lecture 4

Question 1

Object of PSQ (aggregation and generalization)

Answer 1

what is your problem that you want to solve

optimize service no use of servqual

Single event/encounter or multiple experience

Specific organizational or component

Input - process = output

Question 2

5 dimensions of PSQ

Answer 2

reliability

Empathy

Responsiveness

Assurance

Tangibles

Question 3

Servqual super dimensions

Answer 3

Reliability

Assurance

Tangibles

Empathy

Responsiveness

Some of the dimensions overlap

Question 4

PSQ scale (servqual)

Answer 4

22 Items on expectations and perceptions about 5 super dimensions that investigate using a 7 point likert scale

Question 5

Types of construct PSQ servqual

Answer 5

attribute level (serqual items)

Global level (PSQ, satisfaction, loyalty; one or more items per construct, use same scale as on attribute level

Measuring socio-demographics

Question 6

Measuring expectations

Answer 6

Choice of expectations significantly influences the PSQ of a service provider

Which expecations should be taken into account???

Sould (desired) ==> Will

Should ==> Telephone companies should keep their records accurately

Will ==> Telephone comany X will exist on error-free records

Question 7

Shortcomings SERVQUAL (should we measure expectations as a whole in the survey)

Answer 7

People take expectations into account when asked for perception (attitude not disconfirmation)

Correlation between expectations and perceptions

-Increase in amount of items
-Consumers adjust the expectations
+Gives managerial information

If yes ==> when to measure them??
at different moments with different samples

Question 8

ServPerf

Answer 8

service quality is more like an attitude

They say that we should only measure perception and not attitudes

Same questions just didnt do the questions once

The found out that construct validity is higher for SERVPERF and discriminant validity is higher for SERVPERF

Question 9

Validity of Servqual

Answer 9

The existence of the 5 dimensions

Convergent validity ==> the extent to which scale items assumed to represent a construct do in fact converge on the same construct (+)

Discriminant validity ==> the extent to which SERVQUAL has five distinct dimensions (- overlap)

Predictive/concurrent validity ==> extennt to which SERVQUAL scores are associated as hypothesized with other conceptually related measures (r2) (+/-)

Question 10

Points of criticism (servqual/servperf)

Answer 10

generalizability of items and dimensions (qualitative research to find attributes and dimensions)

Measuring importance:
direct: Importance measured in perception, weighted servqual leads to over estimation

Indirect:
Multicollinearity, regression based on factor analyses

Question 11

Formulation of proposition

Answer 11

When you formulate a proposition use the same level of intention (adjectives) consistently

Dont use words with extreme values as “never” or “always”

Question 12

use of tool (survey)

Answer 12

Use of likert scale

7 points or 5 or 9

Translation to 10 point scale not good (neutral center point, results in over estimation, people always bechoosing 3-8 because they know this scale)

Use of “not applicable” category, (otherwise middle chosen, or what if it asks for complaint and you dont have one)

Use of positive vs negative statement (problems in factor analyses; people may stay more alert but eh he likes all positive)

Structured or mixed item order (should similar questions be grouped; grouped they give similar answers ungrouped people get confused )
Halo effect

Question 13

Example of servPerf

Answer 13

service quality is derived from:

Reliability, responsiveness, flexibility and more: –> adapted