Topics in Business Economics – Lectures
Lecture 1 – Introduction
There are only a few theories/concepts that are directly from accounting. Accounting research/theory relies often on
other disciplines (e.g., economics, psychology, organizational behaviour).
- Examples of theories; agency theory, institutional theory, behavioural accounting theory
Theory; a formal set of ideas that is intended to explain why something happens or exists.
Hypothesis; an idea or explanation of something that is based on a few known facts but has not yet been proven to
be true or correct.
Proposition; similar to a hypothesis, but the link cannot be verified by experiment.
Theorem; a rule or principle, especially in mathematics, that can be proven to be true.
Law; a generalized rule, especially in natural sciences, in the form of a mathematical statement.
Basic Levels of research;
- Description: collection and reporting data (observations) and basic statistics
- Classification: grouping (or clustering) of data based on similarities or differences.
- Explanation: make sense of data by attempting to explain relationships (or causality).
o Generally concerned with the ‘Why?’
- Prediction: provide models that allow testable preditions for (future) events.
Other distinctions;
- Deductive reasoning; from theory to observations
- Inductive reasoning; from observations to theory
- Confirmatory research; tests a priori hypothesis, is based on existing study, has stringent research
restrictions, deals with knowns/unknowns.
- Exploratory research; generates a posteriori hypothesis, discovers new knowledge, has less stringent
research restrictions, deals with unknowns.
Examples;
- We collect salary data and find that women earn less than men inductive & confirmatory
- We investigate the effects of the Ukraine war on energy prices using AS/AD model deductive & exploratory
- We use agency theory to explain earnings management in Dutch firms deductive & confirmatory
- We use NASA data on black holes and find that they have an association with earnings smooting inductive & exploratory
! it depends on how much background info you have (much = confirmatory & deductive)
Lecture 2; formulating a research question and transitioning to hypotheses (or propositions)
The kinney three paragraphs; before you start a research, you consider…
- What..
o What is the problem?
o What are you trying to find out?
- Why..
o Why is it an imporant problem?
o To whom?
- How..
o How will it be solved?
o How will it be done?
Example Cade (2018)
What?
- There are different ways for management to react to criticism on social media. How do investors react to
remaining silent, providing an explanation or redirecting attention?
Why?
- We already know much about tranditional communication channels, but communication to stakeholders
increasingly takes place on social media.
- Important to know for companeis’ reputation how to react to criticism.
How?
- Experiment in which participants receive scenario (maniupulating the variable) and answer questions about their
perceptions of the firm and their decisions.
,A research question is a clear, focused, concice and arguable question around which you center our empirial
research project.
- Minimize room for misinterpretations
- It must be specific enough to be covered in a paper.
- It is not about personal opinions and you can formulate an objective answer (falsification is important)
! you just want to explain a part of what is not known about the phenomenon, not all of what is unknown.
What;
Research questions are often (not always) motivated both from theory and practice
Why;
What could be a good reason for you to spend your time and resources exploring the research question?
- What could be a good reason why?
o To improve decision-making in within a ocmpnay
o Because it changes the way people think about certain theories or issues
o It can improve policy making.
o It can help solve societal issues.
o Etc.
How;
- Not all research methods are empirical, for example archival research.
- Using surveys, case studies, controlled/field experiments.
- You make a distinction between collected and generated variables
o Collected; originating from naturally-occuring or human-managed data.
E.g. Number of chocolate bars at de Refter
E.g. Stock price of a publicly listed firm
E.g. Number of hours worked by an employee at a firm in a week
o Generated; data created by us researchers
E.g. Number of correct answeres on a test we created for a study
E.g. Answer to a question about happiness
E.g. Researcher intervention to change manager pays at some units but not others
- We have a controlled and an uncontrolled setting
o Controlled; setting restricts itself to the study of the variables of interest and not others.
In for example an Experiment
E.g. Asking students whether they prefer €5 or 50% change to get €10 in a laboratory
E.g. Eliciting people’s reponses to a hypothetical scenario researchers have created.
o Uncontrolled; setting has other potentially influential vairables besides the ones being studied.
In for example a Case study
E.g. Asking for data collected by an organization.
