designs.¹² A larger meta-analysis of 54 controlled trials by Delia Vîrgă and colleagues, published in 2019, found a similar, weak average effect (a standardized difference of .24); this article draws on its abstract only.¹³

Correlational evidence adds detail. In the 2011 meta-analysis, engagement went with job features such as task variety, task significance, autonomy and social support, and with leader behaviors such as transformational leadership and the quality of the leader-follower relationship; the strongest of these links, for task variety, was a correlation of .53.³ In the authors’ combined model, autonomy, feedback and transformational leadership added almost nothing once other factors were included.³ The leader-follower relationship also comes up in what the research says about leadership, as the route through which leader behaviors were most strongly linked to follower performance.

Reading an engagement score before you act on it

Before acting on an engagement score, check which definition the survey used: the state of the person, the conditions around them, or their behavior.⁶ Then check how answers were turned into a number, because Gallup’s proprietary cut-off and a percent-favorable index produce figures on different scales.⁵

The items tell you which definition you have. Items asking how often people feel energetic, dedicated or absorbed measure the state.² Items about expectations, resources and recognition measure conditions.⁴ A link between scores and results is a correlation unless the study was designed to test cause.³ And because engagement and burnout measures overlap, the burnout picture belongs next to the engagement one.⁹

WiserHours

Before you act on an engagement score

• Read the actual survey items, not only the summary score
• Note whether they ask how people feel (the state) or what they have at work (conditions)
• Ask how "engaged" was turned into a number and whether the cut-off is published
• Compare only with figures scored the same way
• Treat links to business results as correlations unless the study tested cause
• Look at burnout measures alongside engagement

Find this checklist online:
https://wiserhours.com/employee-engagement/what-is-employee-engagement/#checklist-what-is-employee-engagement-score-check

The bottom line

Two measurements share the name employee engagement. Researchers mean a state of energy, dedication and absorption in the work itself. Some surveys, including Gallup’s Q12 by its own description, measure the conditions expected to produce that state, which is useful but not the same thing. When someone quotes an engagement figure, ask which of the two it counts and how “engaged” was scored before you compare it with anything.