Attempt Only Four Case Study
CASE – 1 Consumer
Perception of High-end IT Education
This
case study of recent origin (2001), illustrates the use of free-response
questions which permit respondents to give unstructured answers. The responses
are given in the form of excerpted quotes from the study at the end of the
case. The entire study was bigger in scope and results. These reported results
are only for the purpose of illustration and do not constitute the complete
analysis.
BACKGROUND
SSI,
a computer education centre, has added Internet
to its portfolio. Now SSI plans to re-launch its course called Internet in its updated form. The
course includes ASP, XML, WAP, .NET and BLUETOOTH, the last one being offered
only by SSI’s Internet.
Research Objectives
To
find out
· the deciding factors for
taking up a particular High-End I.T. course.
· whether the course contents
of Internet are actually in “demand”.
· the strengths and weaknesses
of Internet.
Methodology
Collecting information through
· questionnaires
· face-to-face interviews
· telephonic interviews
· internet
Sample Composition
Students
of SSI as well as from competing computer education providers (NIIT, Aptech,
Radiant, Tata Infotech).
Sample size : 80 (25% SSI + 75% others)
Results from Some Free Response
Questions for Students’ Comments
The
following are quotations from some students’ comments on the institute, course,
and so on.
“Right now the I.T. market in U.S.
· Computer professional (New Jersey , USA
“MS (Micro Soft) has come up with the .NET, which
works on the Windows 2000 platform. Anything to do with Internet will be ‘hot’.
And MS won't leave it halfway”.
● Faculty
(Radiant)
“I did my GNIIT, now I am doing Java at RADIANT. Did
not continue there because I wanted to do only Java; and NIIT, though it is
very good, has only long-term courses. Want to get into an I.T. career. From
what I have heard, Aptech is not up to the mark. Don’t know much about SSI or
Internet. .NET is the latest course here.”
· Student (Radiant)
“I am doing Radiant.NET with C#, ASP.NET, XML, SOAP,
and so forth because it is the latest after Java”.
· Student (Radiant)
“I joined Radiant because I heard that the course
material is very good. Faculty is also good. Finished my Java from there. And I
plan to do a post graduate in I.T. NIIT is too expensive. Cost-wise, I guess
SSI and Radiant are comparable. Don’t know more about SSI.”
· Student (Radiant)
“I did my Java from TCI because I stay close by
(Annanagar). Radiant is more expensive. Also TCI gives me a ‘Government of
India’ certificate. I am working as a web page designer. I am being trained in
XML and so on by my company itself.”
· Ex-Student (TCI)
“.NET has not yet come into the market. hence we do
not have the course. We have C#, XML, WAP.”
· Counselor (NIIT)
“Of course NIIT is expensive compared to the other
institutes. But when one is focussed on one’s career, one does not crib about
money. After interacting with my faculty, I have a very good knowledge about
the I.T. world. Now I would not even think of changing. I have a background in
BCA and am doing my Java here.”
· Student (NIIT)
“NIIT has got a name that is recognised the world
over more than any other institute in India
· Student (NIIT)
“I just know about NIIT. So I am here. Plan to do a
short-term course here itself after my GNIIT, which I will finish this year.”
· Student (NIIT)
“I have no background in computers, but I do not
find any difficulty in doing my Internet course. NIIT and APTECH are too
expensive.”
· Student (SSI)
Question
1.
Write don a brief summary of all the answers given above. How does this
differ from the analysis of structured-response questions?
CASE – 2
Chi-square Test
Methodology
1.
A fictitious data set consisting of thirty respondents was created. The
data was mainly constructed to find the relationship between the dependent and
independent variable. Age was taken as the independent variable and choice of a
drink as dependent variable. Six brands of soft drinks were considered as the
different choices for the respondents.
2.
The age group coded into six categories as 1 to 6 and the brands of
soft drinks were coded into six categories and the codings are as follows:
(a) Independent variable
Age Coding
<15 1
16 – 25 2
26 – 35 3
36 – 45 4
46 – 55 5
>55 6
(b) Dependent variable
Different brands Coding
Coke 1
Pepsi 2
Mirinda 3
Sprite 4
Slice 5
Fruit Juice 6
(a) Independent variable
Age Coding
<15 1
16 – 25 2
26 – 35 3
36 – 45 4
46 – 55 5
>55 6
(b) Dependent variable
Different brands Coding
Coke 1
Pepsi 2
Mirinda 3
Sprite 4
Slice 5
Fruit Juice 6
3.
