Table of contents
Chapter I: Why Science?....................................................................................................................2
1. What is Science?........................................................................................................................2
2. The broader purposes of business research...............................................................................3
Chapter II: Thinking like a researcher.................................................................................................3
3. Concepts, constructs and variables............................................................................................3
4. Propositions, hypotheses, theories and models........................................................................6
5. What makes a good theory?......................................................................................................9
6. Unit of analysis.........................................................................................................................11
Chapter III: Overview of the scientific method................................................................................11
7. Model of scientific research in business...................................................................................11
8. Finding a research topic...........................................................................................................12
9. Generating good research questions.......................................................................................14
10. Developing a hypothesis........................................................................................................16
11. Designing a research study.....................................................................................................17
12. Analysing the data..................................................................................................................19
13. Drawing conclusions and reporting the results......................................................................21
Chapter IV: Measurement................................................................................................................21
14. Reliability and validity of measurement.................................................................................21
15. Measurement in practice.......................................................................................................25
16. A look ahead: research design...............................................................................................27
Chapter VIII: Results – Descriptive and inferential statistics.............................................................30
26. The purpose of null hypothesis testing..................................................................................30
27. Expressing you results............................................................................................................32
28. Describing statistical relationships.........................................................................................33
29. Additional considerations in null hypothesis testing..............................................................35
30. Conducting your analyses......................................................................................................37
Chapter IX: Presenting your research...............................................................................................39
32. Citing earlier work in APA style..............................................................................................39
1
,Chapter I: Why Science?
1. What is Science?
Every discipline has a different subject matter, but they share three fundamental features of science:
1. Systematic empiricism: Empiricism refers to learning based on observation, and scientists
learn about the world systematically. They do so by planning, making, recording and
analyzing observations of it. Scientists are unique in their insistence on checking their ideas
about the way the world works against their systematic observations. When these systematic
observations turn out to conflict with their ideas, ideas should be adapted.
2. Empirical questions: Questions about the way the world actually is. These questions can be
answered by systematically observing it.
Example: The question of whether a particular investment strategy yields higher returns than
another is empirical. Either the investment strategy produces higher returns or it does not.
This can be determined by systematically observing and analyzing financial data.
3. Public knowledge: After asking their empirical questions, making their systematic
observations and drawing their conclusions, scientists publish their work
Example: Public article in a professional journal containing: the research question in the
context of previous research, a detailed description of the methods they used to answer their
question, and the presentation of their results and conclusion.
Publication is an essential feature of science for two reasons:
- Science is a remarkably social process. It is a large-scale collaboration among many
researchers distributed across both time and space. Our current scientific knowledge of most
topics is based on many different studies conducted by many different researchers who have
shared their work publicly over many years.
- Publication allows science to be self-correcting. Individual scientists understand that their
data or methods can be flawed, they might make mistakes, and their conclusions may be
incorrect. Publication allows other scientists to detect and correct these errors so that, over
time, scientific knowledge increasingly reflects the way the world actually is.
Because individual studies can be flawed, or specific to the time and place in which they were
conducted, academics typically don’t treat a single study as the final say on a certain subject.
Sometimes there is lots of debate on whether a certain effect, theory or finding actually exists, even
after many studies by many different authors. In order to settle such debates or test how widely
important theories and results apply, many disciplines have recently seen more large-scale
replication studies that test key theories in many different labs and locations at the same time.
2
,2. The broader purposes of business research
Science/business research has three goals:
1. Describe (how does something work?) This goal is achieved by making observations.
Example: Perhaps I know different managers, and observe that some are more successful at
managing their team than others. I could perform qualitative research, follow these
managers, and describe what they do throughout the day. Similarly, I could ask them to
participate in a survey in order to gain some sense of how large the difference between
successful and less-successful managers actually is. Such approaches could help me move on
towards actually explaining what makes some managers more successful than others.
