Introduction to Business Research
Likert Scale = “on a scale of 1-5, how do you rate xxx?” – a quantifiable measuring tool, even
if questions and construct are highly subjective (how happy are you?)
Evidence-based management = the conscientious use of the best current evidence to make
decisions for management.
What are the characteristics of an experiment?
- Different test groups (control group, random sample of people)
- Definition of independent/dependent variable
- Ability to change only one independent variable at a time
Random sampling = test subjects are chosen to reflect the outside world as much as
possible: a wide range of age, sex, socio-economic
Random allocation = the people in the groups are randomly assigned
Validity = are you measuring what you want to measure?
External validity = is the experiment repeatable, applied to the “real world”?
Generalizability = is it representative of the general population?
Internal Validity = causality: does the manipulation of the independent variable cause
change in the dependent variable, all other variables the same. Are all external factors
accounted for, self-selecting groups? Is the model valid?
Construct Validity = how are concepts (concepts= what is being measured: altruism, earning
power) measured: how it is operationalized? How do we define “married”, “earnings”? These
are all “measurement methods”: how valid are they?
Selection hypothesis = are your groups self-selecting? Do they fall into certain categories
first and then get measured? Is the model accounting for this? Otherwise we can only study a
correlation.
Reliability = can the experiment be replicated? Are the tools used to measure reliable? Clear
procedures. If you do the test twice, do you get the same result?
Impact factor = how research is rated, measured and scored. Based on the number of
citations.
“Between-subject” design = one subject per test, they do not repeat it.
“Within-subject” design = participants repeat the test, each group.
Primary data = data we collect ourselves
Secondary data = data that already exists
Research philosophy, research paradigms, theory development
Publication bias = journals are more interested in publishing positive results which have
significance to a wide audience, and also first studies more interesting to them than rebuttal
studies, second studies
Construct = an idea or theory containing various conceptual elements: subjective and not
based on empirical evidence
The “research onion”:
,The outer layers are the first stages of research, as you refine your question and collect data,
it gets closer to the center; we start with our own philosophies and outlook
What is research philosophy?
- How we see and understand knowledge
- How knowledge acquisition is best conducted
Ontology = the nature of reality
, Epistemology = the nature of knowledge acquisition; what kind of researcher are we: natural
scientist or ethnographer?
Axiology = the nature of values, value systems and its effect on research; can research be
value-free or is it always subjective? Our values are the basic drive of all human actions, and
axiology is how we interpret it.
Positivist Interpretivist
Objectivism = social reality is fixed and Subjectivism = we are all social actors, we all
observable, does not depend on the point perceive reality differently. Reality is fluid and
of view of the researcher. There is one depends on the point of view of the researcher.
reality that can be measured and There are multiple realities, which are hard to
quantified. quantify. The world is socially constructed
We can be “value-free” We are all “values-bound”
The natural scientist (epistemology) = The Social Scientist = uses broad, open-ended
uses experiments, observable reality, interviews, a smaller sample. Enters world of
large sample sizes, detached. Structured subjects, studies observation in context of
methodology. Quantifiable approach, subjects environment.
closed-questions. Experiments can be
repeated. Studies data in isolation from Interest in stories over numbers
subjects, does not enter world of
subjects. Induction = starts with broad observations to find
patterns and correlations, ends with a theory:
Interest in numbers over stories.
specific > general: “all swans we see are white,
Deduction = starts with a theory, theory is we can expect that all swans everywhere are
tested and adjusted, conclusion is white as well”
reached
Research outcomes = “grounded theory”
general > specific: “all men are mortal,
Socrates was a man, Socrates was Example = Venkatesh (2008) Gang Leader
mortal”
Constructivism = reality and all our norms and
Research outcomes = conclusions values are not fixed, but constructed in the minds
of each individual (social) actor
Example = List (2007) Dictator Games
experiment
The middle ground
Pragmatist = picks the philosophy that best fits the research question: which research
philosophy to choose depends on the question we wish to answer
Abduction = connects deduction to induction, in the middle.
Critical Realism = a philosophy combining positivism and interpretivist
Realism = “reality” is the most important consideration
Levit and Dubner = “freakonomics” microeconomics researchers, tried to strip away values
and moralities to study prostitution; behavioral economics: we only have one macro economy
to “experiment” with, but micro economies can be studied more carefully.
