20-1

Cluster Analysis
 Cluster analysis is a class of techniques used to
classify objects or cases into relatively homogeneous
groups called clusters. Objects in each cluster tend
to be similar to each other and dissimilar to objects
in the other clusters. Cluster analysis is also called
classification analysis, or numerical taxonomy.
 Both cluster analysis and discriminant analysis are
concerned with classification. However, discriminant
analysis requires prior knowledge of the cluster or
group membership for each object or case included,
to develop the classification rule. In contrast, in
cluster analysis there is no a priori information about
the group or cluster membership for any of the
objects. Groups or clusters are suggested by the
data, not defined a priori.
 20-2

How Cluster Analysis is used in MR

 Segmenting Market : to group customers on the basis of
benefits sought from the purchase of a product.
 Understanding Consumer behavior: to identify
homogeneous groups of buyers-respondents are grouped on
the basis of self-reported importance attached to each factor
for making a purchase.Psychographically,Geographically
 Identifying New product opportunities: by grouping brands
and products, competitive sets within the market can be
determined.
 Selecting test market:Cities can be grouped into
homogeneous clusters to select test market
 Reducing Data.:Clusters can be used as unit of analysis
 20-3

How Cluster Analysis is used in MR

A Foreign market analysis study of 71 countries.


 A perfermance-profile clustering of digital computer

 A product positioning study involving sports car brands.

Example:Segmenting 12 bank branches into three clusters according

to characterstics of no. of men who have borrowed money (x1)
and the number of women who have borrowed money(x2).
•8
No of women borrowers

•10
•1 •2
•11 •12 •6
•3 •5 Branches
•7 •4
•9

No. of men borrowers

 20-4

An Ideal Clustering Situation

Fig. 20.1

Variable 1

Variable 2
 20-5

A Practical Clustering Situation

Fig. 20.2

Variable 1

X
Variable 2
 20-6

Statistics Associated with Cluster Analysis

 Agglomeration schedule. An agglomeration
schedule gives information on the objects or cases
being combined at each stage of a hierarchical
clustering process.
 Cluster centroid. The cluster centroid is the mean
values of the variables for all the cases or objects in
a particular cluster.
 Cluster centers. The cluster centers are the initial
starting points in nonhierarchical clustering. Clusters
are built around these centers, or seeds.
 Cluster membership. Cluster membership
indicates the cluster to which each object or case
belongs.
 20-7

Statistics Associated with Cluster Analysis

 Dendrogram. A dendrogram, or tree graph, is a
graphical device for displaying clustering results.
Vertical lines represent clusters that are joined
together. The position of the line on the scale
indicates the distances at which clusters were joined.
The dendrogram is read from left to right. Figure
20.8 is a dendrogram.
 Distances between cluster centers. These
distances indicate how separated the individual pairs
of clusters are. Clusters that are widely separated
are distinct, and therefore desirable.
 20-8

Statistics Associated with Cluster Analysis

 Icicle diagram. An icicle diagram is a graphical
display of clustering results, so called because it
resembles a row of icicles hanging from the eaves of
a house. The columns correspond to the objects
being clustered, and the rows correspond to the
number of clusters. An icicle diagram is read from
bottom to top. Figure 20.7 is an icicle diagram.
 Similarity/distance coefficient matrix. A
similarity/distance coefficient matrix is a lower-
triangle matrix containing pairwise distances between
objects or cases.
Conducting Cluster Analysis
20-9

Formulate the Problem

 Perhaps the most important part of formulating the
clustering problem is selecting the variables on which
the clustering is based.
 Inclusion of even one or two irrelevant variables may
distort an otherwise useful clustering solution.
 Basically, the set of variables selected should
describe the similarity between objects in terms that
are relevant to the marketing research problem.
 The variables should be selected based on past
research, theory, or a consideration of the
hypotheses being tested. In exploratory research,
the researcher should exercise judgment and
intuition.
Conducting Cluster Analysis
20-10

