How Emotion-Regulation Preferences Relate to Service Performance

How do preferences relate to service performance?

Workplace emotion-regulation preferences may be related to customer-service performance, but the direct evidence is preliminary: one measure-development study reported differences between preference groups. That finding does not tell a customer-facing worker how accurately they will handle a transaction, respond to a customer, or perform under pressure. A group pattern is not an individual forecast.

The distinction matters because “performance” can mean getting the order right, displaying an expected emotion, earning a customer response, or sustaining the work without undue strain. Those outcomes can move differently. The evidence below separates them and asks what the studies actually measured. For a worker interpreting a preference result, the useful question is whether a reported tendency helps explain a behavior worth examining at work—not whether it settles how capable they are. A preference result describes a self-reported inclination; service quality is an evaluation of work, and neither term specifies one universal measure. The studies that follow therefore cannot be collapsed into a single verdict about who is good with customers. Some examine accuracy on a task, others the emotional display or the customer’s response, while still others track strain. Keeping those outcomes distinct lets a worker use the research to frame a concrete question about a work situation without mistaking a broad label for evidence about every part of the job.

What does the direct preference study establish?

A preference is a person’s reported tendency toward particular ways of regulating emotion at work. It describes what someone says they are inclined to do; it is not a record of what they did in a specific customer encounter. The distinction is central to the Workplace Emotion Regulation Preference Inventory (WERPI), the measure examined in “Development and validation of a measure for emotion regulation at work: Introducing the Workplace Emotion Regulation Preference Inventory (WERPI).” The study reported preliminary validity evidence and statistically significant customer-service performance differences among four groups identified by preference. That is a direct connection worth taking seriously: people grouped by their reported regulation preferences did not have identical average performance in the study.

The study was a measure-development project, so the result speaks first to whether WERPI preferences showed a relationship with the study’s customer-service measure in that sample. The accessible record describes 413 undergraduates, ages 17 to 64, who had at least two years of employment history. About 75 percent were women; 40 percent were currently employed, and nearly all had worked within the previous two years. The employment-history requirement gave participants some work experience, but this was still a predominantly undergraduate group rather than a broad sample of established customer-service workers across occupations. Its result should be read within that setting. The age range is wide, but age range alone does not make a sample representative: participants shared an undergraduate status and were recruited under the study’s employment-history condition. A reader should attend to that composition when carrying the finding into a different service setting, such as experienced staff working under distinct scripts, incentives, or customer demands.

A statistically significant difference between four groups addresses a group-level question: did the study detect variation in customer-service performance across those categories? It does not by itself show whether WERPI can estimate a new individual’s future performance accurately. For that, a reader would need evidence about how well the measure predicts people it has not already been used to group, including the size and practical usefulness of its errors. The accessible abstract and article material available for this assignment report the group difference, but do not establish such individual forecasting accuracy. A significant association can exist even when the groups overlap considerably; without prediction results, the amount of overlap and the usefulness of a score for an individual remain unknown. Group comparisons can also be statistically detectable without yielding a useful rule for sorting individuals. The reported result answers whether the means differed under the study’s analysis; it does not report how much knowing a person’s category improves a forecast over a reasonable baseline.

There is a second boundary: a preference inventory and a measure of behavior during an encounter answer different questions. Someone may report an inclination toward one approach yet act differently when a customer is hostile, a policy is restrictive, or a conversation moves quickly. Conversely, one encounter cannot establish a stable preference. WERPI’s reported group differences therefore make workplace regulation preferences a plausible factor to investigate, not a substitute for observing the work behavior or naming the outcome at issue.

For a customer-facing worker, the defensible reading is modest but useful. WERPI offers preliminary group-level evidence that reported preferences and customer-service performance can be related. The result supports asking how regulation tendencies might show up in work; it does not support treating a preference category as a verdict on an individual’s service ability. The distinction also sets up the broader research: studies of emotional labor examine strategies enacted under particular demands, and their findings need to be kept separate from direct evidence about reported preferences.

