Measurement Exercise
Patient Satisfaction Questionnaire (PSQ)
Purpose
Content
Developer
Versions (e.g.,
languages, long or
short form,
modifications)
Number of items
Subscales
Target Population
Other identified
uses
Administration
Method
Training
Time to
administer/
complete
Equipment needed
Cost, availability
General Description
Self-Administered satisfaction survey that would be applicable in general
population studies and would yield reliable and valid measures of
concepts that had both theoretical and practical importance to the
planning, administration, and evaluation of health services delivery
programs.
The PSQ have six have measure six aspects of care, access to
care (emergency care, convenience of services, nonfinancial access);
availability
(availability/family
doctors,
availability/hospitals,
availability/specialists); finances (cost of care, insurance coverage,
payment mechanisms); continuity (continuity of care/family, continuity
of care/self); interpersonal manner (consideration, explanations); quality
of care (doctor’s facilities, prudence/expenses, quality/competence);
access total (access, finances); doctor conduct (interpersonal manner,
quality of care), and general satisfaction.
Developed by Ware, Snyder, and Wright in (1972-1976) at the Southern
Illinois University School of Medicine
- The PSQ have 3 version (PSQ-I, PSQ-II and PSQ-III).
- The PSQ-III had modified from PSQ-II and PSQ-Short is available.
- The PSQ-III is available in English language and have translation
guidelines to translate to other languages.
- These modifications were tested in several pilot studies before they were
adopted for use in the MOS in PSQ-III.
- PSQ-III 50 Items.
- PSQ-Short 18 Items.
It is measure six aspects of care: 1. technical quality, 2. interpersonal
manner, 3. communication, 4. financial aspects of care, 5. time spent with
doctor, and 6. accessibility of care.
In the Medical Outcomes Study (MOS), patient satisfaction with medical
care is measured regardless of utilization and following visits.
Measuring the patient satisfaction for specific physician visits.
Self-Administered Questionnaire
The PSQ required small training, on site before fulfill the data.
- PSQ-III 50 Items need 10 – 15 minutes.
- PSQ-Short 18 Items need 5 minutes.
The PSQ may be printed paper, or via link or tablet and based on the
researcher method of collection.
Very Cheap
Scoring
Responses
Interpretation of
scores
Method of scoring
Time to score
Training to score
Training to
interpret
Norms available
Reliability
Scale: Likert Scale
How was this scale developed and tested?
Score Range: 1 - 5
The PSQ-III given the precoding of item responses (where 1 = strongly
agree and 5 = strongly disagree, as shown above), precoded responses to
all favorably worded items are recoded so that higher item scores will
indicate greater satisfaction.
All subscales are scored so that higher scores indicate greater satisfaction
with the aspect of care named by the subscale’s label, items need to be
scored so that high scores indicate greater satisfaction.
Item scoring rules depend on whether the item represents a favorable or
unfavorable opinion about medical care.
The researcher train the patient on the PSQ and the meaning of the
response which leads the patient to score the item from his/her perspective
and according to score ranges.
The researcher need to train and to be aware of the subscales and the
meaning of each process. The researcher need to be train on the link of
subscales with care process. So, need to read the PSQ manual and to
interpret each item based on the manual and care process.
Norms are available for general adult populations and for representative
samples of the U.S. non-institutionalized population (Aday, Andersen,
and Fleming, 1980; Aday, Fleming, and Andersen, 1984).
Psychometric Information
Internal consistency
Reliability estimates ranged from 0.77 to 0.89
in the MOS baseline sample, and fell below
0.80 only for the two-item Time Spent with
Doctor subscale. As expected, the
Access/Availability/Convenience subscale
proved to be the most heterogeneous, as
reflected in its low homogeneity estimate. As
illustrated in Davies et al. (1986), we
recommend item-by-item analyses before
relying on a summary score where comparing
systems of care in terms of satisfaction with
accessibility, availability, and convenience.
Test-retest/stability
Test-retest methods were used to estimate the
reliability of PSQ-II scales. As summarized
by Ware et al. (1983) for the NCHSR project
and by Davies et al. (forthcoming) for the
Health Insurance Experiment (HIE),
reliability estimates for most multi-item
scales generally met and usually well
exceeded the 0.50 standard recommended for
studies that involve group comparisons.
Inter-rater/ Intra-rater
Validity
The PSQ is self-administered and this
approach was not implemented.
Content/domain/face
The content validity of the PSQ has been
systematically examined against published
satisfaction scales and theory about the
universe of patient satisfaction concepts
(Ware et al., 1976b).
Construct
Multitrait-multimethod analysis of the PSQ
subscales and global scales with measures
using other methods provide convergent and
discriminant validity for the PSQ scales
(Hays, Hayashi, and Ware, 1987; Ware,
Snyder, and Wright, 1976b; Ware et al.,
1983).
Concurrent
A number of validity studies have linked PSQ
scores to health care experiences,
expectations, behavioral intentions, and
various health and illness behaviors (Ware
and Davies, 1983).
Predictive
The predictive validity of the PSQ has
received empirical support as well (Marquis,
Davies, and Ware, 1983; Ware and Davies,
1983).
Comments and Critique

Is this a good test for evaluating your research? Why or why not?
I think is good and will answering the research question. In addition to the number of
items are fair and will may improve the response rate and filling all items of
questionnaire.

Summarize the overall assessment of the reliability and validity of this instrument. Is it
widely used? Under what conditions?
The PSQ reliability had studied with two approach and the four types of validity were
utilized, the PSQ is valid and reliable.

Are the studies of this instrument done well? Have the samples in these studies been
representative of the full scope of subjects for whom this test could be used?
15 studies discussed the PSQ and discussed the Psychometric Information. The studies
had discussed the PSQ from different aspects and provide the reader with good
information.