Select a Distance or Similarity Measure

 The most commonly used measure of similarity is the Euclidean
distance or its square. The Euclidean distance is the square
root of the sum of the squared differences in values for each
variable. Other distance measures are also available. The city-
block or Manhattan distance between two objects is the sum of
the absolute differences in values for each variable. The
Chebychev distance between two objects is the maximum
absolute difference in values for any variable.
 If the variables are measured in vastly different units, the
clustering solution will be influenced by the units of
measurement. In these cases, before clustering respondents,
we must standardize the data by rescaling each variable to have
a mean of zero and a standard deviation of unity. It is also
desirable to eliminate outliers (cases with atypical values).
 Use of different distance measures may lead to different
clustering results. Hence, it is advisable to use different
measures and compare the results.
Conducting Cluster Analysis
20-11

Select a Clustering Procedure – Hierarchical

 Hierarchical clustering is characterized by the
development of a hierarchy or tree-like structure.
Hierarchical methods can be agglomerative or divisive.
 Agglomerative clustering starts with each object in
a separate cluster. Clusters are formed by grouping
objects into bigger and bigger clusters. This process is
continued until all objects are members of a single
cluster.
 Divisive clustering starts with all the objects
grouped in a single cluster. Clusters are divided or
split until each object is in a separate cluster.
 Agglomerative methods are commonly used in
marketing research. They consist of linkage methods,
error sums of squares or variance methods, and
centroid methods.
 20-12

Agglomerative

01
02
Observation No.

03
04
05

06
07
08

0
1 2 3 4 5 6 7
Divisive
Conducting Cluster Analysis
20-13

Select a Clustering Procedure – Linkage Method

 The single linkage method is based on minimum
distance, or the nearest neighbor rule. At every stage,
the distance between two clusters is the distance
between their two closest points (see Figure 20.5).
 The complete linkage method is similar to single
linkage, except that it is based on the maximum
distance or the furthest neighbor approach. In
complete linkage, the distance between two clusters is
calculated as the distance between their two furthest
points.
 The average linkage method works similarly.
However, in this method, the distance between two
clusters is defined as the average of the distances
between all pairs of objects, where one member of the
pair is from each of the clusters (Figure 20.5).
 20-14

Linkage Methods of Clustering

Fig. 20.5 Single Linkage
Minimum Distance

Cluster 1 Cluster 2
Complete Linkage
Maximum
Distance

Cluster 1 Cluster 2
Average Linkage

Average Distance
Cluster 1 Cluster 2
20-15
 20-16

Longest

Shortest

Complete Linkage
Single Linkage
Conducting Cluster Analysis
20-17

Select a Clustering Procedure – Variance Method

 The variance methods attempt to generate clusters to
minimize the within-cluster variance.
 A commonly used variance method is the Ward's
procedure. For each cluster, the means for all the
variables are computed. Then, for each object, the
squared Euclidean distance to the cluster means is
calculated (Figure 20.6). These distances are summed for
all the objects. At each stage, the two clusters with the
smallest increase in the overall sum of squares within
cluster distances are combined.
 In the centroid methods, the distance between two
clusters is the distance between their centroids (means for
all the variables), as shown in Figure 20.6. Every time
objects are grouped, a new centroid is computed.
 Of the hierarchical methods, average linkage and Ward's
methods have been shown to perform better than the
other procedures.
 20-18

Other Agglomerative Clustering Methods

Fig. 20.6
Ward’s Procedure

Centroid Method
Conducting Cluster Analysis
20-19

Select a Clustering Procedure – Nonhierarchical

 The nonhierarchical clustering methods are frequently
referred to as k-means clustering. These methods include
sequential threshold, parallel threshold, and optimizing
partitioning.
 In the sequential threshold method, a cluster center is
selected and all objects within a prespecified threshold value
from the center are grouped together. Then a new cluster
center or seed is selected, and the process is repeated for the
unclustered points. Once an object is clustered with a seed, it
is no longer considered for clustering with subsequent seeds.
 The parallel threshold method operates similarly, except
that several cluster centers are selected simultaneously and
objects within the threshold level are grouped with the nearest
center.
 The optimizing partitioning method differs from the two
threshold procedures in that objects can later be reassigned to
clusters to optimize an overall criterion, such as average within
cluster distance for a given number of clusters.
Conducting Cluster Analysis
20-20