2. Explain (why does something work this way?) determining the causes of behavior.
Example: I could survey many of the aforementioned managers and ask them all kinds of
questions on their personality, management style, or other relevant characteristics.
Subsequently, I could collect data from the members of their team to assess how effective
they think their manager is, or perhaps collect some more objective measure of success. I
could then use statistical methods to see whether these things are related. If I find that they
are, I have improved my understanding of which factors specifically make managers
successful, and how important each of these factors is. If I find no connection between these
factors and manager success, I may need to revise my ideas of which factors are important
and why. Therefore, in each case, we have learned something.
3. Predict (how will something look in the future?) once we have observed with some
regularity that two behaviors or events are systematically related to one another, we can use
that information to predict whether an event or behavior will occur in a certain situation.
Example: Once I understand which factors drive managers’ success, I can predict whether a
job candidate I want to hire is likely to be (or become) a successful manager by thinking
about how they ‘score’ on the set of characteristics that would make a manager successful.
Scientific research is often classified as being either basic or applied:
Basic research Conducted primarily for the sake of achieving a more detailed and
accurate understanding of human behavior, without necessarily trying to address any
particular practical problem.
Applied research Conducted primarily to address some practical problem. Research on
the effects of cell phone use on driving, for example, was prompted by safety concerns and
has led to the enactment of laws to limit this practice.
Many papers have both a ‘basic’ and ‘applied’ element to them, and many excellent studies
contribute both to theoretical and practical understanding.
Chapter II: Thinking like a researcher
3. Concepts, constructs and variables
For many academics, understanding and explaining are the ultimate goal of doing research.
Explanations require development of concepts = generalizable properties or characteristics
associated with objects, events, or people:
3
, Examples: a person’s attitude toward immigrants, a firm’s capacity for innovation, a car’s
weight.
Some concepts have been developed over time through our shared language. Sometimes, we borrow
concepts from other disciplines or languages to explain a phenomenon of interest.
Concepts may have progressive levels of abstraction. Some concepts such as a person’s weight are
precise and objective, while other concepts such as a person’s personality may be more abstract and
difficult to visualize.
Construct = an abstract concept that is specifically chosen (or created) to explain a given
phenomenon. A construct can be unidimensional or multidimensional:
Unidimensional construct a simple concept, such as a person’s weight.
Multidimensional construct a combination of a set of related concepts, such as a person’s
communication skill, which may consist of several underlying concepts such as the person’s
vocabulary, syntax, and spelling.
Constructs used for scientific research must have precise and clear definitions that others can use to
understand exactly what it means and what it does not mean.
Example: A simple construct such as income may refer to monthly or annual income, before
or after-tax income, and personal or family income, and is therefore neither precise nor clear.
When discussing constructs, there are two types of definitions:
Conceptual definitions Involves defining a construct on an abstract and theoretical level.
Example: Perceived service quality the degree and direction of discrepancy between
consumers’ perceptions and expectations of a company’s service performance.
Operational definitions Are necessary when it comes to actually collecting and analyzing
data, and have to be very specific on how you will actually measure your construct.
Example: Perceived service quality a customer’s rating of a company on a 26-item survey
scale that consists of five underlying service quality dimensions: a company’s (service)
reliability, responsiveness, (perceived) empathy, assurance and performance on tangibles.
Example: Income monthly or annual, before- or after-tax, personal or family income.
Example: Temperature measured in Celsius, Fahrenheit or Kelvin scale.
Differences in operational definitions could lead to different findings, even if the conceptual
definition is the same.
A term that is frequently associated with a construct is a variable = a quantity that can vary (from low
to high, negative to positive etc.), in contrast to constants that do not vary (remain constant). In
science, a variable is a measurable representation of an abstract construct.
Constructs are not directly measurable. Therefore, we use variables.
Example: a person’s intelligence is often measured as his or her IQ score. In this case,
intelligence is a construct and IQ score is a variable that measures the intelligence construct.
