| Construct | Source model | What it measures | Typical item count |
|---|---|---|---|
| Service quality gap | SERVQUAL (Parasuraman, Zeithaml & Berry, 1988) | Gap between customer expectation and perceived performance across five dimensions | 22 paired items (44 statements total) |
| Tangibles | SERVQUAL dimension | Physical facilities, equipment, staff appearance | 4 items |
| Reliability | SERVQUAL dimension | Ability to perform the promised service dependably and accurately | 5 items |
| Responsiveness | SERVQUAL dimension | Willingness to help customers and provide prompt service | 4 items |
| Assurance | SERVQUAL dimension | Staff knowledge and courtesy, and their ability to convey trust | 4 items |
| Empathy | SERVQUAL dimension | Caring, individualised attention given to customers | 5 items |
SERVQUAL remains the standard starting framework for a customer satisfaction study in tourism and hospitality, because its five dimensions map cleanly onto the service encounter — a hotel stay, a guided tour, a restaurant meal — in a way generic satisfaction scales do not. South African tourism and hospitality dissertations use it across accommodation grading levels, from budget guesthouses to five-star lodges, adapting only the specific wording to the setting.
What is SERVQUAL, and why does it fit hospitality research?
SERVQUAL, developed by Parasuraman, Zeithaml and Berry (1988) and published in the Journal of Retailing, measures service quality as the gap between what a customer expects from a service and what they perceive they actually received, across the five dimensions in the table above. It was built for and validated across multiple service industries, and hospitality and tourism researchers have adapted its wording extensively since — a hotel study rewords “modern-looking equipment” as something about room facilities or amenities, while keeping the same underlying dimension and gap-score logic.
Can I use SERVQUAL’s exact wording for free?
SERVQUAL is a widely used academic instrument published in a peer-reviewed journal, and researchers routinely adapt its item wording for their own context with a citation to the original authors — this is standard, accepted academic practice and does not require a paid licence in the way a proprietary commercial instrument does. That said, reproducing the full 22-item scale verbatim in an appendix, rather than an adapted version, is worth checking against your own university’s fair-use and copyright guidance before doing so, since practice on exactly how much of a published scale can be reproduced varies by institution. The safer default many South African postgraduates follow is to adapt the wording to their own service context, cite Parasuraman, Zeithaml and Berry (1988) as the source model, and include only a short illustrative subset of items in the body text, with the full adapted instrument in an appendix described as “adapted from SERVQUAL.”
Illustrative adapted items, by dimension
These are illustrative, adapted examples for a guesthouse or small hotel context — reword them to fit your own specific setting (a game lodge, a restaurant, a tour operator) rather than copying them directly, and always cite the original SERVQUAL source.
Tangibles: “The accommodation’s facilities were clean and well maintained.” / “Staff were dressed appropriately for their role.”
Reliability: “The booking I made matched what was actually provided on arrival.” / “Staff resolved my request correctly the first time.”
Responsiveness: “Staff responded promptly when I needed assistance.” / “Staff were willing to help with special requests.”
Assurance: “I felt confident in the competence of the staff.” / “Staff were consistently courteous.”
Empathy: “Staff gave me individual attention.” / “Staff understood my specific needs as a guest.”
Each item is typically paired — one statement capturing the customer’s expectation (“A well-run guesthouse should have clean, well-maintained facilities”) and a matching statement capturing their perception of this specific stay (“This guesthouse’s facilities were clean and well maintained”) — with the gap between the two scored on a Likert scale, commonly 1–7 or 1–5.
How do I score and analyse the gap?
The SERVQUAL gap score for each item is perception minus expectation; a negative score means the service fell short of what the customer expected, a positive score means it exceeded expectations. Dimension-level scores are the mean of their component items, and an overall service quality score is often the mean of the five dimension scores, though some studies report dimension-level results only, since collapsing to one overall number can hide which specific dimension is actually driving dissatisfaction. Analyse dimension scores with descriptive statistics first, then a comparison test (independent-samples t-test or ANOVA) if the research question involves comparing groups — first-time versus repeat guests, or different accommodation types, for example.
A worked illustrative example
This example is entirely illustrative and fictional; the numbers are invented for the demonstration. Working title: “Service Quality and Guest Satisfaction at Small Boutique Guesthouses in the Western Cape: A SERVQUAL-Based Study.” Research objective: to determine which SERVQUAL dimensions most strongly predict overall guest satisfaction at boutique guesthouses in a defined region. Population and sample: guests who completed a stay at 12 participating guesthouses over a defined period, target sample n=250, recruited via a post-stay email survey link. Instrument: the standard 22-item SERVQUAL pairs, adapted for a guesthouse context as illustrated above, plus two revisit-intention items and basic demographic questions. Pilot: the adapted instrument was piloted with 15 recent guests; Cronbach’s alpha per dimension ranged from 0.78 to 0.89, all above 0.70, so no further item revision was needed. Illustrative analysis: multiple regression with overall satisfaction as the outcome and the five dimension scores as predictors, reported alongside descriptive statistics for each of the five dimensions.
