Data Analysis

and
Decision Making
Session 4/14 of DADM
Decision-making in Business!
VP (Sales), Pantaloon: I noticed, users search our products very much on
websites like Flipkart and Amazon.in, but I am unable to understand why
buyers in West India not buying them.

Director (Marketing): So, did you find out about buyer behaviour?

VP (Sales), Pantaloon : I need to figure out!

Director (Marketing): How are sales and profits figures in online channel
in West India?

VP (Sales), Pantaloon: I need to figure out even that too!!

Director (Marketing): I think, you should decide if you need to change

something for my online buyers.
1. Problem definition
• Management decision problem:
• Do I need to do something different for online buyers (e.g. different
marketing mix)?

• Research problem:
• Do online buyers behave differently?

• What factors will affect purchase of online buyers for our products?
Decision-making in Business!
Variable A

Variable B Variable D

Variable C
Research Process
1. Problem definition

2. Developing the approach to the problem

3. Research design formulation

4. Data Collection

5. Data Preparation and Analysis

6. Report preparation and presentation

2. Developing the Approach
Variable A

Variable B Variable D

Variable C

• Approach: (1) QUALitative OR QUANTitative OR Mixed

(2) Data: Secondary OR Primary OR Both

2. Developing the Approach
• Management decision problem:
• Do I need to do something different for online buyers (e.g. different
marketing mix)?

• Research problem:
• What factors will affect purchase of online buyers for our products?

• Approach: (1) QUALitative OR QUANTitative OR Mixed

(2) Data: Secondary OR Primary OR Both

2. Developing the Approach
• Sources of Qualitative–Primary data
• Focus group discussion / interview (exhaustive list of variables)
• In-depth interview (relationship among variables)

• Approach: (1) QUALitative OR QUANTitative OR Mixed

(2) Data: Secondary OR Primary OR Both

IN-DEPTH INTERVIEW

• When responses of respondents may be influenced by

fellow respondents

• Discussion of sensitive topics

• When to uncover respondents’ hidden thoughts OR

motives

• When it is not possible to gather respondents for FGD

IN-DEPTH INTERVIEW
• Also known as Personal Interview

• Verbal conversation between a researcher and a

respondent

• An open-ended, discovery-oriented approach

• Probing-based (technique of encouraging respondents

to answer completely, freely and relevantly)
IN-DEPTH INTERVIEW

• Used when you have some idea of factors

(parameters)

• Conducted after substantial rapport building

• Recording
When NOT to do in-depth
interview

• When you can do FGD . . . . .

• When you have less amount of time

Laddering method: The “Why”
method
• Respondent:  I like A. Product attributes
Interviewer:  Why do you like A?
Respondent:  Because of B.
Consequences of benefits
Interviewer.  And why B?
Respondent:  Because of C.
Interviewer:  And why C? Values
Laddering example (1)
• Respondent:  I like A. Product attributes
Interviewer:  Why do you like A?
Respondent:  Because of B.
Consequences of benefits
Interviewer.  And why B?
Respondent:  Because of C.
Interviewer:  And why C? Values

• Luxury car study

I like Audi Car  Sleek look 

Quality  Comfort  Esteem
Laddering method (Ladder
questioning)
• Product attributes  Consequences of benefits  Values

• Order of questions:
1. “What is it?”
2. “What does it do?”
3. “What benefit do I get from it?”
4. “Why is this important to me?”
5. “What value(s) does this tap into?”
Laddering example (2)
• Product ingredient  Consequences of benefits  Values

• Herbal drink study

1. “What is it?”
2. “What does it do?”
3. “What benefit do I get from it?”
4. “Why is this important to me?”
5. “What value(s) does this tap into?”

Herbal ingredients in a drink  Consume less calories 

Maintain my body shape  Look good in clothes  Self-
confidence
Laddering example (3)

• I dislike . . . . .
In-depth interview
• Why is healthy ingredient important to you?
In-depth interview
• Why is good staff behaviour important to you?
Hidden issue questioning technique
• Starts with an ideal of something
• Continue the line of questioning to determine what traits
would be excluded from their perfect life.
• Focus is on personal values, not on social values (e.g.
safety, self-discipline, care, responsibility)
• Focus is on personal lifestyle, not on general lifestyle
(productive, green)
• Choose 10 existing people you can not imagine your life
• Now remove 5 of the people from your life
• Now remove 2, then another 2, people from your life
• Why did you retain the last one?
Symbolic analysis
• For example, an interviewer might ask the participant if they
regularly use e-mail.
• If so, what would the participant do if e-mail didn’t exist?
• The participant might respond that they would have to write
letters to keep in touch with their family, or take more long
trips to visit them.
• From this, the interviewer can infer that, to the participant, e-
mail represents a way to keep in touch with family.
• This kind of information can help a company determine how to
advertise for emails.
Symbolic analysis
• For example, advertising that emphasizes how easy it is to
exchange photos and home videos with family members
online, is likely to get this participants attention more than
advertising that emphasises the convenience of online sharing.
Symbolic analysis
• What if there is no pure water

