BRM9_G5

Multi-Dimensional
Scaling and Conjoint
Analysis
Overview
01 Multi-Dimensional Scaling

02 Conjoint Analysis

03 Practice
MULTI-DIMENSIONAL
SCALING
Multidimensional scaling
Multidimensional scaling (MDS) is a class of
procedures for representing perceptions and
preferences of respondents spatially by means of a
visual display.

The purpose of MDS is to transform consumer

judgments of similarity or preference (eg. preference for
stores or brands) into distances represented in
multidimensional space.
 Multidimensional
scaling
Perceived or psychological relationships
among stimuli are represented as geometric
relationships among points in a
multidimensional space. These geometric
representations are often called spatial maps

The example is from Pharma brands along

with key dimensions- Gentle, Price, Hard to
Swallow, Effectiveness etc
The use of MDS based on
variables' characteristics
1. The number and nature of dimensions consumers
use to perceive different brands in the marketplace

2. The positioning of current brands on these

dimensions

3. The positioning of consumers’ ideal brand on

these dimensions
Which information can we have
1. Similarity of objects
-> Assess the perception of consumers on
different brands
A survey is conducted on two dimensions:
Price and Fuel efficiency.
The distances between the points would
reflect the perceived similarities and
dissimilarities between the cars.
→ The closer the two brands are, the more
similar they are perceived to be.

Multidimensional scaling
Which information can we have
2. Market segmentation
-> Understand the position of brand in the
market
A survey is conducted on two
dimensions: Price and Sportiness
Four market segmentation: economy
sport cars, luxury sport cars, economy
family cars, and luxury family cars.

Multidimensional scaling
Some business questions
How can we identify and target potential customers for a new product or
service?
Example: Spotify uses MDS to analyze listener preferences and identify new music
genres or artists that might appeal to specific users. -> create personalized playlists
and target marketing campaigns

How can we optimize product features and pricing to appeal to specific

customer segments?
Example: Grab uses MDS to analyze customer feedback and identify key factors that
influence their satisfaction with the ride-hailing service. -> used to optimize features
such as wait times, driver ratings,.... to increase customer retention

Multidimensional scaling
CONJOINT ANALYSIS
Definition
A market research technique used
to measure how consumers make
trade-offs between different
attributes of a product or service.

Helps businesses understand the

preferences and priorities of
consumers when evaluating
different product features or
attributes.
Conjoint Analysis Process

Presenting respondents with a Respondents are then asked Researchers determine the
series of product profiles, each to rank or rate these profiles importance of each
containing a combination of based on their preferences. attribute and the ideal
attributes. combination that would
maximize consumer
satisfaction.
Situations appropriate using Conjoint Analysis

Complex Decision-Making
Product Design & Development

Trade-off Decisions
 Which information can we have ?
A small set of qualities has a big impact on what people choose or decide.

-> Screen size and camera quality

are more important -> Consumers are willing to pay significantly more for
smartphones with larger screens and higher-quality cameras.

Conjoint analysis
Some business questions
What are the relative importance weights of different attributes in influencing
consumer choice?
It helps businesses figure out which things about a product or service matter the
most to people when they're choosing what to buy.

What combination of attribute/feature levels results in the most preferred

product or service offering?
Find out the exact versions of product features or service parts that, when put
together, make customers really want it.

Conjoint analysis
 PRACTICE
Multidimensional scaling
file UnichoiceLikert
The survey collects data regarding some key attributes impacting students’
university choice including:

URE University Reputation ENV Environment

LEC Lecturer EBE Extra Benefits

PRO Training Program CON Consulting services

COS Cost INF Influence of other people

FAC Facilities SUC Satisfaction with the choice

CDE Career Development

The students are from six universities (Code) including: FTU, IU, RMIT, OU, UEH
and UEL. Likert 5-point (1 to 5) scale was used to measure all the key attributes
and the numbers presented in the dataset are the means of students’ evaluation
on each attribute for each university.

Based on this data, which meaningful information can be extracted by employing

multidimensional scaling?
 1.00 - 1.49 (rounded to 1): Strongly disagree
1.50 - 2.49 (rounded to 2): Disagree
2.50 - 3.49 (rounded to 3): Neutral
3.50 - 4.49 (rounded to 4): Agree
4.50 - 5.00 (rounded to 5): Strongly agree

Conclusion:
3.00 - 3.49 including COS, FAC, ENV, INF can be understood as these factors will not affect
their choice of school.
3.50 - 4.49 including the rest variables - which are considered to have affect their choice of
school
file UnichoicePairedcomparison
The survey collects data by asking students to judge how similar pairs of
university are in their perception (1 – very dissimilar to 5 – very similar). All
possible pairs of the six universities including FTU, IU, RMIT, OU, UEH and UEL
have been evaluated and the results of similarity rating (distance between the
two universities) are presented in the dataset.

