You are on page 1of 39

Modern


Research 

Methods


How to write a better thesis


Dr. Sunu Wibirama

sunu@ugm.ac.id
http://wibirama.staff.ugm.ac.id

Department of Electrical Engineering



and Information Technology

Faculty of Engineering 

Universitas Gadjah Mada 

INDONESIA
Sunu Wibirama

Positions

❑ 2008 – now : Lecturer in Department of Electrical Engineering



and Information Technology, UGM, Yogyakarta
❑ 2015 : Research fellow in Tampere Unit for 

Computer-Human Interaction (TAUCHI), 

Tampere University, Finland
❑ 2016 – 2017 : Visiting researcher in Shibaura Institute of Technology, Japan


Education

❑ 2014 Dr.Eng. Science and Technology Tokai University, Tokyo, Japan


❑ 2010 M.Eng Electronics Engineering KMITL, Bangkok, Thailand
❑ 2007 B.Eng. Electrical Engineering UGM, Yogyakarta, Indonesia

Interests
❖ Human-computer interaction
❖ Eye tracking applications
❖ Virtual reality and human factors
❖ Modern research methods, technology commercialization
Warning: 

Content of presentation may have bias due to my
educational background. My background is
engineering / computer science, so some of my
suggestions may not work for different disciplines
How to survive in postgraduate life?
Admit it. It is your reality.
Tips #1: Read carefully, dilligently
Notes: academic research is derived from scientific arguments

not from magic/assumptions
Tips # 2: Prepare for unexpected things
Always prepare your “plan B, C, and even D”
Tips # 3: Work-life balance
Healthy life equals to productivity
A brief overview about thesis
What is Thesis

• Thesis : an extended argument with logical, structured,


and defensible reasoning based on credible evidence
to support original contribution to knowledge

• Master thesis : 1-2 years. 



PhD thesis: 3-5 years.
Scientific Contribution in 

Academic Research
contribution of 

previous work (A)
Border of world’s
knowledge
Your

contribution to the 

Your knowledge
world’s

knowledge
Your 

learning 

progress

contribution of 

previous work (B)

(Dawson, 2009)
From idea to innovation
Existing Idea Creativity New product
Types of novelty (scientific contribution) in
Electrical Engineering, Computer Science, and IT

• Working on old problem with new method (improvement-


type research)

• Implementing old method to new problem

• Implementing new method to new problem

• Proofing a theoretical model (computational) with real


implementation in hardware.

• Development of low-cost method/technique/hardware/


system with equal performance of existing method/
technique/hardware/
Food for thought

What is categorized as “scientific
contribution” in your discipline?
Ask your supervisor.
Please read this article

http://wibirama.staff.ugm.ac.id
Writing thesis is hard

• It is your responsibility, not your supervisor


• It takes quite long time and a lot of work
• Narration (story) is equally important to structure
• You are normally trained technically, not in scientific writing
• Procrastination, mental block, and “do first, write later” adagium
Successful Thesis?
• The thesis makes distinct contribution to knowledge

• The contribution is supported with credible and verifiable


evidence

• The thesis shows that the candidate has a thorough grasp


of appropriate solution for research problem and
awareness of their limitations.

• The thesis shows deep understanding, authority in


candidate’s field and broad knowledge in relevant fields.

• The thesis shows ability to communicate research


findings effectively in the professional arena and an
international context.
Basic questions of thesis examination

• Why you choose this research problem ?

• Why is it important ? (point of view from previous


works)

• How can you show that it is important ?

• How can you measure that your results are good?



(measurement’s metrics)

• How can you know that your results are truly good?
(comparing your results with others)
Some weaknesses inside academic thesis
• No specific research problem (too broad statement). Please avoid assumptions in your
introduction.

• Claim is not supported with correct data

• Scientific contribution (novelty) is not explained clearly. List so many previous works, but
don’t understand the difference between the proposed method and previous works.

• Working in very narrow object, but try to generalize the results for broader objects.