• What if there is no debit / credit card

• What if there is no advertisements in the world

• What if there is no internet / electricity

• What if you don’t avail life insurance (for insurance products)

• Which product do you relate this with?

Is this an example of Laddering OR Hidden
issue questioning OR Symbolic analysis?
Is this an example of Laddering OR Hidden
issue questioning OR Symbolic analysis?
Is this an example of Laddering OR Hidden
issue questioning OR Symbolic analysis?
• Researcher: You said that a shoe’s lacing pattern is important to you in deciding
what brand to buy. Why is that?
• Consumer: A staggered lacing pattern makes the shoe fit more snuggly on my
foot.
• Researcher: Why is it important that the shoe fit more snugly on your foot?
• Consumer: Because it gives me better support.
• Researcher: Why is better support important to you?
• Consumer: So I can run without worrying about injuring my feet.
• Researcher: Why is it important to you to not worry while running?
• Consumer: So I can relax and enjoy the run.
• Researcher: Why is it important to you to relax and enjoy the run?
• Consumer: Because it gets rid of the tension I have built up from work.
• Researcher: Why is it important for you to get rid of tension from work?
• Consumer: So when I go back in the afternoon, I can perform better.
• Researcher: Why is it important for you to perform better?
• Consumer: I feel better about myself.
Advantages of Interview Method
1. Opportunity for feedback and clarifications
2. Probing Complex Answers
3. Ensures, respondent answers all questions asked,
unlike questionnaire where some questions may go
unanswered
4. Props & Visual Aids – Interviewers have the
opportunity of showing respondents items such as
sample products, graphs ands sketches, which can
aid in their answers
Disadvantages of Interview Method
Cost
– Respondents’ geographic proximity
– the number of non-respondents
Identity of respondent is revealed
Takes longer time to complete a study
Variance Effects – It has been shown that the
demographic and cultural characteristics of the
interviewer can influence the answers of the
respondents.
Interviewer’s questioning style and approach may
influence the respondents’ answers
Research framework

Product reviews

Product authenticity Purchase

satisfaction

Economic benefits
Research Process
1. Problem definition

2. Developing the approach to the problem

3. Research design formulation

4. Data Collection

5. Data Preparation and Analysis

6. Report preparation and presentation

3. Research Design
Formulation
• What is research design?
o A PLAN

• Components of research design:

1. THE Factors

2. Research questions (detailed/specific version)

1. What factors will determine purchase of online buyers for our products?

2. Which factor is most impactful? OR, What are the prominent factors?

3. Are factors different for online buyers in one city to another?

3. Research Design
Formulation
• What is research design?
o A PLAN

• Components of research design:

3. Measurement and scaling procedures

4. Pre-testing of questionnaire
5. Sampling process and size

6. Plan for data analysis

• e.g. chi-square test, t-test, ANOVA
Research questions – How to write
them?
Demographics
Age - <18, 19-30, 31-40, 41-50, >50
Gender: Male, Female, Transgender
Marital status: Single, married, Widow(er), Divorcee, Separated
Education: Uneducated, Below, Graduate, Post-graduate, Above
Income: Low, Lower medium, Medium, Upper medium, Rich, Super rich
Household expenditure:
Physical ability:
Employment: Unemployed, Employed (govt./private), Self-employed
Research questions – How to write
them?
Geographic
Research questions – How to write
them?
Behavioural
Research framework

Product reviews

Product authenticity Purchase

satisfaction

Economic benefits
Research questions – How to write
them?
1. Are consumer reviews more important for consumers in age group 21-30
than the consumers in age group 31-40?

2. Are consumer reviews more important for those who access online e-
commerce websites less frequently than those who access more
frequently?

3. Do consumers who buy products online less frequently consider economic

benefits more seriously than those who buy more frequently?

