Vous êtes sur la page 1sur 35

Computational

Research
Methodology
Lecture No. 1
Dr. Akhtar Hussain Jalbani

Recommended Readings
Dawson, Catherine, 2002, Practical
Research Methods, New Delhi, UBS
Publishers Distributors.
Kothari, C.R.,1985, Research MethodologyMethods and Techniques, New Delhi, Wiley
Eastern Limited.
Kumar, Ranjit, 2005, Research MethodologyA Step-by-Step Guide for Beginners,
(2nd.ed.),Singapore, Pearson Education.
How to Research. By Loraine Blaxter
Lecture notes.

What is Research?

Re + Search
Re = Again, Anew or Over again
Search = To examine closely and carefully
Research is a careful, systematic, patient study and
investigation in some field of knowledge, undertaken
to establish facts or principles.
Research is a structured enquiry that utilizes
acceptable scientific methodology to solve problems
and create new knowledge that is generally
applicable.
Research is a process of collecting, analyzing and
interpreting information to answer questions.
A combination of investigation of past work and effort
in the present that will help others in the future.
Finding or developing something new that changes
the world.

Characteristics of Research
Systematic (A logical sequence of steps)
Valid and Verifiable (Your findings are
correct and may be verified by you and
others)
Empirical (Conclusion drawn are based on
evidence collected from real life
experiences or observations)
Critical (The process of investigation must
be foolproof and free from drawbacks)

Types of Research
Research can be classified from three
perspectives:
application of research study
objectives in undertaking the research
inquiry mode employed

Classification based on application:


Pure research
Applied research

Types of Research (Cond)


Pure research, basic research, or
fundamental research is a research carried out
to increase understanding of fundamental
principles.
Pure research explain how this world operates,
what makes things happen, why social relations
are a certain way, and why society changes.
Pure research is the source of most new
scientific ideas and ways of thinking about the
world.
Pure research generates new ideas, principles
and theories, which may not be immediately
utilized; though are the foundations of modern
progress and development in different fields.

Types of Research (Cond)


Applied research is done to solve
specific, practical questions and
understanding of a observable fact.
Applied research can be exploratory,
but is usually descriptive.
Applied research can be carried out by
academic or industrial institutions.
Often, an academic institution such as a
university will have a specific applied
research program funded by an industrial
partner interested in that program.

Types of Research (Cond)


Classification based on objectives:

Descriptive
Correlational
Explanatory
Exploratory

Descriptive research also known as statistical


research, describes data and characteristics about
the population or phenomenon being studied.
Often the best approach, prior to writing
descriptive research, is to conduct a survey
investigation.
Descriptive research deals with everything that
can be counted and must have an impact to the
lives of the people around you.

Types of Research (Cond)


Correlational research tests for
statistical relationships between
variables.
The researcher begins with the idea
that there might be a relationship
between two variables.
She or he then measures both variables
for each of a large number of cases and
checks to see if they are in fact related.

Types of Research (Cond)


For example, imagine that a health
psychologist is interested in testing the
claim that people with more friends tend
to be healthier. She surveys 500 people
in her community, asking them how
many friends they have and getting
some measure of their overall health.
He or she then makes a graph showing
the correlation between these two
variables.

Types of Research (Cond)


Explanatory research attempts to clarify
why and how there is a relationship
between two or more aspects of a situation
or phenomenon.
Explanatory research attempts to identify
the actual reasons a phenomenon occurs.
An example of explanatory research would
be a study that finds that Christian couples
are twice as likely to divorce as Jewish
couples.

Types of Research (Cond)


Exploratory research is conducted into
an issue or problem where there are few or
no earlier studies to refer to.
The focus is on gaining insights and
familiarity for later investigation.
Here data is often quantitative and
statistics applies.
It is used to identify and obtain information
on a particular problem or issue.

Types of Research (Cond)


Classification based on inquiry mode:
Structured approach
Unstructured approach

The structured approach to inquiry is


usually classified as quantitative research.
Here everything that forms the research
process- objectives, design, sample, and
the questions that you plan to ask of
respondents- is predetermined.

Types of Research (Cond)


The structured approach is appropriate
to determine the extent of a problem,
issue or phenomenon.
The structured approach addresses the
problems such as:
how many people have a particular problem?
How many people hold a particular attitude?

The unstructured approach to inquiry is


usually classified as qualitative research.

Types of Research (Cond)


The unstructured approach is more
appropriate to explore the nature of a
problem, issue or phenomenon
without quantifying it.
Common examples include:
An account of different opinions different
people have about a product or an issue.
description of working condition in a
particular industry.

