Summary Knowledge Clips Academic Project – Pre-Master Business Administration UvA
2021-2022
Week 1 knowledge clips
1.1 Foundations of research, research as a language, what is (business administration)
research
What characterized the scientific mindset? Research is about producing new and VALID
knowledge. How do we know which knowledge to trust? -> I doubt therefore I am (Rene
Descartes), doubting everything around him.
In this course, we hope to teach some of the fundamental characteristics and aspects
related to generating new knowledge. Research design, validity of research.
Research is purposeful, systematic, empirical, public, cumulative, critical of itself.
Business and management research is transdisciplinary, double hurdle (theoretical and
practical impact), science -practice gap/translational research. Evidence Based
Management.
Summing it all up
- Researchers systematically question or doubt things to improve some outcome
- Learning about research methods is akin to learning a new language
- Research produces or tests theory – (a set of) expectations about the nature of
reality
- Research is purposeful, systematic, empirical, public, cumulative, and critical of itself
- Business and management research is transdisciplinary.
1.2 The structure of research & the Hourglass model
Most research articles are taking up this Hourglass
model.
Broad, specific, broad again.
Research proposal is half of the study, general,
methods, results.
Empirical academic articles use this structure. You
know where to find what you’re looking for.
Hypotheses at end of introduction, etc.
Summing it all up
- Research reports typically take the for of the Hourglass model
- The Hourglass model is useful for
Efficiently finding that information that you are after
Structuring your own report
, - A research proposal is essentially s good first draft of the introduction and method of
the final report.
1.3 Types of research (and demands on design)
Basic research satisfies our curiosity. Basic research is done a lot in universities.
Applied research is conducted to find answers to ‘real life’ problems.
Types of studies
By evidence:
- Descriptive studies
- Relational studies
- Explanatory / causal studies
By time:
- Cross-sectional (all data collected at one time)
- Longitudinal (data collected over time)
Repeated measures (<20 measurement waves)
Time series (>20 measurement waves)
The holy grail of research is to establish a causal relationship. But it’s also one of the hardest
things to do since you have to exclude all other variables and prove that the one causes the
other and not the other way around.
Types of data
- Qualitative data (‘built theory’)
Data that are in the form of text, pictures, sounds etc.
Much more difficult to analyze. Initial theories are formed through qualitative data,
only then, we are able to use quantitative data (measurements for these constructs)
- Quantitative data (‘test theory’)
Data that is in numeric form
Summing it all up
- We can distinguish studies by the strength of the evidence they generate
(descriptive, exploratory, explanatory) or the type of data they analyze (qualitative vs
quantitative)
- Stronger evidence requires tougher designs
, - Cross-sectional studies cannot be used to identify causal relationships (all data
collected at once)
1.4 The relationship between theory and data; Inductive reasoning; deductive reasoning;
Hypothesis development, types of hypotheses, and hypothesis testing; Types of
relationships
Good theory includes a plausible, coherent, parsimonious explanation for why certain
cause and effect relationships should be expected.
Or more prosaically it is that we think we know about a particular phenomenon.
Relationship between theory and data (zipper between theory and data is knowledge)
Inductive reasoning (from data to theory)
Observation -> Pattern -> testing hypothesis -> theory
Deductive reasoning (from theory to data)
Theory -> Hypothesis -> Observation -> Confirmation
Hypothesis
Specific statement of prediction
Alternative hypothesis
A specific statement of prediction stating what you expect will happen in your study
Null hypothesis
A specific statement that predicts there will be no effect of a program, treatment or
other independent variable you are studying.
, Summing it all up
- Theory allows us to explain what is going on
- Theory is as good as the data that supports it, and needs to be modified if the data
do not
- All else being the same, the simplest theory is best
- We can use inductive reasoning to generate theory and deductive reasoning to test
theory
- Hypothesis allow us to test theory by specifying expected relationships between
constructs
1.5 Introduction to causality; independent/dependent variables; only design can support
causal conclusions.
Causality as the ‘holy grail’ of research
Quantitative variable
The numerical representation of some object
Attribute
A specific value of a carriable
A simplified way of thinking about a quantitative variable is that all attributed are
known and coded with a number prior to data collection
Independent variable
The variable that you manipulate (the cause)
Dependent variable
The variable that is affected by the independent variable
Unit of analysis
The entities about which we want to draw conclusions
Causal relationships as a special type of relationship
Not all relationships that we observe are causal
Reversed causality, third variable. Chance
2021-2022
Week 1 knowledge clips
1.1 Foundations of research, research as a language, what is (business administration)
research
What characterized the scientific mindset? Research is about producing new and VALID
knowledge. How do we know which knowledge to trust? -> I doubt therefore I am (Rene
Descartes), doubting everything around him.