4. Which, among the three, is most important factor for buyers?

Measurement
Scales
Primary Scales
Comparative Scales
Non-Comparative Scales
Applications of Scales
Primary Scales of
Measurement
1. Nominal
2. Ordinal
3. Interval
4. Ratio
1. Nominal Scale
• What is your age? – An example of Nominal scale
• What is your gender?
• Male
• Female
• What is your education? – An example of Nominal scale
• Numbers serve only as labels or tags for identifying and classifying
objects
• Characteristics – description
• Each objects will have only one number assigned to it
• What City do you live in? • What is your blood
• Trichy
group?
• Sangli • A
• Durgapur • B+
• Churu • O
• Which section do you • What is your nationality:
belong to? • Indian
• 1 • American
• 2 • Canadian
• 3 • British
• 4 • French
• 5 • German
• 6 • Australian
• 7 • Chinese
2. Ordinal Scale

• E.g. ICC test team rankings, Forbes list

• What grade did you score in MKRH course?
• A+, A, A-, B+, B, B-, C+, C, C-, D+, D, D-, F
2. Ordinal Scale
• Ranking scale
• Numbers are assigned to the objects to indicate the
relative extent to which they possess certain
characteristics
• Therefore, you can compare – more or less
• Greater than or less than judgement from respondents
3. Interval Scale
• Ordinal + Compare difference
• Which income group do you belong to?
• >10 Lacs, 5-10 Lacs, <5 Lacs
• Which age group do you belong to?
• 1-15, 16-30, 31-45, 46-60, 60-75, >75
Difference between 1 and 2 is equal to distance between 2 and 3
or 3 and 4
Ratio of scale values does not matter
• Celsius scale
• Freezing point and boiling point – divided into 100
partitions
• Is 20˚C twice as hot as 10˚C?
• Therefore, ratios do not make sense
• How much is 30˚C multiplied by 10˚C?
• No multiplication or division can be done directly
4. Ratio Scale
• Identification and classification of objects
• Rank-order the objects
• Compare intervals or differences
• Meaningful to compute ratios
• E.g. height, weight, age, money
• How much money do you have?
• 0
• 100
• 200
• 300
• What does it mean if you select 0?
• No money?
• What does 0 temperature mean?
• No temperature?
Examples of Ratio Scale

• Does height to weight ratio determine food

preference?

• Do those who use fuel efficient cars (km per hour)

have higher tendency to travel?
Comparative Non-Comparative
Scaling Techniques Scaling Techniques
1. Paired Comparison 1. Continuous Rating
Scaling Scale
2. Rank Order Scaling 2. Itemized Rating Scale
3. Constant Sum Scaling • Likert Scale
• Semantic Differential
4. Q-sort Scaling
• Stapel Scale
Comparative Scaling
Techniques
1. Rank Order Scaling
2. Constant Sum Scaling
3. Q-sort Scaling
Rank Order Scaling
• Rank the following mobiles for their battery life:
• Apple __________
• Nokia __________
• Samsung __________
• Asus __________
• Respondents presented with objects/options and asked
to rank them according to some criterion
• Used to measure preference of brands/attributes
Rank Order Scaling
• Preference of different attributes of a brand – ordinal
and rank order scale, constant sum
Constant Sum Scaling

What is Oreo biscuit BEST known for?

Constant Sum Scaling
• Divide 100 points among the following attributes of PC:
• Screen size ________
• RAM ________
• Price ________
• Warranty ________
• Allocate a constant sum of units among a set of attributes
• If an attribute is not important, it will have zero or extremely
less points
• The sum of all points is 100
Q-sort Scaling
• Hang out, Group space, Jobs, Media, Resources,
Privacy, Study, Quiet
• Divide between “Most influential” and “Least
influential”
• And, ask the respondents to explain their
classification.
Q-sort Scaling
• Objects are sorted into piles based on similarity
• 60-140 objects (Not less than 60 and not more than
140)
• 100 statements – place them into 5 piles ranging from
strongly agree to strongly disagree
• Extremes are lesser
Non-comparative Scaling
1. Continuous Rating Scale
2. Itemized Rating Scale
• Likert Scale
• Semantic Differential
• Stapel Scale
1. Continuous Rating Scale
1. Continuous Rating Scale
• Also known as graphic rating scale
• Rating by place a mark at appropriate position
• Not restricted to researcher’s marking
2. Itemized Rating Scale
• Scale with numbers or brief descriptions of each category
• Categories are ordered
• Respondents are supposed to select specified categories
• Types
• Likert
• Semantic differential
• Stapel
2.1. Likert Scale
1 2 3 4 5