Thinking like a researcher is translating between planes:
Theoretical plane Here, we discuss constructs, conceptual definitions and propositions in
an abstract sense.
4
, Empirical plane Here, constructs have operational definitions, are measured by variables,
about which we may have hypotheses.
Thinking like a researcher involves the ability to move back and forth between the more abstract
theoretical plane and the more specific and tangible empirical plane (see figure 1).
See an example of translating between both planes on page 11 of the RSR.
Conclusion of this example: The exact same empirical observation might lead to many
different ideas regarding how and why something occurs or works on the theoretical plane.
Figure 1: The theoretical and empirical planes of research
Depending on their intended use, variables can be classified as followed:
Independent variables Variables that explain other variables.
Dependant variables Variables that are explained by other variables.
Moderating variables (= moderator) Variables that strengthens or weakens the effect of
another variable, and can sometimes change the direction of the effect.
Example: We do research on how intelligence (independent variable) affects academic
success (dependent variable). We may expect that greater amounts of intelligence would be
related to greater amounts of academic success. However, there’s also effort to consider,
which could work as a moderator: effort could ‘strengthen’ the effect of someone’s
intelligence on academic success. Between two equally intelligent students, the student who
puts in more effort achieves greater academic success.
Note: moderators are not the same as independent variables. We’re not stating that students
who put in more effort have a higher IQ for example. Rather, the point is that the effect of
intelligence on academic success depends on a third variable (here: effort).
Mediating variables (intermediate) Variables that are explained by independent
variables, but explain dependent variables. These variables help us understand the
mechanism/process through which something works.
Example:
o Independent variable: service quality
o Hypothesis: service quality influences customer loyalty
o Idea: service quality doesn’t influence customer loyalty directly, but works ‘through’
customer satisfaction: a higher service quality makes customers more satisfied, and
more satisfied customers – in turn – become more loyal.
o Here, customer satisfaction would be the mediating variable.
Control variables Variables that have to take into account a in scientific study, but are
otherwise not of a main interest.
5
Chapter I: Why Science?....................................................................................................................2
1. What is Science?........................................................................................................................2
2. The broader purposes of business research...............................................................................3
Chapter II: Thinking like a researcher.................................................................................................3
3. Concepts, constructs and variables............................................................................................3
4. Propositions, hypotheses, theories and models........................................................................6
5. What makes a good theory?......................................................................................................9
6. Unit of analysis.........................................................................................................................11
Chapter III: Overview of the scientific method................................................................................11
7. Model of scientific research in business...................................................................................11
8. Finding a research topic...........................................................................................................12
9. Generating good research questions.......................................................................................14
10. Developing a hypothesis........................................................................................................16
11. Designing a research study.....................................................................................................17
12. Analysing the data..................................................................................................................19
13. Drawing conclusions and reporting the results......................................................................21
Chapter IV: Measurement................................................................................................................21
14. Reliability and validity of measurement.................................................................................21
15. Measurement in practice.......................................................................................................25
16. A look ahead: research design...............................................................................................27
Chapter VIII: Results – Descriptive and inferential statistics.............................................................30
26. The purpose of null hypothesis testing..................................................................................30
27. Expressing you results............................................................................................................32
28. Describing statistical relationships.........................................................................................33
29. Additional considerations in null hypothesis testing..............................................................35
30. Conducting your analyses......................................................................................................37
Chapter IX: Presenting your research...............................................................................................39
32. Citing earlier work in APA style..............................................................................................39
1
,Chapter I: Why Science?
1. What is Science?
Every discipline has a different subject matter, but they share three fundamental features of science:
1. Systematic empiricism: Empiricism refers to learning based on observation, and scientists
learn about the world systematically. They do so by planning, making, recording and
analyzing observations of it. Scientists are unique in their insistence on checking their ideas
about the way the world works against their systematic observations. When these systematic
observations turn out to conflict with their ideas, ideas should be adapted.
2. Empirical questions: Questions about the way the world actually is. These questions can be
answered by systematically observing it.