In this course, we hope to teach some of the fundamental characteristics and aspects
related to generating new knowledge. Research design, validity of research.
Research is purposeful, systematic, empirical, public, cumulative, critical of itself.
Business and management research is transdisciplinary, double hurdle (theoretical and
practical impact), science -practice gap/translational research. Evidence Based
Management.
Summing it all up
- Researchers systematically question or doubt things to improve some outcome
- Learning about research methods is akin to learning a new language
- Research produces or tests theory – (a set of) expectations about the nature of
reality
- Research is purposeful, systematic, empirical, public, cumulative, and critical of itself
- Business and management research is transdisciplinary.
1.2 The structure of research & the Hourglass model
Most research articles are taking up this Hourglass
model.
Broad, specific, broad again.
Research proposal is half of the study, general,
methods, results.
Empirical academic articles use this structure. You
know where to find what you’re looking for.
Hypotheses at end of introduction, etc.
Summing it all up
- Research reports typically take the for of the Hourglass model
- The Hourglass model is useful for
Efficiently finding that information that you are after
Structuring your own report
, - A research proposal is essentially s good first draft of the introduction and method of
the final report.
1.3 Types of research (and demands on design)
Basic research satisfies our curiosity. Basic research is done a lot in universities.
Applied research is conducted to find answers to ‘real life’ problems.
Types of studies
By evidence:
- Descriptive studies
- Relational studies
- Explanatory / causal studies
By time:
- Cross-sectional (all data collected at one time)
- Longitudinal (data collected over time)
Repeated measures (<20 measurement waves)
Time series (>20 measurement waves)
The holy grail of research is to establish a causal relationship. But it’s also one of the hardest
things to do since you have to exclude all other variables and prove that the one causes the
other and not the other way around.
Types of data
- Qualitative data (‘built theory’)
Data that are in the form of text, pictures, sounds etc.
Much more difficult to analyze. Initial theories are formed through qualitative data,
only then, we are able to use quantitative data (measurements for these constructs)
- Quantitative data (‘test theory’)
Data that is in numeric form
Summing it all up
- We can distinguish studies by the strength of the evidence they generate
(descriptive, exploratory, explanatory) or the type of data they analyze (qualitative vs
quantitative)
- Stronger evidence requires tougher designs
, - Cross-sectional studies cannot be used to identify causal relationships (all data
collected at once)
1.4 The relationship between theory and data; Inductive reasoning; deductive reasoning;
Hypothesis development, types of hypotheses, and hypothesis testing; Types of
relationships
Good theory includes a plausible, coherent, parsimonious explanation for why certain
cause and effect relationships should be expected.
Or more prosaically it is that we think we know about a particular phenomenon.
Relationship between theory and data (zipper between theory and data is knowledge)
Inductive reasoning (from data to theory)
Observation -> Pattern -> testing hypothesis -> theory
Deductive reasoning (from theory to data)
Theory -> Hypothesis -> Observation -> Confirmation
Hypothesis
Specific statement of prediction
Alternative hypothesis
A specific statement of prediction stating what you expect will happen in your study
Null hypothesis
A specific statement that predicts there will be no effect of a program, treatment or
other independent variable you are studying.
, Summing it all up
- Theory allows us to explain what is going on
- Theory is as good as the data that supports it, and needs to be modified if the data
do not
- All else being the same, the simplest theory is best
- We can use inductive reasoning to generate theory and deductive reasoning to test
theory
- Hypothesis allow us to test theory by specifying expected relationships between
constructs
1.5 Introduction to causality; independent/dependent variables; only design can support
causal conclusions.
Causality as the ‘holy grail’ of research
Quantitative variable
The numerical representation of some object
Attribute
A specific value of a carriable
A simplified way of thinking about a quantitative variable is that all attributed are
known and coded with a number prior to data collection
Independent variable
The variable that you manipulate (the cause)
Dependent variable
The variable that is affected by the independent variable
Unit of analysis
The entities about which we want to draw conclusions
Causal relationships as a special type of relationship
Not all relationships that we observe are causal
Reversed causality, third variable. Chance