E.g. Approaching random respondents in the street
E.g. Buying data from a data collection firm
E.g. Observing how managers perform at a firm
- You can use standard or non-standard data sets to study your concept;
o Standard data sets; data from for example CBS, data companies
, o non-standard datasets; Internal documents from a company, Interview, Observations, Social media
The predictive validity framework; deepening the what and how
- You test whether what is said in the theoretical model is really happening in the empirical model (empirical
adequacy). You measure these theoretical concepts by using variables coming from operationalization
o A theory is a conceptual window to look at the overly complex observable world. Theories help you
to build a plausbile answer to the research question (hypothesis).
The concepts are thus abstractions describing parts of reality.
- Concepts; Intelligence, risk aversion, gender, depression, happiness
- Variables; numbers of books purchased, gender, age, stock price, happiness scale
Hypthesis are explicit statements (causally) relating concepts. It is about what you predict.
- H0 (null hypothesis); you don’t predict a relationship = 0
- Ha (alternative hypothesis); you do predict a relationship ≠ 0
! a good hypothesis is falsifiable, meaning that empirical evidence should be able to contradict it.
- Moderation hypothesis; an concept changes/modifies/interacts the effect between two concepts
- Mediation hypothesis; you predict that a concept explains the relationship between two concepts
Validity
- External validity; whether your research makes sense for the outside world (whether it can be generalized).
- Construct validity; the extent to which your test or measure accurately assesses what it's supposed to
- Internal validity; whether the study design/conduct/analysis answer the research questions without bias.
Association and causation
- Empirical research documents association and sometimes also causation.
What are good research papers?
- Published in a top journal
- Impact factor of its journal
- Reputation of the author (such as prizes and affiliations)
- Number of citations on the specific articles
How to search for ‘good’ research papers?
- Keywors technique; start with specific words and use broader words if necessary
- Snowball technique; use a ‘good’ researc h paper to find other ‘good’ papers
- Forward snowballing; looking at which paper is cited most.
How to read research papers?
- Your shouldn’t read everything, the abstract and introduction are most important.
Lecture 3; methodological and theoretical challenges of qualitative research
Perspectives differ. You cannot see two things at the same time. Our interpretations differ.
Every perspectives highlights something and obscures something else. It determines what you see.
- In science, there are also multiple perspectives of how we view reality.
Why multiple perspectives?
- Solving complex problems requires taking different perspectives.
- It learns you how to think out-of-the-box and out-of-the-text books, in order to be creative, novel and
independent.
- The ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function is a
test of intelligence.
Lecture 1 – Introduction
There are only a few theories/concepts that are directly from accounting. Accounting research/theory relies often on
other disciplines (e.g., economics, psychology, organizational behaviour).
- Examples of theories; agency theory, institutional theory, behavioural accounting theory
Theory; a formal set of ideas that is intended to explain why something happens or exists.
Hypothesis; an idea or explanation of something that is based on a few known facts but has not yet been proven to
be true or correct.
Proposition; similar to a hypothesis, but the link cannot be verified by experiment.
Theorem; a rule or principle, especially in mathematics, that can be proven to be true.
Law; a generalized rule, especially in natural sciences, in the form of a mathematical statement.
Basic Levels of research;
- Description: collection and reporting data (observations) and basic statistics
- Classification: grouping (or clustering) of data based on similarities or differences.
- Explanation: make sense of data by attempting to explain relationships (or causality).
o Generally concerned with the ‘Why?’
- Prediction: provide models that allow testable preditions for (future) events.
Other distinctions;
- Deductive reasoning; from theory to observations
- Inductive reasoning; from observations to theory
- Confirmatory research; tests a priori hypothesis, is based on existing study, has stringent research
restrictions, deals with knowns/unknowns.
- Exploratory research; generates a posteriori hypothesis, discovers new knowledge, has less stringent
research restrictions, deals with unknowns.
Examples;
- We collect salary data and find that women earn less than men inductive & confirmatory
- We investigate the effects of the Ukraine war on energy prices using AS/AD model deductive & exploratory
- We use agency theory to explain earnings management in Dutch firms deductive & confirmatory
- We use NASA data on black holes and find that they have an association with earnings smooting inductive & exploratory
! it depends on how much background info you have (much = confirmatory & deductive)
Lecture 2; formulating a research question and transitioning to hypotheses (or propositions)
The kinney three paragraphs; before you start a research, you consider…
- What..
o What is the problem?
o What are you trying to find out?
- Why..
o Why is it an imporant problem?
o To whom?
- How..
o How will it be solved?
o How will it be done?
Example Cade (2018)
What?
- There are different ways for management to react to criticism on social media. How do investors react to
remaining silent, providing an explanation or redirecting attention?
