How to define your topic?
1
MSc Business Information Systems
Project 1: Applying Research Methodologies
Learning goals
■ understand the difference between research questions and
hypotheses
■ learn how to make your topic graspable in the form of a
research question
■ learn how to break down research questions into hypotheses
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
2
MSc Business Information Systems
Project 1: Applying Research Methodologies
MOTIVATION
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
3
MSc Business Information Systems
Project 1: Applying Research Methodologies
Motivation
■ A good thesis statement saves you time because it
♦ defines the direction of investigation
♦ sets boundaries
♦ helps to clearly define what needs to be done, e.g. what data needs to
be gathered, which questions need to be asked in interviews etc.
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
4
MSc Business Information Systems
Project 1: Applying Research Methodologies
DEFINITIONS
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
5
MSc Business Information Systems
Project 1: Applying Research Methodologies
Thesis Statement – Research Question
■ At the beginning you have to define what you will
investigate in your research
■ This can be done as a
♦ Research Question
♦ Hypothesis
■ Hypothesis and research question are equivalent. They differ
only in the kind of phrasing
♦ A hypothesis is an assertion that you put forward as being
(supposedly) true and which you will investigate
♦ Instead of phrasing a statement you can also ask a research question
that you will answer
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
6
MSc Business Information Systems
Project 1: Applying Research Methodologies
Research question
■ “A research question is a narrow, challenging question
addressing an issue, problem, or controversy that is answered
with a conclusion based on the analysis and
interpretation of evidence.” (Lipowski 2008)
■ “The research question is a logical statement that progresses
from what is known or believed to be true (as determined by
the literature review) to that is unknown and requires
validation.” (Marion 2004)
(E. E. Lipowski 2008: Developing great research questions. In American Journal of
Health-System Pharmacy, Vol. 65, Issue 17)
R. Marion (2004). The whole art of deduction. Research skills for new scientists. Webbook
available at http://www.sahs.utmb.edu/pellinore/intro_to_research/wad/wad_home.htm.
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
7
MSc Business Information Systems
Project 1: Applying Research Methodologies
(Directional) Hypothesis
■ A (directional) hypothesis is a statement that expresses the
probable relationship between variables. (Marion 2004)
♦ Variables need to be measurable
♦ Usually there is an independent variable and a dependent variable
♦ Example: “Customer satisfaction increases the profit of a company”
● IV: customer satisfaction
● DV: profit of a company
♦ Question: Are the variables measurable? How can we measure
them?
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
8
MSc Business Information Systems
Project 1: Applying Research Methodologies
(Open) Hypothesis
■ An open hypothesis (or question) asks for possible
relationships between variables
♦ usually formulated as “what are the variables influencing variable X?”
♦ X needs to be a measurable variable!
♦ Example: “what are the factors that help to increase the profit of a
company?”
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
9
MSc Business Information Systems
Project 1: Applying Research Methodologies
DERIVING QUESTIONS AND
HYPOTHESES
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
10
MSc Business Information Systems
Project 1: Applying Research Methodologies
Deriving research question / hypothesis
■ “…progress from what is known to what is unknown…”
♦ what is known:
● primarily literature research
● supported by practice: interviews, focus groups, surveys,
observations
● guided by the description of your topic
♦ questions need to be re-examined after each further step in the
research process, especially after the literature review
● Ideally, literature research and formulation of research questions is
done in parallel
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
11
MSc Business Information Systems
Project 1: Applying Research Methodologies
Thesis Statement – Phrasing a hypothesis
■ A research hypothesis is a guess, an unproven assertion that
you will investigate in your dissertation.
■ A hypothesis is a clear position regarding the problem that
you have identified, and you either argue for that position,
considering the evidence for and against, or you develop a
way of (empirically) testing it.
■ A thesis statement allows the researchers to precisely define
what they will investigate.
♦ It gives clear boundaries and a clear reason to do what you do.
♦ Whenever you do anything in your research, the reason should always
be ‘because it is necessary in order to assess my thesis statement'
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
12
MSc Business Information Systems
Project 1: Applying Research Methodologies
Criteria for Thesis Statements / Research
Questions
■ A thesis statement doesn’t name a topic or a problem that
you’ve identified. It must take a stand about something.
■ It must give you something to argue, to test, to prove, to
probe.
You are not judged whether you proved your thesis statement
right or wrong. You are judged on whether you investigated a
worthwhile issue in a reliable manner and came to a wellsubstantiated conclusion about it.
Source: Alta van der Merwe
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
13
MSc Business Information Systems
Project 1: Applying Research Methodologies
Criteria for Thesis Statements / Research
Questions
■ The following are things to consider when you write your
thesis statement.
♦ the thesis statement should be
● worthwhile
● limited in scope
● feasible in terms of primary sources, i.e. can you get the
data that supports your thesis statement
● doable in a reasonable amount of time
● testable by you
♦ the assertion should be unambiguous
♦ do not focus on more than one topic within one thesis
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
Source: Alta van der Merwe 14
MSc Business Information Systems
Project 1: Applying Research Methodologies
Deriving Research Questions / Thesis Statements
– Example
■ Research topic:
■ An IT system should be developed that supports customer consultants
in a bank in recommending financial products.
■ What is known:
♦ Customer consultants of a bank have to take into account many
information sources in order to recommend financial products that fit
the needs of their customers. It is difficult to observe and consider all
relevant information and to assess future development of the financial
market.
■ What is unknown:
♦ How can the consultants best be supported in their work?
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
15
MSc Business Information Systems
Project 1: Applying Research Methodologies
Examples of Thesis Statements for the Financial
Recommendation System
■ It is possible to develop an IT system that improves the quality of
recommendations for financial products.
■ It is possible to develop an IT system that makes better
recommendations for financial products than a human consultant.
■ It is possible to develop an IT system that makes consulting for
finanical products more efficient.
■
Remark: These are different thesis statements for the same research topic. The thesis
statements differ not only in phrasing but also determine research methodology and
evaluation. While the first two statements are about the effectivity of the system (evaluation is
by comparing the recommendations derived), the third statement is about the efficiency of the
processes (it would accept also recommendations that are not better than former ones).
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
16
MSc Business Information Systems
Project 1: Applying Research Methodologies
Detailing your research
■ The thesis statement resp. the research question leads to
more detailed hypotheses
■ Instead of hypothesis it is also possible to define sub research
questions
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
17
MSc Business Information Systems
Project 1: Applying Research Methodologies
Breaking down your research (1)
■ You can start with a research question or a thesis statement
and then break it down into hypotheses or subquestion
♦ Usually, several hypotheses/research questions can be derived /
thought of, when starting from one research question
■ Breaking down your research works by
♦ thinking about the variables (and how they can be measured)
♦ making guesses about their relationships
■ When you design an artefact or method…
♦ … the dependent variables of hypotheses are often measurable
variables that represent or correlate with the benefit of the designed
artefact/method
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
18
MSc Business Information Systems
Project 1: Applying Research Methodologies
Examples of Sub Questions for the Financial
Recommendation System
■ What information helps to increase the accuracy of the price
prediction of a financial product?
■ Which factors have an influence on the future price of a
financial product?
■ How should the information be presented to support the
customer consultant? (dependent variables: Speed of
decision, quality of decisions)
♦ Which visual presentation helps to increase the speed of decision
making?
♦ Which visual presentation helps to increase the quality of decisions?
Prof. Dr. Knut Hinkelmann
Dr. Hans Friedrich Witschel
19