Sources: Development and validation of a measure for emotion regulation at work: Introducing the Workplace Emotion Regulation Preference Inventory (WERPI)

How do broader emotional-labor findings change the picture?

In workplace research, emotional labor means regulating feeling or its display to meet expectations attached to a job. Those expectations can be explicit, such as a service script, or understood from the role. When a worker feels an emotion that conflicts with the display the job calls for, researchers describe emotion-rule dissonance: the felt state and the expected expression do not match. A worker who is irritated but is expected to sound welcoming faces that mismatch. How the person responds to it matters; the label “emotional labor” by itself does not say which strategy they used or what followed.

The common distinction is between surface acting and deep acting. Surface acting changes the outward expression while the underlying feeling remains much the same: someone suppresses irritation and puts on a polite tone. Deep acting involves trying to shift the feeling itself so that the required expression feels more congruent, perhaps by reappraising what happened. These are strategies used in a situation, shaped by the task and its demands. They are not identical to a reported preference for one strategy, and a person may use different approaches across encounters.

The meta-analysis “On the costs and benefits of emotional labor: a meta-analysis of three decades of research” pooled 494 correlations from 95 independent studies. It found a marked difference between well-being and performance outcomes. Emotion-rule dissonance and surface acting had substantial associations with impaired well-being indicators (corrected correlations, ρ, from .39 to .48), while their associations with job attitudes ranged from −.24 to −.40. In ordinary terms, the studies tended to find that more mismatch or surface acting went along with worse reported strain and attitudes. These are pooled relationships, not a statement that every worker experiences the same cost.

The performance relationships were smaller: the meta-analysis reported a small negative range, ρ = −.20 to −.05. A correlation near the upper end of that range is close to no relationship; even the lower end describes a modest association, not a prediction precise enough to infer how one person will perform. “Performance” also combines outcomes measured in different studies, so this pooled estimate should not be read as a single service-quality score. The finding shifts the emphasis: the broadest and clearest pattern in this synthesis concerns well-being, while links with performance are weaker and depend on what each study counted as performance.

Deep acting complicates a simple story that emotional regulation necessarily damages work. In the same meta-analysis it had weak relationships with impaired well-being and job attitudes, alongside positive associations with emotional performance (ρ = .18) and customer satisfaction (ρ = .37). Those positive values indicate a tendency for higher deep acting to accompany more favorable ratings on those outcomes across the pooled evidence. They do not show that adopting deep acting will cause customers to be more satisfied, or that it improves accuracy on every task. The variation by strategy and criterion is why a claim about “regulation” in general would conceal the result that matters.

For the preference question, this research supplies a mechanism to consider, not a validation test for a preference inventory. It shows that what workers do under emotional display demands has associations with both strain and selected performance measures. A preference score would need its own evidence to show that it captures a stable tendency and relates usefully to work outcomes. The meta-analysis did not test the WERPI or establish that a person’s stated preference predicts the strategy they enact in a particular encounter. Its practical contribution is narrower: asking about a preference can prompt a question about how regulation is experienced at work, but the answer requires observing the relevant behavior and outcome in context.

Because the pooled results combine studies and settings, they cannot say why two similar strategies may have different consequences across encounters. Some service interactions include a receptive customer; others begin with impatience or conflict. That social context could alter the links between acting strategies, exhaustion, and task performance. A daily study of supermarket checkout operators tests that possibility directly, by examining customer affiliative behavior alongside employees’ reported strategies and recorded work outcomes.

Sources: On the costs and benefits of emotional labor: a meta-analysis of three decades of research

When does the customer change the relationship?

A customer’s behavior can be part of the emotional setting in which a service worker regulates emotion. “How does customer affiliative behaviour shape the outcomes of employee emotion regulation? A daily diary study of supermarket checkout operators” studied 49 checkout operators and linked diary data with store performance records. The authors tested two accounts of customer affiliative behavior, which included smiling and brief conversation. One treated it as a pathway through which acting strategies related to outcomes; the other tested whether it changed the strength of those relationships. The distinction matters because a customer’s conduct may be part of the process connecting a strategy with an outcome, or a condition under which that connection differs.