Example: The question of whether a particular investment strategy yields higher returns than
another is empirical. Either the investment strategy produces higher returns or it does not.
This can be determined by systematically observing and analyzing financial data.
3. Public knowledge: After asking their empirical questions, making their systematic
observations and drawing their conclusions, scientists publish their work
Example: Public article in a professional journal containing: the research question in the
context of previous research, a detailed description of the methods they used to answer their
question, and the presentation of their results and conclusion.
Publication is an essential feature of science for two reasons:
- Science is a remarkably social process. It is a large-scale collaboration among many
researchers distributed across both time and space. Our current scientific knowledge of most
topics is based on many different studies conducted by many different researchers who have
shared their work publicly over many years.
- Publication allows science to be self-correcting. Individual scientists understand that their
data or methods can be flawed, they might make mistakes, and their conclusions may be
incorrect. Publication allows other scientists to detect and correct these errors so that, over
time, scientific knowledge increasingly reflects the way the world actually is.
Because individual studies can be flawed, or specific to the time and place in which they were
conducted, academics typically don’t treat a single study as the final say on a certain subject.
Sometimes there is lots of debate on whether a certain effect, theory or finding actually exists, even
after many studies by many different authors. In order to settle such debates or test how widely
important theories and results apply, many disciplines have recently seen more large-scale
replication studies that test key theories in many different labs and locations at the same time.
2
,2. The broader purposes of business research
Science/business research has three goals:
1. Describe (how does something work?) This goal is achieved by making observations.
Example: Perhaps I know different managers, and observe that some are more successful at
managing their team than others. I could perform qualitative research, follow these
managers, and describe what they do throughout the day. Similarly, I could ask them to
participate in a survey in order to gain some sense of how large the difference between
successful and less-successful managers actually is. Such approaches could help me move on
towards actually explaining what makes some managers more successful than others.
2. Explain (why does something work this way?) determining the causes of behavior.
Example: I could survey many of the aforementioned managers and ask them all kinds of
questions on their personality, management style, or other relevant characteristics.
Subsequently, I could collect data from the members of their team to assess how effective
they think their manager is, or perhaps collect some more objective measure of success. I
could then use statistical methods to see whether these things are related. If I find that they
are, I have improved my understanding of which factors specifically make managers
successful, and how important each of these factors is. If I find no connection between these
factors and manager success, I may need to revise my ideas of which factors are important
and why. Therefore, in each case, we have learned something.
3. Predict (how will something look in the future?) once we have observed with some
regularity that two behaviors or events are systematically related to one another, we can use
that information to predict whether an event or behavior will occur in a certain situation.
Example: Once I understand which factors drive managers’ success, I can predict whether a
job candidate I want to hire is likely to be (or become) a successful manager by thinking
about how they ‘score’ on the set of characteristics that would make a manager successful.
Scientific research is often classified as being either basic or applied:
Basic research Conducted primarily for the sake of achieving a more detailed and
accurate understanding of human behavior, without necessarily trying to address any
particular practical problem.
Applied research Conducted primarily to address some practical problem. Research on
the effects of cell phone use on driving, for example, was prompted by safety concerns and
has led to the enactment of laws to limit this practice.
Many papers have both a ‘basic’ and ‘applied’ element to them, and many excellent studies
contribute both to theoretical and practical understanding.
Chapter II: Thinking like a researcher
3. Concepts, constructs and variables
For many academics, understanding and explaining are the ultimate goal of doing research.
Explanations require development of concepts = generalizable properties or characteristics
associated with objects, events, or people:
3
, Examples: a person’s attitude toward immigrants, a firm’s capacity for innovation, a car’s
weight.
Some concepts have been developed over time through our shared language. Sometimes, we borrow
concepts from other disciplines or languages to explain a phenomenon of interest.
Concepts may have progressive levels of abstraction. Some concepts such as a person’s weight are
precise and objective, while other concepts such as a person’s personality may be more abstract and
difficult to visualize.
