Business Analytics – II

Semester – II

Dr. Rajani, IBS Bangalore

Course Outline
•Introduction to Business Research
• Questionnaire Design & Measurement Scales
•Introduction to R
•Data preparation
•Regression Analysis: Scatter plots & Correlations, Simple Linear Regression

•Multiple Regression: Multiple regression as an extension of simple regression, assumptions of multiple

regression, Inference and Regression, Categorical Independent Variables, Dummy variables, Modeling
Non-linear Regression, Model Fitting

•Multiple Regression-Model Building: Regression based Trend Models, Residual analysis, Dealing with
multi-collinearity, Transformation of Variables, Addition / deletion of variables, idea of Autocorrelation

•Logistic Regression

•Time series Forecasting: Regression based trend models, Moving Averages, Exponential Smoothing
Course Outline Conti…

• Factor Analysis: Data reduction and Combining variables

• Cluster Analysis: Segmentation tool, Hierarchical Clustering

• Optimization Modeling: Linear Programming, Two-variable Model (Graphical Method), Sensitivity

Analysis, Properties of Linear Models, Infeasibility & Unboundedness

• Transportation & Assignment Problems: Transportation and Assignment problems as special cases of
LPP and solving using Solver

• Decision Analysis: Decision making under, uncertainty and Risk, Bayesian Approach
 Analytics
What is Business Analytics is the data-driven decision-making
approach for a business problem.

Analytics (BA) Data analysis—includes data description, data

inference, and the search for relationships in data.

• Business analytics (BA) is a set of disciplines and technologies for solving business
problems using data analysis and statistical models.

• Taking in and processing historical business data. Analysing that data to identify trends,
patterns, and root causes. Making data-driven business decisions based on those insights.

• It is the science of analysing data to find out patterns that will be helpful in developing
strategies.
Types of
business
analytics
Download and Install R
https://cran.r-project.org/

Download and Install R Studio

https://www.rstudio.com/products/rstudio/download/

R Certification Course
https://www.mygreatlearning.com/academy/learn-for-free/courses/introduction-to-r
 Business Research

??????

Research - A structured investigation is known as research,

and it is done to gather data for problem-solving.

Specific Reason to do business research

• Business research is a process of acquiring detailed information on all the areas of
business and using such information to maximize the sales and profit of the business.

• Such a study helps companies determine which product/service is most profitable or in

demand.

• Business research is a part of the business intelligence process.

For example: A mobile company wants to launch a new model in the market. But they are
not aware of what are the dimensions of a mobile that are in most demand. Hence, the
company conducts business research using various methods to gather information, and the
same is then evaluated, and conclusions are drawn as to what dimensions are most in
demand.
• This process includes idea and theory development, problem definition, searching for and collecting
information, analysing data, and communicating the findings and their implications.
• Business research is a planned and organized activity generating accurate and objective information
 When Is Business Research Needed?

Not every decision requires research.

The determination of the need for research on

(1) time constraints,
(2) the availability of data,
(3) the nature of the decision to be made, and
(4) the value of the research information in relation to costs.

We have to decide whether the Research is needed or not ???

Time Constraints
Systematic research takes time. In many instances, management believes that a decision must be made immediately,
allowing no time for research.
Decisions sometimes are made without adequate information or thorough understanding of the business situation.
Although making decisions without researching a situation is not ideal, sometimes the urgency of a situation
precludes the use of research. The urgency with which managers usually want to make decisions conflicts with
researchers’ desire for rigor in following the scientific method.

Availability of data - Managers typically have access to sufficient knowledge to make wise judgments without
conducting business research. When there is a lack of data, an investigation must be done. Research can only be
done if data is made available.

Nature of the Decision: The nature of the decision to be made. Simple decisions may not need extensive research.

Benefits versus Costs: The value of the research information in relation to costs. If the cost of research is too high
compared to the benefits, it might not be worthwhile.

Business research should be conducted when these factors support the potential for informed, data-driven decision-
making that improves outcomes.
Types of Business Research

1. Exploratory

2. Descriptive

3. Causal
Exploratory Research

Purpose: To investigate an unknown or new problem.

Exploratory research investigates research questions that have not been studied in depth.

The objective of exploratory research is to gather preliminary information that will help define problems and
suggest hypotheses.

Exploratory research typically seeks to create hypotheses rather than test them. Data from exploratory
studies tends to be qualitative.

Data Collection Methods include brainstorming sessions, interviews with experts, and posting a short survey
to a social networking website.