In the mediation model, customer affiliative behavior was involved in the relationships between both deep and surface acting and employee well-being. For surface acting, the study also found an indirect path to task performance through customer affiliative behavior and emotional exhaustion. In plain terms, the analyzed pattern connected how employees acted with how affiliative customers behaved, and connected that customer behavior and exhaustion with task results. Mediation describes the structure of those observed relationships; it does not demonstrate a causal chain. In particular, it does not establish that a customer’s smile changed an employee’s feelings, which then changed work speed.

The study’s performance measure makes the claim concrete. Rather than asking customers for a general service-quality judgment, the researchers used store records of checkout task performance; the accessible study manuscript specifies scanning rate, or items scanned per minute. That measure captures the pace of one task in a supermarket checkout setting. It does not capture whether the interaction felt respectful, whether the transaction was error-free, or how the employee handled a complicated request. The customer behavior and employee well-being variables therefore help interpret this specific operational outcome, not every dimension that a reader might include under “good service.”

The second, moderation model asked whether customer affiliative behavior altered the association between a strategy and an outcome. The study reported that customer affiliative behavior moderated the relationship between deep acting and emotional exhaustion, and also the indirect relationship between deep acting and task performance through exhaustion. In statistical terms, the estimated relationship differed depending on the level of customer affiliation. That is a conditional pattern: the same reported acting strategy did not have one uniform association across all recorded customer contexts in the model. The abstract does not warrant turning this result into a simple rule that friendly customers protect workers or improve their performance.

Daily diary data bring repeated work observations closer to the encounters being studied than a one-time preference questionnaire can. Linking them with store records also gives the task outcome a behavioral basis. But the diary design remained observational: employees were not randomly assigned to customers who smiled or made conversation, and customer conduct could coincide with other features of the interaction. With 49 operators in supermarket checkout work, the findings add a plausible context-sensitive account of strategy and task performance. They cannot show that friendly customers cause better service or that the same pathways hold in a call center, clinic, or a worker’s next shift.

Sources: How does customer affiliative behaviour shape the outcomes of employee emotion regulation? A daily diary study of supermarket checkout operators

What did a controlled service simulation isolate?

“Display rules versus display autonomy: Emotion regulation, emotional exhaustion, and task performance in a call center simulation” tested a workplace condition rather than asking workers which strategy they generally preferred. Participants completed three simulated customer interactions. In one condition, they were assigned positive display rules: the job required them to present positive emotion. In the comparison condition, they had autonomy over emotional display. Because the display requirement was assigned, the study could compare what happened under those contrasting instructions while holding the brief call sequence within a controlled simulation.

The assigned positive-display condition produced more order-form errors than the autonomy condition. This is a concrete task result: the researchers counted errors in forms completed during the simulated calls, rather than relying on a broad rating of service quality. The prescribed display demand also led to greater emotional exhaustion reported after the simulation. These outcomes belong together in the finding, but they are not the same outcome: one records task accuracy, while the other records participants’ post-task experience of depletion. The study therefore links an imposed display requirement with both a performance decrement on the order-form task and greater exhaustion in this setting.

The authors examined surface-level regulation as a mechanism for the energy-depletion result. The abstract reports that surface-level regulation, rather than deep-level regulation, accounted for the effect of the display rules on exhaustion. In plain terms, the evidence points to managing the outward display without changing the underlying feeling as the process associated with greater depletion under the assigned rule. The experiment did not assign a lasting personal preference, and its result cannot be translated into a claim that some workers are naturally suited to scripted calls. Its relevant organizational contrast is whether the work setting requires a particular emotional display or permits workers more autonomy over how they express themselves.