Select a Clustering Procedure

 It has been suggested that the hierarchical and
nonhierarchical methods be used in tandem. First,
an initial clustering solution is obtained using a
hierarchical procedure, such as average linkage or
Ward's. The number of clusters and cluster centroids
so obtained are used as inputs to the optimizing
partitioning method.
 Choice of a clustering method and choice of a
distance measure are interrelated. For example,
squared Euclidean distances should be used with the
Ward's and centroid methods. Several
nonhierarchical procedures also use squared
Euclidean distances.
Conducting Cluster Analysis
20-21

Decide on the Number of Clusters

 Theoretical, conceptual, or practical considerations
may suggest a certain number of clusters.
 In hierarchical clustering, the distances at which
clusters are combined can be used as criteria. This
information can be obtained from the agglomeration
schedule or from the dendrogram.
 In nonhierarchical clustering, the ratio of total within-
group variance to between-group variance can be
plotted against the number of clusters. The point at
which an elbow or a sharp bend occurs indicates an
appropriate number of clusters.
 The relative sizes of the clusters should be
meaningful.
Conducting Cluster Analysis
20-22

Interpreting and Profiling the Clusters

 Interpreting and profiling clusters involves examining
the cluster centroids. The centroids enable us to
describe each cluster by assigning it a name or label.
 It is often helpful to profile the clusters in terms of
variables that were not used for clustering. These
may include demographic, psychographic, product
usage, media usage, or other variables.
Conducting Cluster Analysis
20-23

Assess Reliability and Validity

1. Perform cluster analysis on the same data using different
distance measures. Compare the results across
measures to determine the stability of the solutions.
2. Use different methods of clustering and compare the
results.
3. Split the data randomly into halves. Perform clustering
separately on each half. Compare cluster centroids
across the two subsamples.
4. Delete variables randomly. Perform clustering based on
the reduced set of variables. Compare the results with
those obtained by clustering based on the entire set of
variables.
5. In nonhierarchical clustering, the solution may depend
on the order of cases in the data set. Make multiple
runs using different order of cases until the solution
stabilizes.
 20-24

Clustering Variables
 In this instance, the units used for analysis are the
variables, and the distance measures are computed
for all pairs of variables.
 Hierarchical clustering of variables can aid in the
identification of unique variables, or variables that
make a unique contribution to the data.
 Clustering can also be used to reduce the number of
variables. Associated with each cluster is a linear
combination of the variables in the cluster, called the
cluster component. A large set of variables can often
be replaced by the set of cluster components with
little loss of information. However, a given number
of cluster components does not generally explain as
much variance as the same number of principal
components.
 20-25

Conducting Cluster Analysis

Fig. 20.3
Formulate the Problem

Select a Distance Measure

Select a Clustering Procedure

Decide on the Number of Clusters

Interpret and Profile Clusters

Assess the Validity of Clustering

 20-26

Student Score   Ajit Balu Chandra Dilip

Ajit 11 Ajit
Balu 11 Balu        
Chandra 13 Chandra        
Dilip 18 Dilip      

  Ajit Balu Chandra Dilip

Ajit 0 0 2 7

Balu  0 0 2 7 

Chandra  2 2  0  5

Dilip  7 7  5  0
 20-27

  Ajit Balu Chandra Dilip

Ajit 0 0 2 7

Balu  0 0 2 7 

Chandra  2 2  0  5

Dilip  7 7  5  0

  Ajit Balu Chandra Dilip

Ajit 0 0 4 49

Balu  0 0 4 49 

Chandra  4 4  0 25

Dilip  49 49 25 0
 20-28

X1 X2 X3 X4 X5

R1

R2

R3

R4

R1 R2 R3 R4

R1

R2

R3

R4
 20-29

X1 X2

R1 25 11
R2 33 11
R3 34 13
. R4 35 18

  R1 R2 R3 R4   R1 R2 R3 R4
R1 0 64 81 100 R1 0 0 4 49
R2 64 0 1 4 R2 0 0 4 49
R3 81 1 0 1 R3 4 4 0 25
R4 100 4 1 0
+ R4 49 49 25 0