Chi-square test has been used to cross-tabulate and to understand the
relationship between the independent and the dependent variable.
4.
Calculation of contingency coefficient and the lambda asymmetric coefficient
is done to find the strength of the association between the two variables.
5.
Sample size is taken as thirty.
6.
Analysis of cross-tabulation.
7.
SPSS software package for the cross tabulation analysis.
Problem
This
is a bivariate problem. The basic intention of the problem is to understand the
relationship between AGE and BRAND PREFERENCE of different brands of soft
drinks.
Input Data Table
|
Serial No.
|
Age
|
AGECODE
|
SOFT DRINK
|
DRINK CODE
|
|
1
|
<15
|
1
|
FRUIT JUICE
|
6
|
|
2
|
<15
|
1
|
SPRITE
|
4
|
|
3
|
<15
|
1
|
MIRINDA
|
3
|
|
4
|
<15
|
1
|
PEPSI
|
2
|
|
5
|
<15
|
1
|
FRUIT JUICE
|
6
|
|
6
|
16-25
|
2
|
COKE
|
1
|
|
7
|
16-25
|
2
|
SLICE
|
5
|
|
8
|
16-25
|
2
|
COKE
|
1
|
|
9
|
16-25
|
2
|
PEPSI
|
2
|
|
10
|
16-25
|
2
|
MIRINDA
|
3
|
|
11
|
26-35
|
3
|
SLICE
|
5
|
|
12
|
26-35
|
3
|
SPRITE
|
4
|
|
13
|
26-35
|
3
|
FRUIT JUICE
|
6
|
|
14
|
26-35
|
3
|
PEPSI
|
2
|
|
15
|
26-35
|
3
|
SLICE
|
5
|
|
16
|
36-45
|
4
|
MIRINDA
|
3
|
|
17
|
36-45
|
4
|
FRUIT JUICE
|
6
|
|
18
|
36-45
|
4
|
FRUIT JUICE
|
6
|
|
19
|
36-45
|
4
|
SLICE
|
5
|
|
20
|
36-45
|
4
|
PEPSI
|
2
|
|
21
|
46-55
|
5
|
COKE
|
1
|
|
22
|
46-55
|
5
|
SPRITE
|
4
|
|
23
|
46-55
|
5
|
SLICE
|
5
|
|
24
|
46-55
|
5
|
FRUIT JUICE
|
6
|
|
25
|
46-55
|
5
|
SLICE
|
5
|
|
26
|
>55
|
6
|
MIRINDA
|
3
|
|
27
|
>55
|
6
|
COKE
|
1
|
|
28
|
>55
|
6
|
COKE
|
1
|
|
29
|
>55
|
6
|
PEPSI
|
2
|
|
30
|
>55
|
6
|
FRUIT JUICE
|
6
|
Output Data
Age by Drink Preference
Age |
Drink Preference
|
Code
|
<15
|
16-25
|
26-35
|
36-45
|
46-55
|
>55
|
Total
|
|
Coke
|
1
|
0
|
2
33.32%
|
0
|
0
|
1
20%
|
1
40%
|
5
16.67%
|
|
Pepsi
|
2
|
1
20%
|
1
16.67%
|
1
25%
|
1
20%
|
0
|
1
20%
|
5
16.67%
|
|
Mirinda
|
3
|
1
20%
|
1
16.67%
|
0
|
1
20%
|
0
|
1
20%
|
4
13.33%
|
|
Sprite
|
4
|
1
20%
|
0
|
1
25%
|
0
|
1
20%
|
0
|
3
30%
|
|
Slice
|
5
|
0
|
1
16.67%
|
2
50%
|
1
20%
|
2
40%
|
0
|
6
40%
|
|
Fruit Juice
|
6
|
2
40%
|
1
16.67%
|
0
|
2
40%
|
1
20%
|
1
20%
|
7
23.33%
|
|
Total
|
|
5
100%
|
6
100%
|
4
100%
|
5
100%
|
5
100%
|
5
100%
|
30
100%
|
|
Chi-Square
|
Value
|
DF
|
Significance
|
|
Pearson
|
18.22857
|
25
|
.08325
|
|
Likelihood Ratio
|
25.52646
|
25
|
.04332
|
|
Mantel-Haenszel
test for linear association
|
.13961
|
1
|
.07086
|
|
|
|
|
|
Minimum
Expected Frequency ─.500
Cells
with Expected Frequency <5─36 of 36 (100.0%)
|
Approximate
Statistics
|
Value
|
ASE 1
|
VAL/ASE 0
|
Significance
|
|
Contigency
Coefficient
|
.61479
|
|
|
.08325*1
|
|
Lambda:
|
|
|
|
|
|
Symmetric
|
.18750
|
.08892
|
1.99754
|
|
|
With 'DRINK CODE'
dependent
|
.21739
|
.12757
|
1.56813
|
|
|
With 'AGE CODE'
dependent
|
.16000
|
.07332
|
2.14834
|
|
|
Goodman &
Kruskal Tau:
|
|
|
|
|
|
With 'DRINK CODE'
dependent
|
.12432
|
.03912
|
|
.08412*2
|
|
With 'AGE CODE'
dependent
|
.12152
|
.02580
|
|
.08580*2
|
*1 Pearson Chi-square probability
*2 Based on Chi-square approximation
Number of
