MA Assignment (MDS) : Section A - Group 03

You might also like

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 10

MA Assignment (MDS)

Section A - Group 03

Soumak Kumar Tah 180103221


Subhra Mukherjee 180103229
Saurabh Lalu Nair 180103189
Tanuj Anand 180103238
Utkarsh Anand 180103244
Context
• You are the product manager of Samsung consumer electronics group

• To identify how Samsung is perceived with respect to its competitors, It is decided to conduct
perceptual mapping (using MDS technique)

• The exercise involves Samsung and another 9 of its competitors


Available Dataset
• The respondents (100 in number) are asked to rate the similarity of each combination of two
brands (Samsung-A; A-B) in a scale of 1 to 7. Here 1 indicates very less similarity and 7 indicates
high similarity

• The average similarity rating score of each of this 45 combination is provided in the excel

• The same set of respondents are also asked to rate these 10 brands on 8 attributes. The average
rating of each of these brands on each attributes are provided

• In the case of brand rating based on attributes, 9 indicates excellent and 1 poor
Data Preparation
• Covert the proximity matrix to distance matrix by reverse coding in SPSS
• Subtract each of the proximity value from 8, do not modify the cells where the value is 0

Name Samsung A B C D E F G H I
Samsung 0.00 1.39 2.06 5.67 5.44 3.94 5.50 5.67 5.56 1.83
A 1.39 0.00 2.07 5.39 5.44 5.61 4.50 5.61 3.06 1.06
B 2.06 2.07 0.00 4.56 3.89 5.83 4.00 4.28 1.39 5.17
C 5.67 5.39 4.56 0.00 1.06 3.94 5.78 5.33 5.50 5.50
D 5.44 5.44 3.89 1.06 0.00 5.61 5.83 5.39 0.94 5.50
E 3.94 5.61 5.83 3.94 5.61 0.00 3.94 4.33 2.39 4.50
F 5.50 4.50 4.00 5.78 5.83 3.94 0.00 5.72 5.17 1.06
G 5.67 5.61 4.28 5.33 5.39 4.33 5.72 0.00 5.44 5.56
H 5.56 3.06 1.39 5.50 0.94 2.39 5.17 5.44 0.00 5.61
I 1.83 1.06 5.17 5.50 5.50 4.50 1.06 5.56 5.61 0.00
2 Dimensional Model
Output

Technical Support

Salesforce image
3 Dimensional Model
Stimulus Coordinates comparison for both Model
Stress & R-Square Comparison

3 Dimensional
Model

2 Dimensional
Model
Analysis
From the output, we conclude that the 3-Dimensional model is a better representation of our dataset than the
2-Dimensional model for the following reasons:

Stress value being lower for 3D than 2D (13% < 18.8%)- Stress measures badness-of-fit, or the proportion of
variance of the optimally scaled data that is not accounted for by the MDS model (Lower the better)

R-squared value being greater in 3D model (80.6% > 73.2)- R-square is a measure of goodness-of-fit that
indicates the proportion of variance of the optimally scaled data can be accounted for by the MDS procedure
(Higher the better)

Also, using the attribute values provided in our dataset, we have assigned Sales force image as the dimension
along X-axis and Technical Support as the dimension along the Y-axis for the following 2 reasons:

Compared the values of parameters where F brand is the highest and corresponding C brands’ value is the
lowest – gave us Sales force image as the dimension along X axis

Similarly, compared the values of parameters where E brand is the highest and corresponding B brands’ value is
the lowest – gave us Technical Support as the dimension along Y axis
Thank you

You might also like