Customer hostility had a separate reported effect. During the hostile call, participants made more order-form errors than during the non-hostile calls, and customer hostility increased exhaustion. The researchers also tested whether hostility and display rules interacted, as if a prescribed positive display would be especially damaging when the customer was hostile. That proposed interaction was unsupported. The results instead identify two experimentally examined pressures with reported effects: the display-rule condition and the hostile interaction. Without support for their interaction, the study does not warrant saying that hostility magnifies the display-rule effect or that the two pressures combine in a particular way.

Random assignment gives this simulation more leverage for interpreting the assigned instructions and hostile-call condition than an observational survey would have. It supports a narrow causal reading within the setup: changing the display requirement changed the measured exhaustion and order-form errors in these simulated interactions. The setting was still three short calls with an artificial task and scripted conditions. It cannot establish whether the same differences persist across actual shifts, how other performance criteria would respond, or what any one employee would do at work. For a workplace decision, the useful implication concerns job design: when a role prescribes emotional display, autonomy is a meaningful condition to examine alongside task accuracy and worker exhaustion, rather than treating those demands as a personal preference or a test of job suitability.

Sources: Display rules versus display autonomy: Emotion regulation, emotional exhaustion, and task performance in a call center simulation

Why can team loyalty and individual task results diverge?

“Revisiting the effect of emotional labor: A Multi-Level Investigation in Front-Line Service Teams” examined employees and customers nested within 82 front-line service teams at one large Chinese electronics provider. The researchers used a time-lagged, multi-source design: employee reports and customer reports supplied different parts of the evidence, collected at separated points in time, and the analysis distinguished team-level from individual-level outcomes. That structure matters for interpreting the findings. Customer loyalty summarized a team’s customer relationship, while task performance concerned an individual employee. A relationship at one level does not automatically describe the other.

The study found a pathway in which surface acting was associated with exhaustion, which in turn was associated with lower team-level customer loyalty and lower individual task performance. The pathway linked the same reported acting pattern and exhaustion to two results, but the results were not interchangeable. Loyalty reflects customers’ continuing orientation toward the service provider or team; task performance captures the employee’s work contribution. A reader who sees lower loyalty cannot infer that each employee made more task errors, just as an individual task result does not reveal whether customers remain loyal to the team. The study’s separate measures preserve that distinction.

The deep-acting results followed a different pattern. Deep acting was not associated with exhaustion in the reported model and was positively associated with members’ task performance. That is a distinct finding from the surface-acting pathway: the study did not report deep acting as a route to exhaustion and poorer task performance. Nor does the positive association mean that asking workers to deep-act would cause performance to rise. The reported evidence is about relationships in the measured sample, and the modeled absence of an exhaustion association should not be broadened into a universal claim that deep acting never carries a cost.

For interpreting service evidence, the design makes it possible to ask whether an employee-level behavior relates differently to an employee result and a team-level customer result. The answer here is that both lower loyalty and lower task performance appeared along the surface-acting-through-exhaustion pathway, while the distinct deep-acting result concerned individual task performance. It would be inaccurate to compress these into one outcome called “service quality,” then claim a single effect on quality. A team may retain or lose customers through a broader relationship than any one employee’s task execution; meanwhile, a worker’s measured task result does not stand in for the whole team’s customer relationship. These are different questions even when they are studied together.

Using multiple sources and separated measurement times strengthens the study’s account by reducing dependence on one person’s report of both workplace behavior and its outcome. The 82-team, nested analysis also fits the fact that employees work within teams while customers respond to service delivered by a team. But the researchers observed one provider rather than randomly assigning emotional-labor strategies. Other features of the provider, its teams, and the work may help explain the associations. The findings therefore support careful level-by-level interpretation in this setting; they do not establish that surface acting caused either outcome or that the same pattern will hold in every customer-facing job.

Sources: Revisiting the effect of emotional labor: A Multi-Level Investigation in Front-Line Service Teams

What counts as a customer response—and what does it leave out?