Missing Observations: 0
Analysis
In a Chi-square test, for a
90 per cent confidence level, if the significance level is greater than or
equal to 0.1, it signifies that there is no association between the two
variables in the cross-tabulation and if significance level is less than 0.1,
then it signifies that there is a significance relationship between the
selected variables.
The result of the cross-tabulation
From the output tables, the
Chi-square test read a significance level of 0.08325 at 90 percent confidence
level. For 90 per cent, significance level is 0.1, that is (1─0.9), so the
above result shows that at 0.08 (which is less than 0.1), there is a
significant relationship between the two variables. At 95 per cent confidence
level, significance level being 0.05, and the above output giving a
significance level of 0.08 which is greater than 0.05, there is no relationship
between the variables:
If
contingency coefficient value is greater than +0.5 then the variables are
strongly associated. In the above case the contingency coefficient value being
0.6 which is greater than 0.5, hence the variables are strongly associated.
The
asymmetric lambda value (with DRINKCODE dependent) 0.21739 means that 21.7% of
error is reduced in predicting brand preference when age is known.
From
the above result we can conclude that there is a significant relationship
between AGE (independent variable) and BRAND PREFERENCE (dependent variable),
of the respondents.
Thus
we can conclude that the age of the respondent plays an important role in the
purchasing intention of a particular brand of soft drink.
Question
Case 2:
Conduct Chi-square test to cross-tabulate and to understand the
relationship between the independent and the dependent variable. Also calculate
contingency coefficient and the lambda asymmetric coefficient to find the
strength of the association between
the two variables. Take Sample size as thirty. Analysis of
cross-tabulation using SPSS software package would be required.
CASE – 3
Tamarind Menswear
Given below is a preliminary
questionnaire for retailers and consumers of a recently launched menswear
brand. Can you list down the research objectives for both questionnaire? Can
you modify the given questionnaires to a final draft?
TAMARIND QUESTIONNAIRE FOR RETAILERS
1.
Do you have
Tamarind? Yes/No
2.
What do you think
about it?
3.
Is there place in
the market for one more readymade garment company?
4.
What kind of
products does Tamarind have? Are they good?
5.
Is it a threat to
any existing brand? If yes, which one?
6.
If it is not a
available, what is your view about advertising so heavily before the product is
launched?
7.
Are people coming
and asking for Tamarind?
8.
The range of clothes with the retailer.
9.
Price range.
10.
Name of the shop and so on.
TAMARIND QUESTIONNAIRE FOR
CONSUMERS
1.
Which ads do you recall?
2.
Which garment ads do you recall?
3.
Have you seen the Tamarind ad?
4.
What do you remember from the ads?
5.
Do you like the ad? Why?
6.
What is the main message?
7.
What kind of clothes are Tamarind?