Construct = an abstract concept that is specifically chosen (or created) to explain a given
phenomenon. A construct can be unidimensional or multidimensional:
Unidimensional construct a simple concept, such as a person’s weight.
Multidimensional construct a combination of a set of related concepts, such as a person’s
communication skill, which may consist of several underlying concepts such as the person’s
vocabulary, syntax, and spelling.
Constructs used for scientific research must have precise and clear definitions that others can use to
understand exactly what it means and what it does not mean.
Example: A simple construct such as income may refer to monthly or annual income, before
or after-tax income, and personal or family income, and is therefore neither precise nor clear.
When discussing constructs, there are two types of definitions:
Conceptual definitions Involves defining a construct on an abstract and theoretical level.
Example: Perceived service quality the degree and direction of discrepancy between
consumers’ perceptions and expectations of a company’s service performance.
Operational definitions Are necessary when it comes to actually collecting and analyzing
data, and have to be very specific on how you will actually measure your construct.
Example: Perceived service quality a customer’s rating of a company on a 26-item survey
scale that consists of five underlying service quality dimensions: a company’s (service)
reliability, responsiveness, (perceived) empathy, assurance and performance on tangibles.
Example: Income monthly or annual, before- or after-tax, personal or family income.
Example: Temperature measured in Celsius, Fahrenheit or Kelvin scale.
Differences in operational definitions could lead to different findings, even if the conceptual
definition is the same.
A term that is frequently associated with a construct is a variable = a quantity that can vary (from low
to high, negative to positive etc.), in contrast to constants that do not vary (remain constant). In
science, a variable is a measurable representation of an abstract construct.
Constructs are not directly measurable. Therefore, we use variables.
Example: a person’s intelligence is often measured as his or her IQ score. In this case,
intelligence is a construct and IQ score is a variable that measures the intelligence construct.
Thinking like a researcher is translating between planes:
Theoretical plane Here, we discuss constructs, conceptual definitions and propositions in
an abstract sense.
4
, Empirical plane Here, constructs have operational definitions, are measured by variables,
about which we may have hypotheses.
Thinking like a researcher involves the ability to move back and forth between the more abstract
theoretical plane and the more specific and tangible empirical plane (see figure 1).
See an example of translating between both planes on page 11 of the RSR.
Conclusion of this example: The exact same empirical observation might lead to many
different ideas regarding how and why something occurs or works on the theoretical plane.
Figure 1: The theoretical and empirical planes of research
Depending on their intended use, variables can be classified as followed:
Independent variables Variables that explain other variables.
Dependant variables Variables that are explained by other variables.
Moderating variables (= moderator) Variables that strengthens or weakens the effect of
another variable, and can sometimes change the direction of the effect.
Example: We do research on how intelligence (independent variable) affects academic
success (dependent variable). We may expect that greater amounts of intelligence would be
related to greater amounts of academic success. However, there’s also effort to consider,
which could work as a moderator: effort could ‘strengthen’ the effect of someone’s
intelligence on academic success. Between two equally intelligent students, the student who
puts in more effort achieves greater academic success.
Note: moderators are not the same as independent variables. We’re not stating that students
who put in more effort have a higher IQ for example. Rather, the point is that the effect of
intelligence on academic success depends on a third variable (here: effort).
Mediating variables (intermediate) Variables that are explained by independent
variables, but explain dependent variables. These variables help us understand the
mechanism/process through which something works.
Example:
o Independent variable: service quality
o Hypothesis: service quality influences customer loyalty
o Idea: service quality doesn’t influence customer loyalty directly, but works ‘through’
customer satisfaction: a higher service quality makes customers more satisfied, and
more satisfied customers – in turn – become more loyal.
o Here, customer satisfaction would be the mediating variable.
Control variables Variables that have to take into account a in scientific study, but are
otherwise not of a main interest.
5