Don’t know Never Less often Often Always

Never Rare Less often Often Always

2.1. Likert Scale
• To indicate the degree of agreement or disagreement
with the statements
• Values assigned from +2 to -2 or 5 to 1
• Analysis – individual or summation
• Easy to construct and administer
• Easy to understand
• e.g. for Functional or behavioural elements
2.2. Semantic Differential
Scale
• E.g. How would you describe More on the following scale?
• Clean ___ ___ ___ ___ ____ ___ ____ Dirty
• Bright ___ ___ ___ ___ ____ ___ ____ Dark
• Inconvenient ___ ___ ___ ___ ____ ___ ____ Convenient
• 7-point rating scale with bipolar labels at the end
• Respondent is supposed to mark the blank indicating his/her
choice
• For emotional elements
2.3. Stapel Scale
• Presented
vertically
• Uni-polar,
10 categories
• For emotional
elements
• How much
better or worse
than an average
level (0 mark)
Non-comparative Itemized Rating
Scale Decisions
• Number of scale categories
• Balanced v/s unbalanced scales
• Forced v/s unforced scales
• Nature and degree of verbal description
APPLICATIONS
What is the image of my brand
amongst other competing brands?
Strong Weak
Enthusiastic Dull
Masculine Feminine
Leader Follower
Luxury Regular
Modern Traditional

Continuous Rating OR Staples

What scale should be chosen to
get this?
What scale should be chosen to
get this?
• Fast growth in height
• Short height Tall
height
• Physically weak Good physical
strength
• Dumb, Slow Quick in
grasping
Rank order scale:
Brand Awareness Score (BAS)
Brands Rank order 1. Recall a TV ad
Colgate _________ 2. A friend mentioned
Sensodyne _________
3. Have seen a poster online
Close Up _________
4. Have seen a hoarding
Pepsodent _________
Dabur Red _________
5. Sponsored a public event
Meswak _________ 6. Got a promo email
Dant Kanti _________
Does age have an effect on
environmentally-friendly behaviour?
• Age 1. I focus on reusing products
20 25 31 27 2. I dispose waste
3. I buy products that consume
less energy
22 26 27 35 4. I participate in cleanliness
drives
5. I appreciate companies
32 31 42 24 involved in waste recycling
Q-Sort scaling:
Happiness OR Excitement
• I am glad. • I am delighted.
• I am jolly. • I feel enjoyment.
• I am aroused. • I have a zeal.
• I am lively. • I am fortunate.
• I am blissful. • I am in paradise.
• I am contended. • I feel the impulse.
Effect of age on message appeal
of an advertisement
• Does age have an effect on message appeal in an
advertisement?
• Does age have an effect on environmentally-friendly
behaviour?
Application of ratio, interval and
Likert scale
• In my hotel business, does # of trips per customer have
an effect on my monthly revenue?
• Here, first we calculate # of trips per customer in ratio
scale.
• Then, we divide them in intervals: # of trips per
customer AND monthly revenue.
• Then, we code the intervals in Likert scale.
• And, then we conduct the analysis.
Which scale will you use?
• Which company is unique for buying its share?
• What is the image of my company amongst investors?
• How does Croma perceive Dell Laptops? Paired
Comparison for two attributes, Semantic Differential
Scale for more than two attributes
• What is image of my company / products amongst
distributors and retailers?
Which scale will you use?
• Director (Marketing) of HUL wants to know, how do
its sales staff perceive different brands? Why does
(s)he want to know?
• Director (Marketing) of HUL wants to know, how do
its brand mangers perceive different segments?
• Rank HUL, P&G, Marico
• Rank Fashion at Big Bazaar (FBB), Pantaloon,
Lifestyle, Zara for product quality
Which scale will you use?
• Joy Store, Big Bazaar, More, D-Mart, Reliance Fresh,
Easy Day
• Rank petrol pumps: HP, BP, Reliance, Adani, Indian
Oil
• Do both to know Zara is way ahead to others.
• Rank order scaling or Constant sum scaling?
• Constant sum scaling or Continuous rating scaling?
Which scale will you use?
• Rank ingredients in Pizza – Q-sort or Constant Sum or
Rank Ordering or Continuous Rating?
• Does Amazon has unique image in minds of
consumers against Flipkart, Snapdeal and other online
shopping websites?
Selecting scales
1. Understand your problem statement
2. Identify if you need data in form of primary scales
3. Identify if you need to compare scales
4. Select a scale OR a set of scales
5. Brainstorm with your project partners
6. Take expert opinions
7. Choose a scale
The objective is: You should be able to solve the problem
Measurement Scales
• Components:
1. Questions / Sentences / Items (Subjective)
2. Form of expression / Value (Objective)

• I feel good
• 0 20 40 60 80 100