8.
What do you think will be the price range?
9.
Will you buy it? Why?
CASE – 4 Logistics Regression
A
pharmaceutical firm that developed particular drug for women wants to
understand the characteristics that cause some of them to have an adverse
reaction to a particular drug. They collect data on 15 women who had such a
reaction and 15 who did not. The variables measured are:
1.
Systolic Blood Pressure
2.
Cholesterol Level
3.
Age of the person
4.
Whether or not the woman was pregnant (1 = yes)
The dependent variable indicates if there was an
adverse reaction (1 = yes)
TABLE
1
|
BP
|
Cholesterol
|
Age
|
Pregnant
|
DrugReaction
|
|
100
|
150
|
20
|
0
|
0
|
|
120
|
160
|
16
|
0
|
0
|
|
110
|
150
|
18
|
0
|
0
|
|
100
|
175
|
25
|
0
|
0
|
|
95
|
250
|
36
|
0
|
0
|
|
110
|
200
|
56
|
0
|
0
|
|
120
|
180
|
59
|
0
|
0
|
|
150
|
175
|
45
|
0
|
0
|
|
160
|
185
|
40
|
0
|
0
|
|
125
|
195
|
20
|
1
|
0
|
|
135
|
190
|
18
|
1
|
0
|
|
165
|
200
|
25
|
1
|
0
|
|
145
|
175
|
30
|
1
|
0
|
|
120
|
180
|
28
|
1
|
0
|
|
100
|
180
|
21
|
1
|
0
|
|
100
|
160
|
19
|
1
|
1
|
|
95
|
250
|
18
|
1
|
1
|
|
120
|
200
|
30
|
1
|
1
|
|
125
|
240
|
29
|
1
|
1
|
|
130
|
172
|
30
|
1
|
1
|
|
120
|
130
|
35
|
1
|
1
|
|
120
|
140
|
38
|
1
|
1
|
|
125
|
160
|
32
|
1
|
1
|
|
115
|
185
|
40
|
1
|
1
|
|
150
|
195
|
65
|
0
|
1
|
|
130
|
175
|
72
|
0
|
1
|
|
170
|
200
|
56
|
0
|
1
|
|
145
|
210
|
58
|
0
|
1
|
|
180
|
200
|
81
|
0
|
1
|
|
140
|
190
|
73
|
0
|
1
|
SPSS Output
TABLE
2 Model Summary
|
Step
|
-2Log likelihood
|
Cox &
|
|
|
1
|
21.84 (a)
|
.482
|
.643
|
Estimation terminated at iteration number 7 because
parameter estimates changed by less than .001.
TABLE
3 Hosmer and Lemeshow Test
|
Step
|
Chi-Square
|
df
|
Sig
|
|
1
|
4.412
|
8
|
.818
|
The
lack of significance of the Chi-Squared test indicates that the model is a good
fit
TABLE
4 Classification Table
|
Observed
|
Predicted
|
||
|
DrugReaction
|
Percentage
Correct
|
||
|
0 1
|
|||
|
Step
1 DrugReaction
Overall Percentage
|
0
1
|
11 4
2 13
|
73.3
86.7
80.0
|
The cut value is .500.
The classification table shows that
the model makes a correct prediction 80% of the time overall. Of the 15 women
with no reaction, the model correctly identified 11 of them as not likely to
have one. Similarly, of the 15 who did have a reaction, the model correctly
identifies 13 as likely to have one.
TABLE
5 Variables in the Equation
|
|
B
|
S.E.
|
Wald
|
df
|
Sig
|
Exp (B)
|
|
Step 1 (a) BP
|
-.018
|
.27
|
.463
|
1
|
.496
|
.982
|
|
Cholesterol
|
.027
|
.025
|
1.182
|
1
|
.277
|
1.027
|
|
Age
|
.265
|
.114
|
5.404
|
1
|
.20
|
1.304
|
|
Pregnant
|
8.501
|
3.884
|
4.790
|
1
|
0.29
|
4918.147
|
|
Constant
|
-17.874
|
10.158
|
3.096
|
1
|
0.78
|
.000
|
Variable(s) entered on Step 1: BP, Cholesterol, Age,
Pregnant.