The two studies in “Want a tip? Service performance as a function of emotion regulation and extraversion” make the word performance harder to use casually. In the restaurant study, deep acting related to customer expectations and tips. In the call-center simulation, emotional performance and extra-role service behavior were among the outcomes examined; the reported surface-acting relationships varied with extraversion and whether the behavior was in-role or extra-role. These are findings about how emotional display and customers’ responses or judgments relate, not a common score applied across settings. Even within the same broad service occupation, an outcome depends on whose response is recorded and what they are responding to. The server study brings customers’ expectations and gratuities into view; the simulation attends to emotional display and conduct beyond the prescribed role. Those measures cannot be swapped as if they were different names for the same task result.

A tip is a concrete response with economic meaning for a server. It can reflect a customer’s experience of an encounter, but it is not a record of whether an order was entered correctly. Customer expectations capture an evaluation of service against what the customer anticipated. Emotional performance concerns the quality of emotional display, while extra-role service refers to behavior beyond the formal task. The latter outcomes may matter to customers and employers, yet they still answer different questions from transaction accuracy or task completion. The distinction is especially important when a result is carried from a study into practice. A manager interested in reliable order handling needs evidence about errors or accuracy; someone studying whether customers felt well served may reasonably examine expectations or tips. Each criterion can support a useful question, but only the one measured supports the corresponding claim.

The study’s results therefore make a broad claim about “service quality” difficult to interpret unless the writer names what was actually measured. A favorable tip result cannot silently stand in for fewer errors; an association with emotional performance does not establish faster or more accurate work. Likewise, extra-role behavior may enrich an interaction without showing whether required duties were completed. The abstract does not provide enough measurement detail to translate these distinct outcomes into one scale or to estimate how well any individual worker would perform. That restraint matters for the conditional surface-acting findings, too: the relationship differed with extraversion and with whether behavior was part of the role or went beyond it. The result does not support a single directional rule that surface acting helps or harms every worker. It points to variation tied to the measured person characteristic and criterion, while leaving the underlying measurement detail unavailable in the abstract.

This is an interpretive rule derived from the contrast across these measures: state the criterion before saying that performance changed. If the outcome is customer expectation, report that; if it is a tip, call it a tip; if it is task accuracy, identify the accuracy measure. Such precision preserves the value of customer response and emotional delivery instead of dismissing them as soft outcomes, while preventing one positive result from being stretched into evidence about every part of service. Customer-facing work has several legitimate endpoints. The evidence becomes useful when the endpoint stays visible. It also keeps disagreement about “good service” productive: people can see whether they differ over the customer’s experience, the quality of emotional delivery, discretionary help, or completion of the required task. A study may illuminate one of these without settling the others.

Sources: Want a tip? Service performance as a function of emotion regulation and extraversion

What should you do with an EQ reflection?

Use a reflection to choose one observable part of an interaction to notice repeatedly. For example, after a customer asks about a delayed order, you might note whether your explanation was clear, then compare that with whether the transaction record was accurate. This illustration does not identify a cause; it separates an interpersonal behavior from a work result so you can decide what to practice or ask about. Keep the observation small enough to repeat: one clear behavior, one kind of work evidence, and more than a single encounter where possible. If the explanation is unclear, you have a concrete communication behavior to examine; if records are accurate while customers remain confused, you have a different question to raise about the information or process. Either way, the observation gives reflection a specific place to begin.

The private 32-item Emotional Skills Profile offers a way to reflect on recent emotional habits and judgments about authored scenarios. It is educational, not a measure of job performance. Its value here is as a prompt for examining what you tend to notice, regulate, or do in emotionally charged situations, then checking that reflection against evidence from the work itself. A self-description and a customer response answer different questions; placing them side by side may help you choose what to observe next without treating one as proof of the other. Explore the profile at [/assessment](/assessment), then compare what you notice in the reflection with relevant work evidence, such as feedback or transaction accuracy.

Sources: Emotional Skills Profile

Questions readers ask