Why?
- We already know much about tranditional communication channels, but communication to stakeholders
increasingly takes place on social media.
- Important to know for companeis’ reputation how to react to criticism.
How?
- Experiment in which participants receive scenario (maniupulating the variable) and answer questions about their
perceptions of the firm and their decisions.
,A research question is a clear, focused, concice and arguable question around which you center our empirial
research project.
- Minimize room for misinterpretations
- It must be specific enough to be covered in a paper.
- It is not about personal opinions and you can formulate an objective answer (falsification is important)
! you just want to explain a part of what is not known about the phenomenon, not all of what is unknown.
What;
Research questions are often (not always) motivated both from theory and practice
Why;
What could be a good reason for you to spend your time and resources exploring the research question?
- What could be a good reason why?
o To improve decision-making in within a ocmpnay
o Because it changes the way people think about certain theories or issues
o It can improve policy making.
o It can help solve societal issues.
o Etc.
How;
- Not all research methods are empirical, for example archival research.
- Using surveys, case studies, controlled/field experiments.
- You make a distinction between collected and generated variables
o Collected; originating from naturally-occuring or human-managed data.
E.g. Number of chocolate bars at de Refter
E.g. Stock price of a publicly listed firm
E.g. Number of hours worked by an employee at a firm in a week
o Generated; data created by us researchers
E.g. Number of correct answeres on a test we created for a study
E.g. Answer to a question about happiness
E.g. Researcher intervention to change manager pays at some units but not others
- We have a controlled and an uncontrolled setting
o Controlled; setting restricts itself to the study of the variables of interest and not others.
In for example an Experiment
E.g. Asking students whether they prefer €5 or 50% change to get €10 in a laboratory
E.g. Eliciting people’s reponses to a hypothetical scenario researchers have created.
o Uncontrolled; setting has other potentially influential vairables besides the ones being studied.
In for example a Case study
E.g. Asking for data collected by an organization.
E.g. Approaching random respondents in the street
E.g. Buying data from a data collection firm
E.g. Observing how managers perform at a firm
- You can use standard or non-standard data sets to study your concept;
o Standard data sets; data from for example CBS, data companies
, o non-standard datasets; Internal documents from a company, Interview, Observations, Social media
The predictive validity framework; deepening the what and how
- You test whether what is said in the theoretical model is really happening in the empirical model (empirical
adequacy). You measure these theoretical concepts by using variables coming from operationalization
o A theory is a conceptual window to look at the overly complex observable world. Theories help you
to build a plausbile answer to the research question (hypothesis).
The concepts are thus abstractions describing parts of reality.
- Concepts; Intelligence, risk aversion, gender, depression, happiness
- Variables; numbers of books purchased, gender, age, stock price, happiness scale
Hypthesis are explicit statements (causally) relating concepts. It is about what you predict.
- H0 (null hypothesis); you don’t predict a relationship = 0
- Ha (alternative hypothesis); you do predict a relationship ≠ 0
! a good hypothesis is falsifiable, meaning that empirical evidence should be able to contradict it.
- Moderation hypothesis; an concept changes/modifies/interacts the effect between two concepts
- Mediation hypothesis; you predict that a concept explains the relationship between two concepts
Validity
- External validity; whether your research makes sense for the outside world (whether it can be generalized).
- Construct validity; the extent to which your test or measure accurately assesses what it's supposed to
- Internal validity; whether the study design/conduct/analysis answer the research questions without bias.
Association and causation
- Empirical research documents association and sometimes also causation.
What are good research papers?
- Published in a top journal
- Impact factor of its journal
- Reputation of the author (such as prizes and affiliations)
- Number of citations on the specific articles
How to search for ‘good’ research papers?
- Keywors technique; start with specific words and use broader words if necessary
- Snowball technique; use a ‘good’ researc h paper to find other ‘good’ papers
- Forward snowballing; looking at which paper is cited most.
How to read research papers?
- Your shouldn’t read everything, the abstract and introduction are most important.
Lecture 3; methodological and theoretical challenges of qualitative research
Perspectives differ. You cannot see two things at the same time. Our interpretations differ.
Every perspectives highlights something and obscures something else. It determines what you see.
- In science, there are also multiple perspectives of how we view reality.
Why multiple perspectives?
- Solving complex problems requires taking different perspectives.
- It learns you how to think out-of-the-box and out-of-the-text books, in order to be creative, novel and
independent.
- The ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function is a
test of intelligence.