Since BP and Cholesterol show up as not significant, one can
try to run the regression again without those variables to see how it impacts
the prediction accuracy. Since the sample size is low, one cannot assume that
they are insignificant. Wald’s test is best suited to large sample sizes.
The prediction equation is:
Log (odds of a reaction to drug) =
─17.874─0.018(BP) + (Cholesterol) + 0.265 (Age) + 8.501 (Pregnant)
As with any regression, the positive
coefficients indicate a positive relationship with the dependent variable.
TABLE
6 Predicted Probabilities and Classification
|
BP
|
Cholesterol
|
Age
|
Pregnant
|
Drug Reaction
|
Pred_Prob
|
Pred_Class
|
|
100
|
150
|
20
|
0
|
0
|
.00003
|
0
|
|
120
|
160
|
16
|
0
|
0
|
.00001
|
0
|
|
110
|
150
|
18
|
0
|
0
|
.00002
|
0
|
|
100
|
175
|
25
|
0
|
0
|
.00023
|
0
|
|
95
|
250
|
36
|
0
|
0
|
.03352
|
0
|
|
110
|
200
|
56
|
0
|
0
|
.58319
|
1
|
|
120
|
180
|
59
|
0
|
0
|
.60219
|
1
|
|
150
|
175
|
45
|
0
|
0
|
.01829
|
0
|
|
160
|
185
|
40
|
0
|
0
|
.00535
|
0
|
|
125
|
195
|
20
|
1
|
0
|
.24475
|
0
|
|
135
|
190
|
18
|
1
|
0
|
.12197
|
0
|
|
165
|
200
|
25
|
1
|
0
|
.40238
|
0
|
|
145
|
175
|
30
|
1
|
0
|
.65193
|
1
|
|
120
|
180
|
28
|
1
|
0
|
.66520
|
1
|
|
100
|
180
|
21
|
1
|
0
|
.30860
|
0
|
|
100
|
160
|
19
|
1
|
1
|
.13323
|
0
|
|
95
|
250
|
18
|
1
|
1
|
.58936
|
1
|
|
120
|
200
|
30
|
1
|
1
|
.85228
|
1
|
|
125
|
240
|
29
|
1
|
1
|
.92175
|
|
|
130
|
172
|
30
|
1
|
1
|
.69443
|
1
|
|
120
|
130
|
35
|
1
|
1
|
.76972
|
1
|
|
120
|
140
|
38
|
1
|
1
|
.90642
|
1
|
|
125
|
160
|
32
|
1
|
1
|
.75435
|
1
|
|
115
|
185
|
40
|
1
|
1
|
.98365
|
1
|
|
150
|
195
|
65
|
0
|
1
|
.86545
|
1
|
|
130
|
175
|
72
|
0
|
1
|
.97205
|
1
|
|
170
|
200
|
56
|
0
|
1
|
.31892
|
0
|
|
145
|
210
|
58
|
0
|
1
|
.62148
|
1
|
|
180
|
200
|
81
|
0
|
1
|
.99665
|
1
|
|
140
|
190
|
73
|
0
|
1
|
.98260
|
1
|
The table
above shows the predicted probabilities of an adverse reaction, and the
classification of each into group 0 or 1 on the basis of that probability,
using 0.5 as the cut-off score.
Question:
Case 4: Using
logistic regression proof that particular drug for women has characteristics
that cause some of them an adverse reaction to a particular drug.
CASE – 5
Conjoint Analysis
Problem
XYZ
paint company identified the attributes which are important to their customers
and also classified each of the attributes into their levels. Based on this, they
want to use the technique of conjoint analysis to determine from a potential
customer’s point of view, how important each attribute is to him. They also
want to know how much utility the customer derives from a given combination of
these levels of attributes. It also helps to understand the feasible offerings
from the marketer’s point of view. The three important attributes identified
for the paint are:
1.
Life—this is the number of years the paint coat lasts.
2.
Price—the price of one litre of paint.
3.
Colour—the colour of paint.
The levels of the above mentioned
attributes are as follows:
·
Life—3 years, 4 years, 5 years
·
Price—Rs. 50 per litre, Rs. 60 per litre, Rs. 70 per litre
·
Colour—Green, Blue, Cream
Input data
After the attributes and their levels are decided,
the next stage is to collect from the respondent, the ranking of all 27
combinations of levels. This can be seen from Table 1.1.