• No supporting data for “improvement-type” research (no comparison, no benchmark).

• No judgment on “pre-defined variable”, e.g.: participants, how to set threshold, good or


bad results, etc.

• Inconsistency of jargon / terminology.

• CompSci / IT: Wrong comparison 



(your work vs. result in paper, but using your own data, not public dataset). 

Solution: (1) using your own data: do your method and benchmark method(s)

(2) using public dataset: only do your method
Reasons for not writing
I don’t have any time for writing
I can’t write in my office
I’m not ambitious
My teaching comes first
I review papers regularly, but I don’t write myself
I don’t want to play the publications game
I’m too tired when I get home to do any writing
I resent giving up so much of my personal time
I do a lot of writing, just not for publication
No one will read it anyway
I’m probably just afraid of rejection
I don’t write well
Writing thesis
• Reading is hard, but writing is harder

• Use simple structure in sentences

• Start early, revise often.

• Drive reading with your writing

• Choose good and cozy environment

• Finding best performance (day or night?)

• Scheduled time
Writing process
• Prewriting (60%)
• Organize and synthesize information
• Define a take-away message
• Outline and road map


• Write draft (10%)


• Just keep writing 


• Rewrite and Revision (30%)


• Read out loud, cut the clutter, simple structure 

tips: read On Writing Well and PUEBI
• Correct mistakes
• Get feedback (conference, research group meeting)
Warning: 

Content of presentation may have bias due to my
educational background. My background is
engineering / computer science, so some of my
suggestions may not work for different disciplines
Thesis outline
and suggested writing order

• Abstract (6)
Background

Research problems

• Introduction (4/5) Aim and scope

Contribution
• Chapter 2: (1)

Related works (research positioning)
Literature Review
 Theoretical foundation (justification of selected

Theoretical background
 approach)

Research questions

• Materials and Methods (2)

• Results and Discussion (3)

• Conclusions (4/5)
Storyline of your thesis
Why your research is important for
general society ? What is important
1 General issue in society issue faced by society?

General problem that motivates

2 General problem research in your field

Try not to dump references by saying:



“A proposed B, resulting X… “ 


3 Literature synthesis 

Critically analyze advantage and
disadvantage of previous work. FInd
research gap from previous research

4 Research problems List all of research problems that you to


solve with your proposed solution.

Type of contribution:
- solving new problem with old method

- solving old problem with new method

Proposed method / solution Your interpretation to answer research


problems (no. 4):

• How far is your improvement, compared


5 Results with previous work?
• Implication of your research to previous
work (support or contradict ? Why?)
• Implications to general problem
(answering no. 1 and 2)
• Limitation of your method

6 Analysis and Discussion


• Generalization, explanation of any pattern
in your data, how error is produced, how
future work can be done to improve your
work

7 Conclusions
Summary of research problem and your solution 

to solve the problem. Take-away message from your 

research and general implications of your research
(c) sunu wibirama 2016
Tahapan penelitian v.s Tesis
Tahapan&Peneli+an& Susunan&Tesis&
1.#Pendahuluan:#
####3#Latar#Belakang#
1.#Iden(fikasi#Masalah& ####3#Rumusan#Masalah#
####3#Tujuan#Peneli(an#
####3#Manfaat#Peneli(an#
2.#Landasan#Teori:#
####3#Peneli(an#yang#Berhubungan#
2.#Perumusan#Hipotesis& ####3#Landasan#Teori#
####3#Kerangka#Pemikiran#
3.#Metodologi#Peneli(an:#
####3#Metode#Peneli(an#
3.#Pengujian&Hipotesis& ####3#Metode#Pengumpulan#Data#
####3#Metode#Analisis#Data#
dan#Analisis#Hasil#
####3#Metode#Pengukuran#Peneli(an#
4.#Analisis#Hasil#dan#Pembahasan#
4.#Kesimpulan& 5.#Kesimpulan#dan#Saran#

(Romi Satria Wahono, 2013) 25


Abstract
• Completely independent of your
thesis
• People read abstract to know if
they will read your thesis or not
• Overview of the main story:
• Why you are doing it 

(background, your goal)
• How you are doing it 

(methods, key results)
• How it relates to the real world
(conclusions, implications)
Introduction
• Introduce / define your area in general.

