Business research methods - Quantitative
1. Introduction
With knowledge of this course you will be able to perform business research.
Research Process: the typical research process consists of 6 consecutive stages. We will
shortly dive into the first 3 stages, the last 3 will be examined later.
Stage 1 – Problem definition:
In this first stage the researcher tries to narrow down the gap that might exist between the
business/decision problem (manager focused) and the research problem (research
focused). Example: when Adidas is facing a problem with the sale of flip flops in Alaska
(business/decision problem), research could investigate to what extent a polar climate
impacts the use of flip flops and whether it is dependent of the absence of infrastructure
(research problem)
,Stage 2 – Research approach development:
The second stage is concerned with what is called the theoretical framework. This consists
of 3 elements:
- A description of all relevant variables and their definitions
- The hypotheses (i.e., expected relationship between variables)
- A conceptual model (graphical representation)
This is an example of conceptual
model with several variables. H1-3
and H4-6 refer to the hypotheses.
In this example a hypothesis could
be H1:Brand cognition has a
positive influence on advertising
effects.
Stage 3 – Research design:
A research design can be defined as “a framework or plan for conducting a research project.
It details the procedures necessary for obtaining the information needed to structure or
solve research problems”. It involves the following components or tasks:
- Define the information needed
- Decide on nature of research
- Decide on techniques and measurement
- Construct and pre-test the research
- Decide on sampling process and sample size
- Develop a data analysis plan
,The research design can be classified in two distinct types: exploratory research and
conclusive research. It is important to know the difference between these two!
- Exploratory research > used to understand phenomena about which little is
known, and little theory exist. To understand phenomena that is difficult to
measure. New theory may evolve through an in-depth and flexible
approach.
- Conclusive research > aims to tests hypotheses about clearly defined
phenomena, measured by means of quantitative data.
▪ Descriptive research: testing the correlational relationship
between two or more variables (e.g., by means of a survey or
archival data)
▪ Causal research: testing the causal relationship between two or
more variables by means of an experiment
Correlation versus causality
In sum, the conditions for causality are:
- X and Y co-occur (correlate)
- A logical explanation for the effect of X on Y is needed
- X proceeds Y in time
- No other cause (Z) explains the co-occurrence (correlation) of X and Y
Hallmarks
In order to reach a sound and reliable research, each researcher should keep the 8
hallmarks in mind:
1. Purposiveness – why you are doing research
2. Rigor – a sound theoretical base and design
3. Testability – ability to test your ideas
4. Replicability – finding consistent results when the research is repeated in
similar conditions
5. Precision & confidence – accurate conclusions with a high degree of confidence
6. Objectivity – conclusions based on facts
7. Generalizability – ability to apply your findings in a variety of settings
8. Parsimony – explaining a lot with a little
Polls lecture 1:
Exercise 2: Would you use a boxplot or plot function to inspect how age relates to satisfaction?
> Plot: provides insights to the spread of data, as well as the relationship
Knowledge poll: How can you distinguish between causality and correlation on existing dataset?
> You cannot distinguish between causality and correlation based on statistics
2. Measurement & Scales
, In order to analyze variables, we assign numbers to characteristics of objects >
measurement. Consecutively, these measured objects can be placed on a generated
continuum > scaling.
There are 4 primary scales of measurement: nominal, ordinal, interval and ratio. They
differ on 4 basic characteristics: difference, order, distance and origin
Nominal scale: a scale where the numbers assigned to variables only serve as labels or tags
classifying objects. The size of the numbers has no meaning.
- Example: Male (1) and female (2), Hispanic or Black
Ordinal scale: a scale where the numbers assigned to objects indicate the relative extent to
which some characteristic is possessed (i.e., more or less of a characteristic than another
object)
- Example: Large, Medium, Small
- Example: Position in a race (1st, 2nd, 3rd, 4th)
- Order has meaning
- Distance between scale points is not the same
Interval scale: a scale where numbers are used to rank objects such that numerically equal
distances on the scale represent numerical equal distances in the characteristics being
measured
- Example: “2018”, temperature
- Distances between scale points is the same
Ratio scale: a scale to identify or classify objects, rank-order the objects AND compare
intervals or differences. It is a scale with an absolute zero point > the number 0 indicates
the absence of the attribute being measured
- Example: Money, age, ratio on a map
- Distances between scale points is the same
The more information the scale provides, the more powerful are the statistical techniques
you can use. An important thing to remember:
- Nominal & ordinal > Non-metric
- Interval & Ratio > Metric
1. Introduction
With knowledge of this course you will be able to perform business research.