TABLE 1.1
Input Data for Conjoint Analysis
|
S.No.
|
Life (in years)
|
Price (Rs/Litre)
|
Colour
|
Rating (27 to 10
|
|
1
|
5
|
50
|
Green
|
27
|
|
2
|
4
|
50
|
Green
|
26
|
|
3
|
5
|
50
|
Cream
|
25
|
|
4
|
5
|
50
|
Blue
|
24
|
|
5
|
5
|
60
|
Green
|
23
|
|
6
|
4
|
60
|
Green
|
22
|
|
7
|
5
|
70
|
Green
|
21
|
|
8
|
5
|
60
|
Blue
|
20
|
|
9
|
5
|
60
|
Cream
|
19
|
|
10
|
4
|
50
|
Blue
|
18
|
|
11
|
4
|
50
|
Cream
|
17
|
|
12
|
5
|
70
|
Blue
|
16
|
|
13
|
3
|
50
|
Green
|
15
|
|
14
|
5
|
70
|
Cream
|
14
|
|
15
|
3
|
50
|
Blue
|
13
|
|
16
|
4
|
60
|
Blue
|
12
|
|
17
|
4
|
60
|
Cream
|
11
|
|
18
|
3
|
50
|
Cream
|
10
|
|
19
|
4
|
70
|
Green
|
9
|
|
20
|
3
|
60
|
Green
|
8
|
|
21
|
4
|
70
|
Blue
|
7
|
|
22
|
3
|
60
|
Blue
|
6
|
|
23
|
4
|
70
|
Cream
|
5
|
|
24
|
3
|
60
|
Cream
|
4
|
|
25
|
3
|
70
|
Green
|
3
|
|
26
|
3
|
70
|
Blue
|
2
|
|
27
|
3
|
70
|
Cream
|
1
|
Table 1.2
Shows different codes assumed for various levels of attributes for a
regression run. The coding of the attribute levels for this purpose is known as
‘effects coding’. In this table, which is similar to the coding of dummy
variables, the three levels of life are coded as follows:
|
Life in years
|
Var 1
|
Var 2
|
|
3
|
1
|
0
|
|
4
|
0
|
1
|
|
5
|
─1
|
─1
|
Thus,
the two variables, Var 1 and Var 2 are used to indicate the 3 levels of life,
as per the coding scheme mentioned above.
Similarly
the coding scheme for the three levels of the price is as shown as follows:
|
Price
(Rs. Per liter)
|
Var 3
|
Var 4
|
|
50
|
1
|
0
|
|
60
|
0
|
1
|
|
70
|
─1
|
─1
|
Finally, the coding scheme for colour
is as shown below:
|
Colour
|
Var 3
|
Var 4
|
|
Green
|
1
|
0
|
|
Blue
|
0
|
1
|
|
Cream
|
─1
|
─1
|
Thus, 6 variables, that is Var 1
─ Var 6 are used to
represent the 3 levels of life of the paint (3, 4, 5), 3 levels of price per
litre (50, 60 & 70) and 3 levels of colour (green, blue and cream). All the
six variables are independent variables in the regression run. Var 7 is the
rating of each combination given by the respondent, and forms the dependent
variable for the regression curve. The recoded input data are shown in Table
1.3.
If the conjoint analysis is run as a
regression model, the rating (which is the reverse of ranking) is used as a
dependent variable. All combinations from the first to the twenty-seventh are
ranked by the respondent. Rank 1 can be considered as the highest rating and
given a rating of 27. Rank 2 can be given a rating of 26 and so on. This is not
an interval-scaled rating, and should have only ordinal interpretation.