• Put general issue faced by society to put your thesis in broader context

• What is known

• What is unknown (research gap)

• Your question and your goal

• How will you solve your question

• Why is your approach different and important

• “So what” test

• Roadmap of your thesis


140 Research Gap Chapter 2

Reference: M.J. Katz, “From research to manuscript”, 2nd Edition, Springer, 2009
Literature review and theoretical background
• Literature review:

• Give the big picture of your field

• Give context and introduce state-of-the-art work in your field

• Introduce the limitations of these works and why these works don’t
solve your problems

• Give hints on what your solution will have to address

• Theoretical background:

• Basic explanation to help others appreciating your proposed solution

• State your judgment of your proposed solution. Remember: every


part of proposed solution must come with scientific reason. 

No “MAGIC NUMBER” is allowed.

• State your hypothesis (optional)


Proposed solution
• Design of proposed solution: 

approach, system, algorithms, frameworks, models, etc.

• What is the data for your research (dataset) and how to


obtain it ? Primary data: you get your own data
Secondary data: use other’s data.

• Configuration of devices

• Design of experiment

• How to evaluate your proposed solution


Results and Discussion
• Improvement of previous work ? How much
improvement ?

• Explanation of pattern from your results

• Limitations of your work (trade-off)

• Implications of your results to answer your research


problem and to achieve your goal
Remember: Storyline of your thesis
Why your research is important for
general society ? What is important
1 General issue in society issue faced by society?

General problem that motivates

2 General problem research in your field

Try not to dump references by saying:



“A proposed B, resulting X… “ 


3 Literature synthesis 

Critically analyze advantage and
disadvantage of previous work. FInd
research gap from previous research

4 Research problems List all of research problems that you to


solve with your proposed solution.

Type of contribution:
- solving new problem with old method

- solving old problem with new method

Proposed method / solution Your interpretation to answer research


problems (no. 4):

• How far is your improvement, compared


5 Results with previous work?
• Implication of your research to previous
work (support or contradict ? Why?)
• Implications to general problem
(answering no. 1 and 2)
• Limitation of your method

6 Analysis and Discussion


• Generalization, explanation of any pattern
in your data, how error is produced, how
future work can be done to improve your
work

7 Conclusions
Summary of research problem and your solution 

to solve the problem. Take-away message from your 

research and general implications of your research
(c) sunu wibirama 2016
Preparing dissemination plan

• PhD and some Master program are research


degree. Publication is obligatory —> Publication
milestones.

• You should break down your work to smaller “goal”,


each of which presents a novelty.

• Expertise of supervisor: breaking down a big topic


to smaller topics. Consult your supervisor
Best practices to graduate on-time
• Know your limitation. Select only research topic that is suitable and
realiable.

• Prepare RESEARCH LOGBOOK.

• Set a definite schedule and stick with it.

• Allocate minimum 4 hours/day for research 



(if you are grad student in Japanese university, you spend 8 hours/day for
research).

• Do it in your lab / campus. 



Don’t bring it home working along with your family.

• Be proactive. Know how to deal with your advisor.

• Report your progress weekly / bi-weekly to your advisor.

• Scheduled writing session. Disconnect internet.

• Take your research project seriously.


Ending your holiday without a new headache
RESEARCH LOGBOOK
LEARNING ROADMAP
Do you know your destination and when you will arrive?
Gantt Chart

(Dawson, 2009)
Warning: 

Content of presentation may have bias due to my
educational background. My background is
engineering / computer science, so some of my
suggestions may not work for different disciplines
“Thank you”
http://wibirama.staff.ugm.ac.id

You might also like