•Example: A company wants to introduce a new product but doesn’t know what features customers want. They
conduct focus groups and open-ended interviews to explore customer preferences.
Descriptive Research

Purpose: To describe characteristics or functions of a population or phenomenon.

Descriptive research aims to accurately and systematically describe a population, situation or phenomenon. It
can answer what, where, when and how questions. A descriptive research design can use a wide variety of
research methods to investigate one or more variables.

Exploratory research is usually conducted when a researcher has just begun an investigation and wishes to
understand the topic generally. Descriptive research aims to describe or define the topic at hand.
Example: A market research firm surveys 1,000 people to determine the average age and gender distribution of
customers who buy a specific type of product.
Causal Research
Purpose: To determine cause-and-effect relationships.

Causal research is a type of market research that aims to establish cause-and-effect relationships between variables.

It can help you understand how changing one factor, such as price, packaging, or promotion, can affect another
factor, such as sales, customer satisfaction, or brand loyalty.

Causal research aims to establish whether one variable (the cause) directly affects another variable (the effect),
explanatory research goes a step further by providing a detailed understanding of how and why that effect
happens.

Rain causes grass to get wet. Rain is the cause, and wet grass is the effect.

Example: A company conducts an experiment to see whether

increasing the price of a product (independent variable) will lead
to a drop in sales (dependent variable). The research aims to
establish the cause-and-effect relationship between price changes
and sales behavior.
CAUSALITY

Causality refers to the relationship between two events where one event (the cause) directly influences another event
(the effect). In research, establishing causality is key to understanding cause-and-effect relationships.

Ideally, managers want to know how a change in one event will change another event of interest.

For example, how will implementing a new employee training program change job performance?

Example: If you eat a healthy diet (cause), you may experience better health (effect). Causality implies that there is a
direct, logical connection between the two events.

Causality is a concept that refers to the idea of cause and effect.

Causal Research is a methodological approach used to test and establish causality by examining relationships
between variables.
Key Aspects of Causality:

1.Temporal Precedence: The cause must occur before the effect.

For example, if you hypothesize that exercise leads to weight loss, the exercise (cause) must come before the observed
weight loss (effect).

2. Covariation of Cause and Effect: The cause and effect must be related. When the cause occurs, the effect must also
occur, and when the cause does not occur, the effect should be absent.

For example, an increase in advertising spending (cause) should be followed by an increase in sales (effect).

3. Elimination of Confounding Variables: To confirm that one variable is causing another, alternative explanations or
confounding factors must be ruled out. This is often done using control groups or statistical methods to isolate the cause
and its effect.

Example: A study is conducted to determine whether drinking coffee (independent variable) increases productivity
(dependent variable) at work.
Potential Confounding Variable: The amount of sleep participants get the night before. (Eliminating the confounding
variable to get more accurate result)
DEGREES OF CAUSALITY

Absolute Causality: Absolute causality means that the cause is necessary and sufficient to bring about the
effect.

One variable directly causes a change in another, with no intervening variables.

For example, When temperatures fall below 32º F, unsalted water begins to freeze.

Example: Pushing a light switch (cause) directly turns on the light (effect).

Strength: This is the strongest form of causality where the cause is known to immediately produce the effect.
Conditional Causality: Conditional causality means that a cause is necessary, but not sufficient to bring about
an effect.

A variable contributes to an effect but is not the sole cause. Multiple factors may be involved in producing the
outcome.

For example “smoking causes cancer.”

Example: A poor diet contributes to weight gain, but other factors like physical activity and genetics also play a
role.

Strength: This is a moderate level of causality, as the cause is one of several contributors.
Contributory Causality: Contributory causality is the weakest form of causality. Contributory
causality is when the cause is neither necessary nor sufficient to bring about the effect.

The cause increases the likelihood of the effect but does not guarantee it will happen.

Example: Smoking increases the probability of developing lung cancer, but not every smoker will
develop cancer.

Strength: This level suggests a correlation that is strong but not definitive. The cause increases the risk,
but other factors may also influence the outcome.
Stages in the Research Process
Market Research Simulation for a New Product Launch - Sustainable fashion (e.g., clothing made
from recycled fabrics).

Market Research Plan

The target market (demographics, location, income level, etc.)

Competitors (existing companies offering similar products)
Research methods (surveys, interviews, focus groups, etc.)
Key questions to answer (e.g., What features do customers prefer? What pricing model is best?)
A startup wants to explore potential markets for an innovative app.

For Exploratory Research: Focus on open-ended approaches, like secondary data analysis or informal
interviews, to gather initial insights.