Table
1.3 Conjoint Problem Input Data Coded
for Regression
|
Var 1
|
Var 2
|
Var 3
|
Var 4
|
Var 5
|
Var 6
|
Var 7
|
|
─1.00
|
─1.00
|
1.00
|
0.00
|
1.00
|
0.00
|
27.00
|
|
0.00
|
1.00
|
1.00
|
0.00
|
1.00
|
0.00
|
26.00
|
|
─1.00
|
─1.00
|
1.00
|
0.00
|
─1.00
|
─1.00
|
25.00
|
|
─1.00
|
─1.00
|
1.00
|
0.00
|
0.00
|
1.00
|
24.00
|
|
─1.00
|
─1.00
|
0.00
|
1.00
|
1.00
|
0.00
|
23.00
|
|
0.00
|
1.00
|
0.00
|
1.00
|
1.00
|
0.00
|
22.00
|
|
─1.00
|
─1.00
|
─1.00
|
─1.00
|
1.00
|
0.00
|
21.00
|
|
─1.00
|
─1.00
|
0.00
|
1.00
|
0.00
|
1.00
|
20.00
|
|
─1.00
|
─1.00
|
0.00
|
1.00
|
─1.00
|
─1.00
|
19.00
|
|
0.00
|
1.00
|
1.00
|
0.00
|
0.00
|
1.00
|
18.00
|
|
0.00
|
1.00
|
1.00
|
0.00
|
─1.00
|
─1.00
|
17.00
|
|
─1.00
|
─1.00
|
─1.00
|
─1.00
|
0.00
|
1.00
|
16.00
|
|
1.00
|
0.00
|
1.00
|
0.00
|
1.00
|
0.00
|
15.00
|
|
─1.00
|
─1.00
|
─1.00
|
─1.00
|
─1.00
|
─1.00
|
14.00
|
|
1.00
|
0.00
|
1.00
|
0.00
|
0.00
|
1.00
|
13.00
|
|
0.00
|
1.00
|
0.00
|
1.00
|
0.00
|
1.00
|
12.00
|
|
0.00
|
1.00
|
0.00
|
1.00
|
─1.00
|
─1.00
|
11.00
|
|
1.00
|
0.00
|
1.00
|
0.00
|
─1.00
|
─1.00
|
10.00
|
|
0.00
|
1.00
|
─1.00
|
─1.00
|
1.00
|
0.00
|
9.00
|
|
1.00
|
0.00
|
0.00
|
1.00
|
1.00
|
0.00
|
8.00
|
|
0.00
|
1.00
|
─1.00
|
─1.00
|
0.00
|
1.00
|
7.00
|
|
1.00
|
0.00
|
0.00
|
1.00
|
0.00
|
1.00
|
6.00
|
|
0.00
|
1.00
|
─1.00
|
─1.00
|
─1.00
|
─1.00
|
5.00
|
|
1.00
|
0.00
|
0.00
|
1.00
|
─1.00
|
─1.00
|
4.00
|
|
1.00
|
0.00
|
─1.00
|
─1.00
|
1.00
|
0.00
|
3.00
|
|
1.00
|
0.00
|
─1.00
|
─1.00
|
0.00
|
1.00
|
2.00
|
|
1.00
|
0.00
|
─1.00
|
─1.00
|
─1.00
|
─1.00
|
1.00
|
OUTPUT AND ITS INTERPRETATION
The
output of the regression model is shown in Table 1.4. Variables 1 to 6 are
treated as independent variables. The column titled ‘B’ (the regression
coefficient column) provides the part utility of each level of attributes.
Table
1.4 Multiple regression output for
conjoint problem (partial output shown)
|
Variables in the regression equation
|
|
|
VARIABLE
|
B
|
|
Var 1
|
─7.00
|
|
Var 2
|
0.11
|
|
Var 3
|
5.44
|
|
Var 4
|
─0.11
|
|
Var 5
|
3.11
|
|
Var 6
|
─0.88
|
For example, the life
of 3 years is represented by variable 1 as per our coding scheme. Its
utility is equal to ─7.11 (looking under column ‘B’ of Table 1.4 for variable
1). Similarly the utility for variable 2, representing life of 4 years is 0.11.
The utility for the 3rd level of life, is not in the table, but is derived from
the property of this coding, that all the utilities for a given attributes
should sum to 0. Thus, utility for life
of 5 years should be equal to 7 (─7.11
+ 0.11).
Similarly for price, the utilities of Rs. 50/litre and Rs.