Based on this data, which meaningful information can be extracted by employing

multidimensional scaling?
<0.10
D.A.F = 1-NRS

TCC = √DAF
OU, UEL, and IU on the second
quadrant similar in nature
UEH and FTU on the third
quadrant similar in nature
 PRACTICE
CONJOINT ANALYSIS
Stage 1
Developing product profiles to collect
data on consumers’ preferences
The new product has 5 key attributes that have different
options as follows:

Product Package 3 options including A style, B style and C style

Brand image 3 options including luxury (LUX), common (COM) and cheap (CHE)

Price 3 levels of price including 200$, 500$ and 700$

Delivery Consumers can choose home delivery service for this product or
buy it at company store.

Money-back Consumers can choose whether to participate in money-back

guarantee guarantee program for this product or not.
Help the team to identify the number of
product profile that is feasible to collect
data.
Select ProductDesign_plan
The Card list table displays sequential numbers for all
product profiles.

The Profiles table breaks down each product profile and

shows details about their attributes.
Stage 2
Launching new product to the market
In which:

Specify the ranking preferences from the most to the least. Must be
Sequence
exact variable names in the data file and no blanks in between (PREF1 TO
PREF22).

Subject Specify the ID of participants in the data file.

Factors Specify the factors (exact factor names in the data file.

Discrete Assumed if a factor is not labeled with one of the four alternatives
(DISCRETE, LINEAR, IDEAL, ANTIIDEAL).
In which:

Linear Used for the factors indicating that the data are expected to be
linearly related to the preference.

An expected direction for the relationship between the factor and

Less
preference: higher preference for lower factor. For instance:
higher preference for lower price.

An expected direction for the relationship between the factor and

More
preference: higher preference for higher factor. For instance:
Higher preference for higher quality.
Help the team to identify the product
profiles that can potentially meet
consumers’ demands.
Select ProductDesign_prefs
Click Run selection
RESULTS

This table shows the utility scores and their standard errors for
each factor level.
Higher utility values indicate greater preference.
⇨ There is an inverse relationship between price and utility,
with higher prices corresponding to lower utility (larger negative
values mean lower utility).
Utilities

The presence of a delivery or money-back guarantee correspond to a higher utility

Utilities

Since the utilities are all expressed in a common unit, they

can be added together to give the total utility of any
combination.

E.g: The total utility of a combination of product package B,

brand LUX, price $500, and no delivery or money-guarantee is:
1.867 + 0.367 + (-2.158) + 2.000 + 1.250 + 7.181 = 10.507
Importance Values

The results show that the product package has the most
influence on overall preference.
In contrast, a money-back guarantee plays the least important
role in determining overall preference.
Price plays a significant role but not as significant as package
design.
Coefficients

This table shows the linear regression coefficients for those factors specifies as LINEAR
(for IDEAL and ANTIIDEAL models, there would also be a quadratic term). The utility for a
particular factor level by the coefficient.
Coefficients
E.g. The predicted utility for a price of $200 was listed as -0.083 in the utilities table

This is simply the value of the price level (200) multiplied by the
price coefficient (-0.004).
200 x (-0.004) = -0.8 (approximately -0.863).
Correlations

This table displays 2 statistics, Pearson’s R and Kendall’s tau, which provide measures
of the correlation between the observed and estimated preferences.
 Preference probabilities of simulations

This table gives the predicted probabilities of choosing each of the simulation cases as the
most preferred one, under three different probability-of-choice models.

⇒ Across the 10 subjects in this study, all three models indicated that simulation profile 2
would be preferred.
Number of reversals

A count of how many times these inconsistent preferences occur

across all the choice sets or rating tasks in the conjoint analysis.
⇨ A higher number of reversals may indicate that the respondent's
preferences are less stable or more difficult to interpret.
⇨ This table displays the number of reversals for each factor and
for each subject.

E.g. Three subjects showed a reversal for price. That is, they
preferred product profiles with higher prices
⇨ Different customers have given different kinds of priorities
preferences
Thank you for listening!