For Descriptive Research: Focus on describing characteristics of the market, such as demographics, market
size, or product usage patterns.

For Causal Research: Focus on experiments or testing to understand cause-and-effect relationships, such
as how price changes impact sales.
Measurement and Scaling
Measurement and Scaling

Gender - male/female, hair

colour, nationalities, names
of people

education level (“high

school”,”BS”,”MS”,”PhD”),
income level (“less than 50K”,
“50K-100K”, “over 100K”)

temperature (in Celsius or

Fahrenheit), mark grading and
CGPA.

length, duration, income, weight

Chi-square test
correlation
Always numerical data
Numerical data
it is possible to perform On a ratio scale true zero means
mathematical operations - there is a total absence of the
Example - the number of
variable of interest.
children in a household or
years of work experience.
Interval scales hold no true zero and can represent values below zero.

For example, you can measure temperatures below 0 degrees Celsius,

such as -10 degrees.

Ratio variables, on the other hand, never fall below zero. Height and
weight measure from 0 and above, but never fall below it.
 Example - interval
and ration scale

Temperature in Celsius or Fahrenheit is at an interval scale because zero is not the lowest
possible temperature. In the Kelvin scale, a ratio scale, zero represents a total lack of
thermal energy.
EXHIBIT 1.8 Facts About
the Four Levels of Numerical Descriptive
Level Examples Operations Statistics
Scales
Nominal Student ID number Counting Frequencies
Yes – No Mode
Male – Female
Buy – Did Not Buy
East region
Central region
West region
Ordinal Student class rank Counting Frequencies
Please rank your three favorite movies. Ordering Mode
Choose from the following: Median
Dissatisfied Range
Satisfied
Very satisfied
Delighted
Indicate your level of education:
Some high school
High school diploma
Some college
College degree
Graduate degree
 Graduate degree
Interval Student grade point average (GPA) Common Frequencies
Temperature (Celsius and Fahrenheit) arithmetic Mode
Points given on an essay question operations Median
100-point job performance rating provided by Range
supervisor
Mean
Variance
Standard deviation
Ratio Amount spent on last purchase All Frequencies
Salesperson sales volume arithmetic Mode
Number of stores visited on a shopping trip operations Median
Annual family income Range
Time spent viewing a Web page Mean
Variance
Standard deviation
Attitude Measurement - Questionnaire
Attitude measurement is an important aspect of data collection in marketing research. It involves measuring the
attitudes and opinions of individuals towards a particular product, service, or brand. Attitude measurement can
provide valuable insights into customer behaviour and preferences.

Attitude measurement is a key concept in understanding people's preferences or biases.

There are three components of attitude:

Affective component – refers to an individual’s general feelings or emotions toward an object.

(e.g., liking or disliking a product).

Cognitive component – represents an individual’s awareness and knowledge about an object.

(e.g., thinking a service is useful or harmful).

Behavioural component – the effect of attitude on behaviour. (e.g., the intention to buy or avoid something).

Attitude measurement is highly important in a questionnaire because it helps capture people's beliefs, feelings, and
predispositions toward various objects, concepts, or behaviors.
Techniques for Measuring Attitudes – Rating scale and Ranking Scale
A ranking scale question asks the respondent to put items in order, usually order of preference.
A rating question is where the respondent assigns a score to each item.

Rating scale - the most common practice in business research

•Simple Attitude Scales

The simplest rating scale contains only two response categories: agree/disagree.

This type of self-rating scale merely classifies respondents into one of two categories; it has only the properties of a
nominal scale.
• Category Attitude Scales
Give more flexibility in the rating of a task.

A category scale is more sensitive than a simple scale categories—it provides more information.

More than two response categories: Strongly agree/ Agree / Uncertain / Disagree / Strongly disagree.
Different Method of Rating Scale

• The Likert Scale

The Likert scale is an extremely popular means for measuring attitudes.

Individuals generally choose from approximately five (although alternatives may range from three to nine)
response alternatives: “strongly agree,” “agree,” “uncertain,” “disagree,” and “strongly disagree.”
Researchers assign scores, or weights, to each possible response (e.g., 1, 2, 3, 4, and 5 assigned to each level
of agreement).

•Reverse Recoding
If a statement is framed negatively, the numerical scores would need to be reversed.
e.g.
Old Value - 1, 2, 3, 4, and 5 assigned to each level of agreement
New Value – 5, 4, 3, 2 and 1
Semantic Differential
Semantic differential scale questions - ask respondents to mark their position on a scale between two opposite adjectives

A semantic differential scale uses two bipolar adjectives, one negative adjective on the left of the scale, and one positive
adjective on the right of the scale.