Research Process: the typical research process consists of 6 consecutive stages. We will
shortly dive into the first 3 stages, the last 3 will be examined later.
Stage 1 – Problem definition:
In this first stage the researcher tries to narrow down the gap that might exist between the
business/decision problem (manager focused) and the research problem (research
focused). Example: when Adidas is facing a problem with the sale of flip flops in Alaska
(business/decision problem), research could investigate to what extent a polar climate
impacts the use of flip flops and whether it is dependent of the absence of infrastructure
(research problem)
,Stage 2 – Research approach development:
The second stage is concerned with what is called the theoretical framework. This consists
of 3 elements:
- A description of all relevant variables and their definitions
- The hypotheses (i.e., expected relationship between variables)
- A conceptual model (graphical representation)
This is an example of conceptual
model with several variables. H1-3
and H4-6 refer to the hypotheses.
In this example a hypothesis could
be H1:Brand cognition has a
positive influence on advertising
effects.
Stage 3 – Research design:
A research design can be defined as “a framework or plan for conducting a research project.
It details the procedures necessary for obtaining the information needed to structure or
solve research problems”. It involves the following components or tasks:
- Define the information needed
- Decide on nature of research
- Decide on techniques and measurement
- Construct and pre-test the research
- Decide on sampling process and sample size
- Develop a data analysis plan
,The research design can be classified in two distinct types: exploratory research and
conclusive research. It is important to know the difference between these two!
- Exploratory research > used to understand phenomena about which little is
known, and little theory exist. To understand phenomena that is difficult to
measure. New theory may evolve through an in-depth and flexible
approach.
- Conclusive research > aims to tests hypotheses about clearly defined
phenomena, measured by means of quantitative data.
▪ Descriptive research: testing the correlational relationship
between two or more variables (e.g., by means of a survey or
archival data)
▪ Causal research: testing the causal relationship between two or
more variables by means of an experiment
Correlation versus causality
In sum, the conditions for causality are:
- X and Y co-occur (correlate)
- A logical explanation for the effect of X on Y is needed
- X proceeds Y in time
- No other cause (Z) explains the co-occurrence (correlation) of X and Y
Hallmarks
In order to reach a sound and reliable research, each researcher should keep the 8
hallmarks in mind:
1. Purposiveness – why you are doing research
2. Rigor – a sound theoretical base and design
3. Testability – ability to test your ideas
4. Replicability – finding consistent results when the research is repeated in
similar conditions
5. Precision & confidence – accurate conclusions with a high degree of confidence
6. Objectivity – conclusions based on facts
7. Generalizability – ability to apply your findings in a variety of settings
8. Parsimony – explaining a lot with a little
Polls lecture 1:
Exercise 2: Would you use a boxplot or plot function to inspect how age relates to satisfaction?
> Plot: provides insights to the spread of data, as well as the relationship
Knowledge poll: How can you distinguish between causality and correlation on existing dataset?
> You cannot distinguish between causality and correlation based on statistics
2. Measurement & Scales
, In order to analyze variables, we assign numbers to characteristics of objects >
measurement. Consecutively, these measured objects can be placed on a generated
continuum > scaling.
There are 4 primary scales of measurement: nominal, ordinal, interval and ratio. They
differ on 4 basic characteristics: difference, order, distance and origin
Nominal scale: a scale where the numbers assigned to variables only serve as labels or tags
classifying objects. The size of the numbers has no meaning.
- Example: Male (1) and female (2), Hispanic or Black
Ordinal scale: a scale where the numbers assigned to objects indicate the relative extent to
which some characteristic is possessed (i.e., more or less of a characteristic than another
object)
- Example: Large, Medium, Small
- Example: Position in a race (1st, 2nd, 3rd, 4th)
- Order has meaning
- Distance between scale points is not the same
Interval scale: a scale where numbers are used to rank objects such that numerically equal
distances on the scale represent numerical equal distances in the characteristics being
measured
- Example: “2018”, temperature
- Distances between scale points is the same
Ratio scale: a scale to identify or classify objects, rank-order the objects AND compare
intervals or differences. It is a scale with an absolute zero point > the number 0 indicates
the absence of the attribute being measured
- Example: Money, age, ratio on a map
- Distances between scale points is the same
The more information the scale provides, the more powerful are the statistical techniques
you can use. An important thing to remember:
- Nominal & ordinal > Non-metric
- Interval & Ratio > Metric
