4. • Their principal use is twofold:
1. Within organizations, the collection, analysis
g , , y
and dissemination of these data ensuring
that they have a positive experience with
y p p
the company’s goods and services
services.
2.
2 Although sales or market share can indicate
how well a firm is performing currently,
satisfaction is an indicator of how many
customers will make further purchases in
the future
future.
5. • There are various ways to obtain customer
g
feedback with regard to their satisfaction
levels with the product(s) and the
company. Some of them are,
company Some of them are
– Telephone follow‐up
Telephone follow‐
– Customer complaint data
Customer complaint d
l data
– Direct customer visits
Direct customer visits
– Customer advisory councils
Customer advisory councils
– User conferences
User conferences
6. • Th
There are ththree common methods t gather survey
th d to th
data: face‐to‐face interviews, telephone interviews,
and mailed questionnaires.
• The personal interview method requires the
interviewer to ask questions based on a pre‐structured
pre‐
questionnaire and to record the answers
answers.
• Telephone interviews are less expensive than face‐to‐
face interviews. These can be monitored by the
research t
h team t ensure th t th specified i t i
to that the ifi d interview
procedure is followed.
• The mailed questionnaire is less expensive However,
expensive.
this savings is usually at the expense of response rates
rates.
7. Face‐to‐Face Telephone Interview
Telephone Interview Mailed
Mailed
Interview Questionnaire
Gives very high Less expensive and special Very Less
degree of validity methods needed to increase Expensive but
of the data efficiency. more chances
for low response
p
rate.
The major The limitations of this Questionnaire
limitations are method are the lack of direct development
costs and factors observation, the lack of using requires
concerning the exhibits for explanation, and professional
interviewer.
i i the li i d group of potential
h limited f i l knowledge and
k l d d
respondents—those who can experience and
be reached by telephone. should be dealt
with accordingly.
8. Survey
se Rate
ations
Length of
bility
view
pling
bits
dity
ed
st
Spee
ype of S
Observa
Exhib
Cos
Valid
Interv
Flexib
espons
Samp
Re
O
Ty
In Person – – + – + – + – ++ ++ + + ++
Phone + + ++ + + – – – +
Mail ++ – –– – – – + +
+– –
– = Disadvantage – – = Worst
g
+ = Advantage ++ = Best
+ – = Could be an Advantage or a Disadvantage
9. • Wh th customer b
When the t base i l
is large it i t costly t
large, is too tl to
survey all customers.
customers
• Estimating the satisfaction level of the entire
customer population through a representative
sample is more efficient
efficient.
• To do that Scientific probability sampling methods
are used.
• There are four basic types of probability sampling
sampling:
– Simple Random Sampling
– Systematic Sampling
– Stratified Sampling
– Cluster Sampling.
Sampling.
10.
11. • I systematic sampling one simply goes d
In t ti li i l down a
list taking every kth individual, starting with a
randomly selected case among the first k
y g f
individuals.
• Systematic sampling is simpler than random
sampling if a li t i extremely l
li list is t l long or a l large
sample is to be drawn.
• However there are two types of situations in
However,
which systematic sampling may introduce biases:
– (1) The entries on the list may have been ordered so
that a trend occurs and
– (2) the list may possess some cyclical characteristic
that coincides with the k value.
12. • I
In a stratified sample, we first
t tifi d l fi t
, and
.
• The strata are usually based on important variables
pertaining to the parameter of interest.
t i i t th t fi t t
• Stratified sampling, when properly designed, is
more efficient than simple random sampling and
more efficient than simple random sampling and
systematic sampling.
• Stratified samples can be designed to yield greater
p g y g
accuracy for the same cost, or for the same
accuracy with less cost.
13. • I t tifi d
In stratified sampling we sample within each
li l ithi h
stratum. Sometimes it is advantageous to divide
the
.
• A cluster sample is a simple random sample in
which each sampling unit is a cluster of elements.
hi h h li it i l t f l t
• Usually geographical units such as
.
• Cluster sampling is generally less efficient than
p g g y
simple random sampling, but it is
.
14. • H l
How large a sample i sufficient?
l is ffi i t?
• The answer to this question
and
.
• The higher the level of confidence we want from
the sample estimate, and the smaller the error
margin, , and vice
versa.
versa
• For each probability sampling method,
,
some of which (such as that for cluster sampling)
are quite complicated.
15. • The following formula is for
(e.g., percent satisfied) based
on
on :
NZ × p(1 − p )
2
n=
NB 2 + [ Z 2 × p (1 − p)]
N – Population Size for 80% Confidence Level, Z = 1.28
Z – Z Statistic from normal distribution for 85% Confidence Level, Z = 1.45
P – Estimated Satisfaction Level for 90% Confidence Level, Z = 1.65
B – Margin of Error for 95% Confidence Level, Z = 1.96
16. • The five‐point satisfaction scale (
h fi i if i l (very
satisfied, satisfied, neutral, dissatisfied,
and very dissatisfied) is often used in
customer satisfaction surveys.
• The data are usually summarized in terms
p
of percent satisfied.
• In presentation, run charts or bar charts to
show the trend of percent satisfied are
show the trend of percent satisfied are
often used.