70/litre are given by the numbers 5.44 and ─0.11, as shown against 3 and 4 in
Table 1.4 in Table 1.4 but the utility for Rs. 80/litre is derived from the
same property, that the sum of the utilities for different levels of price
should sum to 0. Therefore the price Rs. 80/litre has the utility of 5.33 (5.44
+ (─0.11).
Finally for colour, green has the utility of 3.11 and blue
has the utility of ─0.88. Cream has a derived utility of 2.23 (3.11 + (─0.88).
TABLE 1.5 Utilities Table for Conjoint Analysis
|
Attributes
|
Levels
|
Part Utility
|
(Max ─ Min)
|
|
|
Life
|
3 years
|
─7.11
|
= 7.00 ─ (─7.11)
|
|
|
|
4 years
|
0.11
|
= 14.11
|
|
|
|
5 years
|
7.00
|
|
|
|
Price
|
Rs. 50/litre
|
5.44
|
|
|
|
|
Rs. 60/litre
|
─0.11
|
= 5.44 ─ (─0.11)
|
|
|
|
Rs. 70/litre
|
5.33
|
= 5.55
|
|
|
Colour
|
Green
|
3.11
|
= 3.11 ─ (─0.88)
|
|
|
|
Blue
|
─0.88
|
= 3.99
|
|
|
|
Cream
|
2.23
|
|
|
From the Table 1.5 we can conclude
that the life or the number of years the paint lasts is the most important
attribute for the customer. There are two indicators for this.
1. The range of utility value
is highest (14.11) for the life. (From Range of Utility
2. The highest individual value
of this attributes is at its 3rd level that is, i.e., 7.00.
Both these figures indicate that the number of years
the paint lasts is the most important attribute at given levels of attributes.
The price/litre seems to be the second most important attribute, as its range
of utilities is 5.55. The last attribute in relative importance is the colour,
with the utility range of 3.99.
Combination Utilities
The
total utility of any combination can be calculated by picking the attribute
levels of our choice. For example, the combined utility of the combination 4
years of life, Rs. 70/litre, and cream colour is 0.11 + 5.33 + 2.33 = 7.67. If
we want to know the best combination, it is advisable to pick the highest
utilities from each attribute, and add them. The possible combination is 5
years of life, Rs. 50/litre, and green colour, that is, 7.00 + 5.44 + 3.11 =
15.55. The next best combination is 5 years of life, Rs. 70/litre, and green
colour, with the combined utility of 7 + 5.33 + 3.11 = 15.44.
Individual Attributes
The
difference in utility with the change of one level in one attribute can also be
checked. For the life of 3 years to 4 years, there is increase in utility value
of 7.22 units, but the next level, that is, 4 years to 5 years has an increase
in utility of 6.89.
Similarly, increase in price from Rs.
50/litre to Rs. 60/litre induces a utility drop of 5.55, whereas from Rs.
60/litre to Rs. 70/litre there is an increase in utility of 5.44.
Finally, colour green to colour blue
induces 3.99 drop in utility. Next, from colour blue to colour cream there is
an increase in utility of 3.11.
Question:
Case 5: Use conjoint analysis to determine from a potential customer’s point of view, how important each attribute is to him. Also determine how much utility the customer derives from a given combination of these levels of attributes. The attributes are life, price and colour.
CASE 6
A recent case study for a
cellular phone service provider in Chennai listed its research objectives and
methodology (including sampling plan) for a marketing research study as
follows:
SKCELL,
A CELLULAR OPERATOR/STUDY ON VALUE ADDED SERVICES LIKE SMS (SHORT MESSAGING
SERVICE), VOICE MAIL, AND SO ON
Research Objectives
To find out
·
whether people actually use the mobile
phone just for talking
·
to what extent the mobile phone is used
for its VAS (Value Added Services)
·
factors influencing choice of service
provider
·
awareness of Skycell’s improved coverage
Locations Covered
Chennai city and the suburbs
Methodology
Primary data:
Through questionnaires
Sample Composition
·
Mobile phone users
·
Business pesons
·
Executives
·
Youth
Sample size: 75
Age group: 18 – 45 years
Questions:
1.
Can you add to methodology section?
2.
Distribute the sample of 75 among the
different categories of respondents mentioned under “Sample Composition”.
No comments:
Post a Comment