  R1 R2 R3 R4
R1 0 64 85 149
R2 64 0 5 53
= R3 85 5 0 26
R4 149 53 26 0
 20-30

Ajit
Balu
Chandra
Dilip
Eswar
Farook
Govind
Hari
Indira
Kumar
 20-31

  Ajit Balu Chandra Dilip Esawar Farook

. Ajit 0 4 36 81 196 225

Balu 4 0 16 49 144 169

Chandra 36 16 0 9 64 81

Dilip 81 49 9 0 25 36

Eswar 196 144 64 25 0 1

Farook 225 169 81 36 1 0

  Ajit Balu Chandra Dilip Esawar

Ajit 0 4 36 81 196

Balu 4 0 16 49 144

Chandra 36 16 0 9 64

Dilip 81 49 9 0 25

Eswar & 196 144 64 25 0

Farook
 20-32

  Ajit Balu Chandra Dilip Esawar

Ajit 0 4 36 81 196

Balu 4 0 16 49 144

Chandra 36 16 0 9 64

Dilip 81 49 9 0 25

Eswar & 196 144 64 25 0

Farook

  Ajit & Chandra Dilip Esawar

Balu
Ajit & Balu 0 16 49 144

Chandra 16 0 9 64

Dilip 49 9 0 25

Eswar & 144 64 25 0

Farook
 20-33

.
  Ajit & Balu Chandra & Dilip Eswar & Farook

Ajit & Balu 0 16 144

Chandra & Dilip 16 0 25

Eswar & Farook 144 25 0

 20-34

Respondents
Clustering
Varaibles
A B C D E F G
V1 3 4 4 2 6 7 6
V2 2 5 7 7 6 7 4
 Scatterplot
10
9
8

D C F
7

E
6

V2
B
5
4

G
3

A
2
1
0

0 1 2 3 4 5 6 7 8 9 10
V1
 20-36

Observation

Observation A B C D E F G

A
B 3.162
C 5.099 2.000
D 5.099 2.828 2.000
E 5.000 2.236 2.236 4.123
F 6.403 3.606 3.000 5.000 1.414
G 3.606 2.306 3.606 5.000 2.000 3.162
 20-37

Agglomeration Process Cluster solution

Minimum Overall Similarity
Distance b/w Measure(Average
Unclustered Observation Within-Cluster
Observations Pair Cluster No. of Distance)
Step Membership Clusters

Initial Solution (A) (B) (C) (D) (E) (F) (G) 7 0

1 1.414 E-F (A) (B) (C) (D) (E-F) (G) 6 1.414

2 2.000 E-G (A) (B) (C) (D) (E-F-G) 5 2.192
3 2.000 C-D (A) (B) (C-D) (E-F-G) 4 2.144
4 2.000 B-C (A) (B-C-D) (E-F-G) 3 2.234
5 2.236 B-E (A) (B-C-D-E-F-G) 2 2.896
6 3.162 A-B (A-B-C-D-E-F-G) 1 3.420
 20-38

10
9
8

D 3 C 1 F
7

E
6

4 2
B
5
4

G
5
3

A 6
2
1
0

0 1 2 3 4 5 6 7 8 9 10
 Observation 20-39

Observation A B C D E F G

A
B 3.162
C 5.099 2.000
D 5.099 2.828 2.000
E 5.000 2.236 2.236 4.123
F 6.403 3.606 3.000 5.000 1.414
G 3.606 2.306 3.606 5.000 2.000 3.162
0 1 2 3 4 5 6 7 8 9 10

D 3 C 1F
E
4 B 2
G
5
A 6

0 1 2 3 4 5 6 7 8 9 10
 A
B
4 6
C
Observation

3
D
5
E
1
F 2
G

0 1 2 3 4
Distance at Combination
 Problem: Page 590
Clustering of consumers based on attitudes
towards shopping:
Six attitudinal variables and 20 respondents
V1: Shopping is a fun
V2:Shopping is bad for your budget
V3:I combine shopping with eating out
V4: I try to get the best buys when shopping
V5: I do not care about shopping
V6: You can save a lot of money by
comparing prices.
 20-42