For example "Good-Bad", "Hot-Cold", "Black-White.

For scoring purposes, a numerical score is assigned to each position on the rating scale (i.e., 1, 2, 3, 4, 5, 6, 7 or -3, -2, -1, 0,
+1, +2, +3).

Example: How do you feel about the user interface of the new mobile app?

Unfriendly -3 -2 -1 0 +1 +2 +3 Friendly

Scale: The scale uses a continuum between two opposite adjectives: "Unfriendly" and "Friendly." The respondent can
choose a position from -3 (most negative) to +3 (most positive), with 0 being neutral.
Numerical Scales
In numerical scales, numbers, rather than semantic space or verbal descriptions, serve as the response options to
identify categories (response positions).

If the scale items have five response positions, the scale is called a 5-point numerical scale, six options is called a
6-point scale, etc.

The numerical scale utilizes bipolar adjectives in the same manner as the semantic differential.

Example: How do you feel about the user interface of the new mobile app?

1 2 3 4 5 6 7 8 9 10
Very
Excellent
Poor

This allows the respondent to give a precise numerical rating, with 1 representing a very poor user
interface and 10 representing an excellent user interface.
Stapel Scale

The Stapel scale is a unipolar rating scale with ten categories numbered from -5 to +5, without a neutral
point (zero).

Question: How would you rate the user interface of the new mobile app based on the following
characteristics?

+5 +4 +3 +2 +1 Efficient -1 -2 -3 -4 -5

Respondents choose a number between +5 (very efficient) and -5 (very inefficient). You can apply this to
various adjectives like "Efficient," "User-friendly," or "Attractive" to gauge specific aspects of the app's
interface.
Monadic or Comparative Scale

• A monadic rating scale uses no comparison; it asks a respondent to rate a single concept in isolation.

• On a seven point scale Product A may be rated as 6, Product B as 4, and Product C as 1.

• A comparative rating scale asks a respondent to rate a concept in comparison with a benchmark

• McDonald's, to compare with a different fast food restaurant like Burger King. For example: Rate Burger
King in comparison to McDonald's
A researcher wishes to compare two hotels on the following attributes:

Convenience of location
Friendly personnel
Value for the money

a. Design a Likert scale to accomplish this task.

A researcher wishes to compare two hotels on the following attributes:

Convenience of location
Friendly personnel
Value for the money
Questionnaire
A questionnaire is a research instrument that consists of a set of questions or other types of
prompts that aims to collect information from a respondent.

A research questionnaire is typically a mix of close-ended questions (Structured) and open-ended

questions (Unstructured).

Open-ended, long-form questions offer the respondent the ability to elaborate on their thoughts.

Closed-ended or fixed-alternative questions give respondents specific limited-alternative

responses and ask them to choose the one closest to their own viewpoint.
Types of Fixed-Alternative Questions

1. Simple questions – require the respondent to choose one of two alternatives. The answer can be a simple “yes” or
“no”

2. Several types of questions provide the respondent with multi-choice alternatives:

•The determinant-choice question requires the respondent to choose one and only one—response from among
several possible alternatives.
•The frequency-determination question is a determinant-choice question that asks for an answer about the general
frequency of occurrence.
•Attitude rating scales, such as the Likert scale, semantic differential, and Stapel scale are also fixed-alternative
questions
•The checklist question allows respondents to provide multiple answers to a single question. In many cases, the
choices are adjectives that describe a particular object.
GUIDELINES FOR CONSTRUCTING QUESTIONS

In developing a questionnaire, there are no hard-and-fast-rules.

Avoid Complexity: Use Simple, Conversational Language Words used in questionnaires should be readily
understandable to all respondents.

Avoid Leading and Loaded Questions – Answers are very specific

Example - "Do you agree that our company's product is the best on the market?”

Avoid Ambiguity: Be as Specific as Possible

Avoid Making Assumptions

Avoid Burdensome Questions - Questions that give no clue as to the brand of interest are referred to as
unaided recall questions.
Find the problem and define objective

Set Hypothesis

Design the questionnaire

Collect the data using questionnaire

Prepare and analysis with the help of different techniques

A STUDY ON CUSTOMER SATISFACTION OF NAGA PRODUCTS & MARKET POTENTIAL TOWARDS NAGA
FROZEN CHAPATI & PARATHA
Hypothesis
Design the questionnaire and Collect the data using questionnaire
Analysis with the help of different techniques

Analysis for Age of respondents

Analysis for Gender of respondents