Quantitative vs. Qualitative


Research
Quantitative use of statistical, formulaic
or numerical analysis to generate results
Main approach: analysis; causal
determination, prediction, generalization of
findings
Results: This solution is N% better

Qualitative not quantitative; use of nonnumeric techniques


Main approach: discovery; illumination,
understanding, extrapolation to similar
circumstances
Results: This is a new way of solving our
problem

Scope of Research
Varies by level of work
Ph.D. students contribution expected at
world level; e.g.
background investigation on all past work
make meaningful addition to world knowledge

Undergraduate students contribution can


be at local to national to world level; e.g.
background investigation at university up to
world level
make meaningful addition to university up to
world level of knowledge

What Isnt Research

Playing with technology


Book report
Programming project
Doing what others have already done

However, each of these can be done


as part of research

Who Does Research?


Graduate Students
Masters Degree (lower standard)
Ph.D. Degree (higher standard)

Researchers at universities
Post-Doctoral students
Faculty members

Researchers in industry
Research scientists
Many other technical workers

Undergraduate students

Who Does Research? (2)


Individuals
Teams
Teams almost always make the
process easier
Division of labor
Feedback from team members
Each member can work to own strengths

Research Process
(Methodology)

Initial Idea
Background Investigation
Refinement of Idea
Core Work
Investigation and Development
Documentation
Prototype (if appropriate)

Evaluation
Identification of Future Work
Presentation

Research Process Initial


Idea
Stems from critical thinking
Be on the lookout for and open to seeing problems
Gaps in framework
Repetitive behavior thats slightly different (and can be
generalized)
Manual solutions (that can be automated)
Inelegant solutions

Ask questions
Is something missing here?
Can this be done in a better way?
Is there a need for a new approach?

Should be an area youre interested in, as:


Youll be spending a lot of time with it
It wont always be easy/fun to continue

Research Process
Background Investigation
Given an idea, need to determine:
Has this work been done previously?
What similar work has been done leading up to
this point?
How is any previous work distinguished from
what Im planning to do?
What group of people will be positively impacted
by the research?

Tools
Literature Review using library resources (e.g.
online databases such as ACM and IEEE, popular
magazines)
WWW search

Research Process
Refinement of Idea
Based on background investigation,
need to refine idea
Issues:
Precision focus on precisely identifying:
Problem
Possible solutions (plural!)

Scope need to build fences


Whats an essential part of this work? (fence in)
Whats tangential, additional, or for any other
reason best left for later/someone else? (fence
out)

Research Process Core Work,


Investigation and Development
Provide yourself with infrastructure
equipment / software
additional knowledge (get up to speed)

Do the work
Experimentation (scientific process)
Develop opinions
Look for better ways of solving problem
Can you generalize?
Can you develop a framework?

Discuss, brainstorm
Reevaluate as you proceed
Look for improvements, changes to your original ideas

Research Process Core Work,


Investigation and Development
(2)

Process

Work regularly
Easier to keep going if have a commitment
to a regular work time
Helps you keep your past work in mind

Allocate large block of time for research


Takes time to get going/back to speed
Make sure can do something significant each
work session

Research Process
Core Work, Documentation
Need to document as you go
Dont want to lose any information

1) Maintain a diary for day-to-day thoughts


Can be paper, electronic, ...
Keep it with you at all times
Never know when good ideas will hit

2) Keep an updated task list


Focus on accomplishing something each work session

3) Write up your work


Periodically, write a few pages on a subset of your work
Summarize work, accomplishments, problems

At end, write up a summary document


Can be based on steps discussed here

Research Process
Core Work, Prototype
Need to demonstrate the merit of
your ideas
If work is non-theoretical, do this
through a developed system
No need to build the entire system
Just need to demonstrate the value of
the core ideas

Research Process Evaluation


Perhaps the most difficult part.
Best if can show others are already using your work

Quantitative
Test your prototype
What improvements exist over currently available
alternative?
How much of an improvement do you see?

Qualitative
What can you do now that couldnt be done before?
What are the benefits of your solution?

Research Process
Identification of Future Work
Helps you organize any future efforts
Helps others build on your work
Sources:
What you excluded in your idea
refinement
New problems that have surfaced during
your work

Research Process Presentation


Its not a contribution to the field if
no one knows about it or can use it
Presentation/Dissemination
Conferences, Journals, Web
e.g. National Undergraduate Research
conference

Papers, Talks, Poster Sessions

Example
my Masters project
an example for each of you
choose a CS subject youre interested in
think of a problem or issue you see in that area
refine your interest to a possible project that
involves one or more ways of solving that problem
outline the steps youd take to do the project work
and test your ideas
what is your hypothetical conclusion?
how would you evaluate the quality of your
solution?

Resources
http://www.cs.indiana.edu/mit.resear
ch.how.to/mit.research.how.to.html
http://www-2.cs.cmu.edu/afs/cs.cmu
.edu/user/mleone/web/how-to.html
http://www.honors.ucr.edu/research.
htm
http://www.utexas.edu/research/eure
ka/resources/why/index.php