You are on page 1of 39

Research Process & Scientific Method

Dr. A. K. Dey
Research Process: Dr. Dey 1

Learning Outcome

Classification of research based on objectives Characteristics (Principles) of good research methodology Scientific method: Validity & Reliability Research Designs Different types of Techniques of data collection Characteristics (Principles) of good research design Difficulty in employing scientific method in marketing research Steps of a marketing research process Problems in employing scientific method at each research process step and how to minimize errors
Research Process: Dr. Dey 2

Objectives of Research

To gain further insight into the problem Exploratory or Formulative Research To describe the characteristics of a situation or a group Descriptive Research To determine the frequency with which some variable occurs Diagnostic Research To test a hypothesis of causal relationship between variables Hypothesis testing research

Research Process: Dr. Dey

Characteristics of a Good Research Methodology

Systematic: Structured with specified steps to be taken in predetermined sequence according to a specific set of rules Logical: Should be guided by rules of logical reasoning Empirical: Related to one or more aspects of real situation Replicable: Results can be verified by replicating the study
Research Process: Dr. Dey 4

Scientific Method

Two main characteristics of scientific method


The claim of measuring the identified

variable Reliability or Repeatability: A repeat of the study should lead to the same outcome; like experiments of Physics or Chemistry

Very few Marketing Research projects could qualify as experiments; much less reproducible experiments
Research Process: Dr. Dey 5

What is Research Design?

A plan for selecting the sources and types of information used to answer research questions A framework for specifying the relationships among the study variables A blueprint that outlines each procedure from the hypothesis to the analysis

Research Process: Dr. Dey

Descriptors of Research Design


3. 4. 5. 6. 7. 8.

Degree of crystallization of research question Method of Data Collection Power to produce effects Purpose of the study Time dimension Topical scope breadth & depth Research environment Participants perceptions
Research Process: Dr. Dey 7

1. Degree of crystallization of Research question

Exploratory study is usually to develop hypotheses or questions for further research Formal study is to test the hypotheses or answer the research questions posed

Research Process: Dr. Dey

2. Methods of Data Collection

Monitoring, which includes observational studies Interrogation/communication studies

Research Process: Dr. Dey

3. Power to Produce Effects

In an experiment, the researcher attempts to control and/or manipulate the variables in the study In an ex post facto design, the researcher has no control over the variables; they can only report what has happened

Research Process: Dr. Dey


4. Purpose of the Study

Exploratory Studies Tries to define the problem more objectively Descriptive study tries to explain relationships among variables

Causal study is how one variable produces changes in another

Research Process: Dr. Dey 11

5. The Time Dimension

Cross-sectional studies are carried out once and represent a snapshot of one point in time Longitudinal studies are repeated over an extended period

Research Process: Dr. Dey


6. The Topical Scope

Statistical studies attempt to capture a populations characteristics by making inferences from a samples characteristics Case studies place more emphasis on a full contextual analysis of fewer events or conditions and their interrelations

Research Process: Dr. Dey


7. The Research Environment

Field conditions: Actual environmental conditions Laboratory conditions: Manipulated conditions Simulations: Major characteristics of various conditions and relationships in actual situations are often represented in mathematical models: Role plays, Mystery shoppers
Research Process: Dr. Dey 14

8. A Participants Perceptions

Usefulness of a design may be reduced when people in a disguised study perceive that research is being conducted

Participants perceptions influence the outcomes of the research: Example if the sales girl comes to know that her performance is being watched by mystery shoppers, the performance itself may change
Research Process: Dr. Dey 15

Why do Exploratory Studies?

Exploration is particularly useful when researchers lack a clear idea of the problems Outcome of an Exploratory study is a set of Hypotheses

Research Process: Dr. Dey


Data Collection Techniques

Two types: Qualitative & Quantitative Qualitative techniques: Suitable for exploratory studies

Secondary data Experience surveys In-depth interviews Focus groups Two-stage design

Research Process: Dr. Dey 17

Data Collection Techniques

Quantitative techniques: Suitable for descriptive and causal studies Only two types: Ask or Observe Observation Survey Experimentation

Research Process: Dr. Dey



The essential element of causation is A produces B or A forces B to occur

Research Process: Dr. Dey


Causal Study Relationships

Possible relationship between two variables


Variables fluctuate together but the changes in neither variable are due to the changes in the other


Variables mutually influence or reinforce each other


Changes in one variable (independent) are responsible for the changes in the other (Dependent)
Research Process: Dr. Dey 20

Asymmetrical Relationships

Four types of Asymmetrical Causal Relationship


A change in work rules leads to higher worker output


Age and attitudes about saving, Social class and opinion about taxation


Job satisfaction & work output


Age and participation in sports

Research Process: Dr. Dey 21

Principles of Experimental Design

Main objective is to achieve

Validity Reliability

Principles are


When ever the researcher faces a situation to exercise choice, random method must be used


Design should have built-in replications so as to compare outcomes

Local Control

Using best available method instruments

Research Process: Dr. Dey 22

MR & Scientific Method

It is very difficult to measure identified variable: Example TV viewer ship Controlling variables

Highly interactive; almost impossible to control in a market place Projects mostly carried out one time hence no way to test reliability MR projects are proprietary hence no way to check reproducibility