Attitudinal Data For Clustering

Table 20.1
Case No. V1 V2 V3 V4 V5 V6
1 6 4 7 3 2 3
2 2 3 1 4 5 4
3 7 2 6 4 1 3
4 4 6 4 5 3 6
5 1 3 2 2 6 4
6 6 4 6 3 3 4
7 5 3 6 3 3 4
8 7 3 7 4 1 4
9 2 4 3 3 6 3
10 3 5 3 6 4 6
11 1 3 2 3 5 3
12 5 4 5 4 2 4
13 2 2 1 5 4 4
14 4 6 4 6 4 7
15 6 5 4 2 1 4
16 3 5 4 6 4 7
17 4 4 7 2 2 5
18 3 7 2 6 4 3
19 4 6 3 7 2 7
20 2 3 2 4 7 2
 20-43

Results of Hierarchical Clustering

Table 20.2

Agglomeration Schedule Using Ward’s Procedure

Stage cluster
Clusters combined first appears
Stage Cluster 1 Cluster 2 Coefficient Cluster 1 Cluster 2 Next stage
1 14 16 1.000000 0 0 6
2 6 7 2.000000 0 0 7
3 2 13 3.500000 0 0 15
4 5 11 5.000000 0 0 11
5 3 8 6.500000 0 0 16
6 10 14 8.160000 0 1 9
7 6 12 10.166667 2 0 10
8 9 20 13.000000 0 0 11
9 4 10 15.583000 0 6 12
10 1 6 18.500000 6 7 13
11 5 9 23.000000 4 8 15
12 4 19 27.750000 9 0 17
13 1 17 33.100000 10 0 14
14 1 15 41.333000 13 0 16
15 2 5 51.833000 3 11 18
16 1 3 64.500000 14 5 19
17 4 18 79.667000 12 0 18
18 2 4 172.662000 15 17 19
19 1 2 328.600000 16 18 0
 20-44

Results of Hierarchical Clustering

Table 20.2 cont.
Cluster Membership of Cases Using Ward’s Procedure
Number of Clusters
Label case 4 3 2

1 1 1 1
2 2 2 2
3 1 1 1
4 3 3 2
5 2 2 2
6 1 1 1
7 1 1 1
8 1 1 1
9 2 2 2
10 3 3 2
11 2 2 2
12 1 1 1
13 2 2 2
14 3 3 2
15 1 1 1
16 3 3 2
17 1 1 1
18 4 3 2
19 3 3 2
20 2 2 2
 20-45

Cluster Centroids
Table 20.3

Means of Variables

Cluster No. V1 V2 V3 V4 V5 V6

1 5.750 3.625 6.000 3.125 1.750 3.875

2 1.667 3.000 1.833 3.500 5.500 3.333

3 3.500 5.833 3.333 6.000 3.500 6.000

 20-46

Results of Nonhierarchical Clustering

Cluster Membership
Table 20.4 cont.
Case Number Cluster Distance
1 3 1.414
2 2 1.323
3 3 2.550
4 1 1.404
5 2 1.848
6 3 1.225
7 3 1.500
8 3 2.121
9 2 1.756
10 1 1.143
11 2 1.041
12 3 1.581
13 2 2.598
14 1 1.404
15 3 2.828
16 1 1.624
17 3 2.598
18 1 3.555
19 1 2.154
20 2 2.102
 20-47

Dendrogram Using Ward’s Method

Fig. 20.8
 20-48

Vertical Icicle Plot Using Ward’s Method

Fig. 20.7
 20-49

Results of Nonhierarchical Clustering

Table 20.4 cont.

Final Cluster Centers

Cluster
1 2 3
V1 4 2 6
V2 6 3 4
V3 3 2 6
V4 6 4 3
V5 4 6 2
V6 6 3 4

Distances between Final Cluster Centers

Cluster 1 2 3
1 5.568 5.698
2 5.568 6.928
3 5.698 6.928
 20-50

SPSS Windows
To select this procedures using SPSS for Windows click:

Analyze>Classify>Hierarchical Cluster …

Analyze>Classify>K-Means Cluster …