Marketing & consumer behavior
Customer satisfaction scale: what it is and how to measure
- Concept
- Customer Satisfaction
- Field
- Marketing & consumer behavior
- Scales in the library
- 3
- Items
- 10
Customer satisfaction is a customer's overall evaluation after comparing the actual experience with what they expected. It is central to expectation-confirmation theory (Oliver, 1980), Bhattacherjee's (2001) ECM for continued use, and DeLone and McLean's (2003) information systems success model. In theses in Vietnam, satisfaction is often a mediator between service quality and loyalty, with topics such as digital banking, food delivery apps, online learning systems, or hospitals.
Comparing the scales
| Scale | Author (year) | Items | Response format | When to use it |
|---|---|---|---|---|
| Satisfaction (SAT) IS continuance model (ECM) | Vedadi & Warkentin (2016) | 4 | Likert 7-point | Choose this to measure overall satisfaction after using a digital app or service; 4 semantic-differential adjective pairs, not tied to any industry. |
| User Satisfaction (US) IS success scale | Cidral, Aparicio & Oliveira (2020) | 4 | Likert 7-point | Choose this for models following DeLone and McLean, for example satisfaction with an e-learning system or management software. |
| General Satisfaction (GS) Patient satisfaction (PSQ-18) | Marshall & Hays (1994) | 2 | Likert 5-point | Choose this for patient surveys; with only 2 items it works as a supplementary measure and should not be the only dependent variable. |
Items of each scale
Original English wording from the published source. Replace placeholders such as [system] or
[brand] with what your study is about, and pilot the questionnaire with 20 to 30 respondents first.
Satisfaction (SAT) · IS continuance model (ECM)
| Code | Item |
|---|---|
| SAT1 | Very dissatisfied/Very satisfied |
| SAT2 | Very displeased/Very pleased |
| SAT3 | Very frustrated/Very contented |
| SAT4 | Absolutely terrible/Absolutely delighted |
Source: Vedadi & Warkentin (2016). Response scale, licence and APA reference: IS continuance model (ECM).
User Satisfaction (US) · IS success scale
| Code | Item |
|---|---|
| US1 | How adequately does the [e-learning system] support your area of study? |
| US2 | How efficient is the [e-learning system]? |
| US3 | How effective is the [e-learning system]? |
| US4 | Are you satisfied with the [e-learning system] on the whole? |
Source: Cidral, Aparicio & Oliveira (2020). Response scale, licence and APA reference: IS success scale.
General Satisfaction (GS) · Patient satisfaction (PSQ-18)
| Code | Item |
|---|---|
| PSQ3 | The medical care I have been receiving is just about perfect. |
| PSQ17 (R) | I am dissatisfied with some things about the medical care I receive. |
Source: Marshall & Hays (1994). Response scale, licence and APA reference: Patient satisfaction (PSQ-18).
(R): reverse-coded item; recode before analysis. Patient satisfaction (PSQ-18): 1–5 scale: new score = 1 + 5 − score = 6 − score. If you keep RAND's numbering (1 = Strongly agree), follow the original scoring key instead of this formula.
Tips
- The ECM SAT scale uses 7-point semantic-differential adjective pairs, not agree-disagree Likert. Put a shared lead-in such as 'Your feeling about using [service] is:' before the 4 adjective pairs.
- The user satisfaction scale by Cidral et al. was written for online learning (US1). For other fields, adapt this item to your context and state that you adapted it.
- The PSQ-18 has a reverse-worded item (PSQ17). Reverse-score it before computing Cronbach's alpha, otherwise alpha will be unusually low.
- The library has no general customer satisfaction scale for retail or hotels. If you use the ECM SAT scale for these fields, adapt the lead-in to your target, cite the adaptation from Bhattacherjee (2001), and re-test reliability and convergent validity.
Once you have a scale, put the concept in the research model drawer to state your hypotheses, work out the minimum sample size, and after data collection check reliability with the Cronbach's alpha calculator.
Frequently asked questions
How do customer satisfaction and service quality differ?
Service quality is an evaluation of specific aspects of a service (reliability, responsiveness, empathy, etc.). Satisfaction is the overall feeling after using the service. In models, service quality is usually the independent variable affecting satisfaction.
How many items should a satisfaction scale have?
3 to 5 items are enough for a unidimensional construct. The ECM SAT scale and DeLone and McLean's user satisfaction scale both have 4 items.
Can I measure satisfaction with a single item?
You can in app surveys, but a thesis running PLS-SEM or CFA needs a multi-item scale to test reliability and convergent validity. Use a scale of 3 or more items.