Research Process: Dr. Dey 23

Scientific Vs Non Scientific Methods

Major differences between Scientific & Non Scientific Method that affect Validity & Reliability of the results are
Objectivity of the investigator Accuracy of measurements Degree to which the investigation is continuing & exhaustive

Research Process: Dr. Dey


Scientific Vs Non Scientific Methods

Objectivity of the investigator

A researcher must base his judgment on facts & not on preconceived notion or intuition Two major factors that reduce the objectivity on the part of the investigator


Researchers report to a strong willed executive who thinks that he knows the market well Marketers often tend to exploit comfortable segments; They ignore segments that are different, unattractive or threatening
Research Process: Dr. Dey 25

Scientific Vs Non Scientific Methods

Accuracy of measurements

In experiments of Physics or Chemistry measuring devices of great accuracy are used In MR attitudes, intensions, behaviours etc are measured so far no instruments can measure these accurately Such variables are dynamic checking reproducibility is not possible Even crude devices like Attitude Rating Scales are administered by different interviewers & interviews are conducted in different settings results not reproducible
Research Process: Dr. Dey 26

Scientific Vs Non Scientific Methods

Continuing & Exhaustive nature of investigation

Scientists are continuously & aggressively searching for additional evidence, they are not sure that ultimate truth has been found Marketing Research tends to be less continuous than science Results of MR projects not published or shared no refinement done

Research Process: Dr. Dey 27

Difficulties in Applying Scientific Method to Marketing Research

Investigator involved in use of results Imprecise measuring devices Influence of measurement process on the results Time pressure for results Difficulty in using experiments to Test Hypothesis Great complexity of the subject : Complexity of human behaviourial response makes it difficult to achieve high level of Validity & Reliability
Research Process: Dr. Dey 28

Research Process

Seven inter-related steps

2. 3.

5. 6.


Specifying research objectives Preparing a list of needed information Designing the data collection project Selecting a sample type Determining sample size Organizing & carrying out the field work Analyzing the collected data & report the findings
Research Process: Dr. Dey 29

Research Process & Problems in Achieving Scientific Method

To achieve Validity & Reliability, Marketing Research should be conducted as a proper Scientific Method At each of the seven steps let us analyze
A. B.

Problems in achieving Scientific Method Steps to minimize the potential sources of errors

Research Process: Dr. Dey


1. Specifying Research Objectives


Problems in achieving Scientific Method


Managers Expectations of research results

Write research objectives Manager & researcher must discuss the objective statements & if necessary modify


Minimizing potential sources of errors

a. b.

Research Process: Dr. Dey


2. Preparing a list of needed information

Problems in achieving Scientific Method



Due to busy schedule manager may not get adequately involved May think researcher knows what to do

Minimizing potential sources of errors

Manager & researcher should develop List of needed information together & evaluate usefulness Research is not needed if manager is forced to select a particular course of action irrespective of research findings
Research Process: Dr. Dey 32

3. Designing data collection project

Problems in achieving Scientific Method

b. c. d.


Using inappropriate research design Wrong selection of respondents Asking unclear or ambiguous questions Using large scale study instead of small scale & vice versa Using poor experimental design

Research Process: Dr. Dey


3. Designing data collection project

Minimizing potential sources of errors

Five important issues that must be addressed: Should the research be exploratory or conclusive? Who should be interviewed & how? Should only few cases be studied or large samples? How well experiments be incorporated? How should data collection form be designed?
Research Process: Dr. Dey 34

4. Selecting a sample type

Why samples? Probability Vs. Non Probability Problems in achieving Scientific Method

Sample not representative of the population Define sampling frame carefully Select proper sampling method Simple Random
Research Process: Dr. Dey 35

Minimizing potential sources of errors

a. b.

5. Determining Sample size

i. ii. iii.

Problems in achieving Scientific Method

Sample size depends upon
Nature of the problem Budget Accuracy needed


Small sample Lower reliability Large sample Likely to give higher reliability

a. b.

Minimizing potential sources of errors

Use Sampling Statistics to calculate sample size for a given accuracy (Confidence Interval) Care exercised in determining sample size & sample type will minimize errors
Research Process: Dr. Dey 36

6. Organizing & Carrying out field work

Field work: Selecting, Training, Controlling & Evaluating field force Involves substantial portion of budget Potential source of errors through lack of Validity & Reliability

Problems in achieving Scientific Method

a. Varying

skills of field workers b. Forms filled without interview c. May not follow instruction d. Investigators bias e. Respondents bias
Research Process: Dr. Dey 37

6. Organizing & Carrying out field work

Minimizing potential sources of errors

a. Follow

good practices in selection, training, controlling & evaluating field workers b. Incorporate Back Checks & Spot Checks c. Motivate supervisors d. Deploy adequate field force release time pressure

Research Process: Dr. Dey


7. Analyzing Data & Report Preparation

Problems in achieving Scientific Method

Care & precaution not taken during editing, coding & data entry b. List of needed information not prepared properly c. Research objectives not established correctly

Minimizing potential sources of errors

Editing & Coding done carefully b. Incorporate extensive validity checks c. Inferences to be drawn based on factual data & not based upon researchers personal understanding
Research Process: Dr. Dey 39