What if my study also wants to measure revisit intention or loyalty?
Customer satisfaction and behavioural intention (would this customer return, or recommend the destination or property to others) are related but distinct constructs, and a dissertation studying both should treat them as separate sections of the questionnaire, with satisfaction items preceding intention items. A simple, commonly used approach adapts two or three direct items — for example, “I intend to return to this destination/property in the future” and “I would recommend this destination/property to others” — scored on the same Likert scale as the satisfaction items, rather than importing an entire separate loyalty scale without checking that it is genuinely needed for the research question.
How does this connect to the theoretical framework chapter?
SERVQUAL is itself grounded in the disconfirmation paradigm — the idea that satisfaction results from comparing expectation against perceived performance — so a dissertation using it should introduce that underlying theory in the theoretical framework chapter, not only describe SERVQUAL as a questionnaire. Naming the disconfirmation paradigm explicitly, and citing Oliver’s foundational work on expectation-disconfirmation theory alongside Parasuraman, Zeithaml and Berry, gives the methodology a theoretical anchor beyond “this is a well-known instrument” — which is a distinction examiners in tourism and hospitality programmes specifically look for.
How do I validate the adapted instrument before using it?
Adapting SERVQUAL’s wording to a new context does not exempt the instrument from validation. Run the adapted questionnaire past two or three people familiar with the specific hospitality context (content validity), pilot it with a small sample before full data collection, and report Cronbach’s alpha for each dimension once data are collected — a coefficient below roughly 0.70 signals that a dimension’s items are not measuring the same underlying construct consistently and may need revision. The site’s guide to proving a questionnaire is valid covers this process in more depth for any adapted instrument, not only SERVQUAL.
What mistakes do examiners flag most often in SERVQUAL-based hospitality studies?
Three recur. First, using SERVQUAL’s original retail or banking wording without adapting it to the hospitality context at all — a question about “modern equipment” makes little sense to a guest unless it is reworded around rooms, amenities or facilities. Second, reporting only a single overall satisfaction score and discarding the five dimension scores, which throws away exactly the diagnostic detail SERVQUAL is designed to provide — a hotel might score well overall while still scoring poorly on responsiveness specifically, and that distinction disappears if only the average is reported. Third, treating “satisfaction” and “service quality” as identical constructs in the write-up when the study has actually measured one or the other — SERVQUAL measures perceived service quality, which is related to but conceptually distinct from overall satisfaction, and a precise dissertation keeps the two terms separate rather than using them interchangeably.
How does this compare to other survey tools on this site?
The site’s comparison of online survey tools covers the platforms you would use to actually field a questionnaire like this one (Google Forms, Qualtrics, SurveyMonkey and similar), which is a separate decision from which items or scale to put inside that questionnaire — this piece addresses the content of the instrument itself.
Where do I go once the data are collected?
Once satisfaction data are collected and the reliability checks above are complete, the next step is choosing the right statistical test for your specific comparison or relationship — the site’s guide to choosing the right statistical test for a dissertation covers the options for comparing groups or testing relationships between satisfaction and other variables such as loyalty or spend.
Adapting an established instrument like SERVQUAL correctly — wording, scoring, validation — is exactly the kind of structuring work Tesify helps with, while your specific context and every word of the final questionnaire stay entirely your own.
Frequently asked questions
Do I need to use all five SERVQUAL dimensions, or can I drop one?
Some hospitality studies drop or merge dimensions that overlap heavily in their specific context (tangibles and assurance sometimes blur in a small guesthouse setting, for example) — this is defensible if you state and justify the adaptation, rather than silently dropping a dimension without explanation.
Is SERVQUAL the only option for measuring service quality in tourism?
No — SERVPERF (a performance-only variant that drops the expectation half of each item pair) and HOLSERV (a hospitality-specific adaptation) are also used in the literature. SERVQUAL remains the most widely cited starting point, but name whichever variant you actually use precisely in your methodology chapter.
How many respondents do I need for a SERVQUAL-based hospitality study?
This depends on your planned analysis rather than a fixed rule — a study comparing two groups with a t-test needs a smaller sample than one running a full factor analysis on all 22 items; consult your supervisor or a sample-size calculation tool once your specific analysis plan is set.
Can I administer the survey in a language other than English?
Yes, and for a South African tourism or hospitality study with local guests this is often more appropriate — translate the adapted items carefully, ideally with a back-translation check, and note the translation process in your methodology chapter.
Should I survey guests during their stay or after they leave?
Both are used in the literature; during-stay surveys typically get higher response rates but may understate dissatisfaction guests are reluctant to voice in person, while post-stay surveys (email or online, after checkout) can capture more considered responses but often have lower response rates. State and justify whichever timing you choose.
Can this item bank be adapted for a restaurant or tour-operator dissertation instead of accommodation?
Yes — the five SERVQUAL dimensions and the gap-score logic apply across hospitality and tourism sub-sectors; only the specific item wording needs to change to match the service encounter being studied.