SERVQUAL for Public Services
- Introduced by
- A. Parasuraman, Valarie A. Zeithaml & Leonard L. Berry
- Year
- 1988
- Type
- Multi-item service-quality survey scale
- Family
- Gap-model service quality measurement
What does it do?
SERVQUAL is a multi-item survey instrument for measuring perceived service quality as the gap between what service users expect and what they perceive they received.
Developed by A. Parasuraman, Valarie Zeithaml and Leonard Berry — conceptually in their 1985 gap model and operationally in the 1988 22-item scale — it assesses quality along five dimensions: tangibles, reliability, responsiveness, assurance and empathy. Applied to public services such as healthcare, licensing, social services and local government, SERVQUAL reframes citizens as service users whose expectations and perceptions can be measured, producing diagnostic gap scores that pinpoint where a public service is falling short of what the public expects.
The intuition behind the method
Think of service quality not as an absolute but as the distance between a hope and an experience. Before visiting a government office to renew a permit, a citizen carries expectations — they hope it will be quick, the staff will be helpful, the forms will be clear. Afterward they have perceptions of what actually happened. If the experience matched or exceeded the hope, quality feels high; if it fell short, quality feels poor, even if the office was objectively efficient. SERVQUAL measures both the hope and the experience on the same questions and subtracts one from the other, so a negative gap on 'responsiveness' tells managers precisely that citizens expected faster help than they got, rather than leaving them to guess.
History
Service quality was hard to measure until A. Parasuraman, Valarie Zeithaml and Leonard Berry developed the gap model of service quality, set out in their 1985 Journal of Marketing article, which located quality in the gaps between expectations, management perceptions, service standards, delivery and communication. In 1988 they operationalized the model as SERVQUAL, a 22-item, five-dimension scale published in the Journal of Retailing, which became one of the most widely used instruments in services research. From the 1990s onward public-administration and health-services scholars adapted SERVQUAL to measure the quality of government and public services, prompting both refinements and the perceptions-only SERVPERF debate.
When to use it
Use SERVQUAL when a public organization wants a structured, comparable measure of how well a citizen-facing service meets user expectations and where specifically it falls short. It is well suited to recurring service-quality monitoring, before-and-after evaluation of service improvements, and benchmarking comparable services across offices or regions. The method assumes that users can meaningfully separate expectations from perceptions, that the five dimensions are relevant to the service, and that the gap between expectation and perception is the right operationalization of quality. It is less appropriate when respondents struggle to recall or articulate prior expectations, when the service has no comparable 'excellent' reference point, or when objective process metrics (waiting times, error rates) answer the question more directly. Some critics prefer a perceptions-only variant (SERVPERF) when expectation measurement proves unstable.
Limitations and considerations
- Measuring expectations is contentious: respondents may report expectations after the experience, biasing the gap, and some users have only vague prior expectations.
- The five-dimension structure does not always replicate cleanly in public-service or non-Western samples, requiring revalidation when adapted.
- Difference (gap) scores can have weaker statistical properties than direct perception scores, motivating the SERVPERF alternative.
- As a self-report perception measure it captures how the service feels, which may diverge from objective performance and from equity considerations central to public services.
Source conditions do not replace independent scientific review or validation by an analysis engine.
Strengths
- Provides a diagnostic, dimension-level picture of service quality that pinpoints exactly where a service falls short rather than yielding a single satisfaction number.
- Built on the widely validated five-dimension gap model, enabling comparison across services, offices and over time.
- The expectation-perception design captures quality relative to what users actually want, which pure satisfaction scales miss.
- Adaptable across many public-service settings — healthcare, local government, licensing, social services — with modest rewording of items.
Applications
- Measuring patient-perceived quality in public hospitals and primary-care clinics.
- Assessing citizen experience at government service counters such as licensing, registration and tax offices.
- Evaluating local-government services like waste collection, libraries and social services.
- Benchmarking service quality across regional offices or branches of a public agency to target improvement.
- Tracking the impact of service-redesign or digital-transformation initiatives through before-and-after gap measurement.