Reliability vs validity
Likert Scale = “on a scale of 1-5, how do you rate xxx?” – a quantifiable measuring tool, even
if questions and construct are highly subjective (how happy are you?)
Evidence-based management = the conscientious use of the best current evidence to make
decisions for management.
What are the characteristics of an experiment?
- Different test groups (control group, random sample of people)
- Definition of independent/dependent variable
- Ability to change only one independent variable at a time
Random sampling = test subjects are chosen to reflect the outside world as much as
possible: a wide range of age, sex, socio-economic
Random allocation = the people in the groups are randomly assigned
Validity = are you measuring what you want to measure?
External validity = is the experiment repeatable, applied to the “real world”?
Generalizability = is it representative of the general population?
Internal Validity = causality: does the manipulation of the independent variable cause
change in the dependent variable, all other variables the same. Are all external factors
accounted for, self-selecting groups? Is the model valid?
Construct Validity = how are concepts (concepts= what is being measured: altruism, earning
power) measured: how it is operationalized? How do we define “married”, “earnings”? These
are all “measurement methods”: how valid are they?
Selection hypothesis = are your groups self-selecting? Do they fall into certain categories
first and then get measured? Is the model accounting for this? Otherwise we can only study a
correlation.
Reliability = can the experiment be replicated? Are the tools used to measure reliable? Clear
procedures. If you do the test twice, do you get the same result?
Impact factor = how research is rated, measured and scored. Based on the number of
citations.
“Between-subject” design = one subject per test, they do not repeat it.
“Within-subject” design = participants repeat the test, each group.
Primary data = data we collect ourselves
Secondary data = data that already exists
Research philosophy, research paradigms, theory development
Publication bias = journals are more interested in publishing positive results which have
significance to a wide audience, and also first studies more interesting to them than rebuttal
studies, second studies
Construct = an idea or theory containing various conceptual elements: subjective and not
based on empirical evidence
The “research onion”:
,The outer layers are the first stages of research, as you refine your question and collect data,
it gets closer to the center; we start with our own philosophies and outlook
What is research philosophy?
- How we see and understand knowledge
- How knowledge acquisition is best conducted
Ontology = the nature of reality
, Epistemology = the nature of knowledge acquisition; what kind of researcher are we: natural
scientist or ethnographer?
Axiology = the nature of values, value systems and its effect on research; can research be
value-free or is it always subjective? Our values are the basic drive of all human actions, and
axiology is how we interpret it.
Positivist Interpretivist
Objectivism = social reality is fixed and Subjectivism = we are all social actors, we all
observable, does not depend on the point perceive reality differently. Reality is fluid and
of view of the researcher. There is one depends on the point of view of the researcher.
reality that can be measured and There are multiple realities, which are hard to
quantified. quantify. The world is socially constructed
We can be “value-free” We are all “values-bound”
The natural scientist (epistemology) = The Social Scientist = uses broad, open-ended
uses experiments, observable reality, interviews, a smaller sample. Enters world of
large sample sizes, detached. Structured subjects, studies observation in context of
methodology. Quantifiable approach, subjects environment.
closed-questions. Experiments can be
repeated. Studies data in isolation from Interest in stories over numbers
subjects, does not enter world of
subjects. Induction = starts with broad observations to find
patterns and correlations, ends with a theory:
Interest in numbers over stories.
specific > general: “all swans we see are white,
Deduction = starts with a theory, theory is we can expect that all swans everywhere are
tested and adjusted, conclusion is white as well”
reached
Research outcomes = “grounded theory”
general > specific: “all men are mortal,
Socrates was a man, Socrates was Example = Venkatesh (2008) Gang Leader
mortal”
Constructivism = reality and all our norms and
Research outcomes = conclusions values are not fixed, but constructed in the minds
of each individual (social) actor
Example = List (2007) Dictator Games
experiment
The middle ground
Pragmatist = picks the philosophy that best fits the research question: which research
philosophy to choose depends on the question we wish to answer
Abduction = connects deduction to induction, in the middle.
Critical Realism = a philosophy combining positivism and interpretivist
Realism = “reality” is the most important consideration
Levit and Dubner = “freakonomics” microeconomics researchers, tried to strip away values
and moralities to study prostitution; behavioral economics: we only have one macro economy
to “experiment” with, but micro economies can be studied more carefully.
Reliability vs validity