How it works
How it works
The instrument is organized around five service-quality dimensions distilled from the original gap model: tangibles (physical facilities, equipment and appearance of staff), reliability (delivering the promised service dependably and accurately), responsiveness (willingness to help and provide prompt service), assurance (knowledge and courtesy of staff and their ability to inspire trust) and empathy (caring, individualized attention). For public services these dimensions are interpreted in terms of citizen-facing service encounters rather than commercial transactions.
Respondents complete two parallel batteries of items — conventionally 22 items grouped under the five dimensions. One battery measures their expectations of an excellent service of that type, and the other measures their perceptions of the specific service they actually received, each rated on a Likert scale. Pairing the same items across the two batteries is what allows quality to be expressed as a directly comparable difference.
For each item the gap score is computed as the perception rating minus the corresponding expectation rating. A negative gap indicates that perceived performance fell short of expectations, a zero gap indicates expectations were just met, and a positive gap indicates they were exceeded. Item-level gaps are then averaged within each of the five dimensions to give dimension-level gap scores.
Formula
Q_i = P_i - E_i, \qquad i = 1,\dots,22
Because the five dimensions are not equally important to users, respondents are often asked to distribute importance weights across them, and an overall weighted service-quality score is formed as the sum of each dimension's mean gap times its importance weight, with the weights summing to one. Weighting ensures that improvement priorities reflect what matters most to citizens rather than treating all dimensions as interchangeable.
Formula
\text{SQ} = \sum_{d=1}^{5} w_d \, \bar{Q}_d, \quad \sum_{d=1}^{5} w_d = 1The internal consistency of each dimension's item set is checked with Cronbach's alpha to confirm the items cohere as a scale, and the instrument's validity is examined through factor analysis. Reliability checks matter especially when SERVQUAL is adapted to a new public-service context, because the standard items may behave differently for, say, hospital outpatients than for retail customers.
Formula
\alpha = \frac{K}{K-1}\left(1 - \frac{\sum_{i=1}^{K}\sigma_i^2}{\sigma_T^2}\right)The dimension-level and overall gap scores are interpreted to diagnose where the service falls short and to prioritize improvement. Persistent negative gaps point managers to specific failures — in the gap-model logic, gaps between citizen expectations and management perceptions, service standards, actual delivery, or external communication — guiding targeted action rather than generic satisfaction drives.
Common pitfalls
- Administering the expectation battery after the service encounter, so reported expectations are contaminated by the experience just had.
- Importing the standard 22 items verbatim into a public-service context without checking that the dimensions still hold via factor analysis.
- Reporting only an overall score and hiding which dimension drives a poor result, defeating the instrument's diagnostic purpose.
- Treating citizens purely as customers and ignoring public-service values such as equity, due process and universal access that satisfaction gaps do not capture.
Frequently asked questions
What is the difference between SERVQUAL and SERVPERF?
SERVQUAL measures quality as perceptions minus expectations, using two parallel item batteries. SERVPERF, proposed by Cronin and Taylor, drops the expectation battery and measures perceptions only, arguing that performance perceptions alone predict overall quality and satisfaction as well or better while avoiding the statistical and practical problems of difference scores. Many public-service studies report both, or use SERVPERF when expectation measurement proves unstable.
Can citizens really be treated as customers in public services?
Partially. SERVQUAL is useful because much of a public-service encounter — being treated promptly, courteously and reliably — genuinely resembles a service transaction, and citizens do hold expectations about it. But public services also carry obligations of equity, universal access, due process and accountability that a customer-satisfaction lens can miss. Best practice uses SERVQUAL alongside, not instead of, measures of fairness, access and outcomes so that responsiveness to articulate users does not crowd out service to the underserved.
How many respondents and items does SERVQUAL need?
The standard instrument uses 22 paired items across five dimensions, and samples should be large enough to support reliable dimension scores and, if used, factor analysis to confirm the structure — typically at least several hundred respondents for a robust validation. When adapting items for a specific public service, a pilot and a check of Cronbach's alpha per dimension are advisable before fielding the full survey.
Related methods
Sources
Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1988). SERVQUAL: A Multiple-Item Scale for Measuring Consumer Perceptions of Service Quality. Journal of Retailing, 64(1), 12–40.
Open publication referenceCopied.Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1985). A Conceptual Model of Service Quality and Its Implications for Future Research. Journal of Marketing, 49(4), 41–50.
Open DOI referenceCopied.