URBS 2005

Research Methods in Urban Studies

Lecture 5: 09/10/2024

Research Design
PREVIOUS LECTURE

lHistory and definition of

case studies
lTypologies and Values of
case studies
lStrengths & Weaknesses
TODAY’S LECTURE

lResearch Design
lMeasurement
lResearch Proposal
 Research design
A research design is a plan, structure and strategy of
investigation so conceived as to obtain answers to
research questions or problems.
• Conceptualize an operational plan to undertake the
various procedures and tasks required to complete
your study.
• Ensure that these procedures are adequate to obtain
valid, objective and accurate answers to the research
questions.
• Strategies of inquiry (Denzin & Lincoln, 2011)
 Research design
Key aspects of research design:

• Research methodology

• Participant/sample collection

• Data collection procedures and Instruments

Examples of Research Designs
 Good research design

• Logical structure for the enquiry

• Clear plan for data collection and analysis

• Convincing conclusions
Types of research design

• Experimental design
• Cross-sectional design
• Longitudinal design
• Case-study design
 Experimental design

Useful if your objective is to find out the effect of a

treatment/intervention.

E.g. Effect of a drug/chemical/policy

Human/animal behavior
 Classic experimental design
Experimental group: the group which is exposed
to the treatment.
Control group: the group which is NOT exposed
to the treatment.

Non-blind experiments, single-blind

experiments, and double-blind experiments.
• Placebo effect
 Classic experimental design
Experimental group: the group which is exposed
to the treatment.
Control group: the group which is NOT exposed
to the treatment.

Non-blind experiments, single-blind

experiments, and double-blind experiments.
• Placebo effect
Random
Random assignment: randomly assign people to
different groups at the beginning of a study.

To make sure that any differences between the groups

should be random rather than systematic.

The variation between the two groups is confined to

the intervention. Then it is likely that the effect of the
intervention on the experimental group has been
caused by the intervention.
Limitations of experimental design
• Difficult to carry out experiments on human society.
• Difficult to exclude the impact of external factors
(difficult to establish causality)
• Age, gender, occupation, ethnicity, etc.
• Ethical issues (e.g. deception, protection of
participants, right to withdrawal, etc.)
Quasi-experiment: similar to experiment, but lack the
element of random assignment to treatment or control.
Cross-sectional design (Observational
Design)

Useful if your objective is to

find out existing differences
between groups of people.
Data collected at one point
in time.
e.g. census
 Limitations of cross-sectional design
• Inclusion of control variables

• Lack of a time dimension

• Difficult to establish causal direction (X causes Y, or Y

causes X)

• Low response rate à selection bias

• Difficult to study uncommon or rare illnesses efficiently

Example
Example (cont.)
source
 Longitudinal design
Useful if your objective is to find out the effect of a
treatment/intervention over time.
Usually involves collecting data from the same cohort
of people at multiple points in time.
A few weeks… or as long as several decades
E.g.
https://www.instagram.com/reel/CxVl8PussU9/?igshid
=MzRlODBiNWFlZA%3D%3D
 Limitations of longitudinal design

• Time consuming and costly

• Difficult to track same set of people after many years

(missing data)

• Lack of control: any change over time could be due to

multiple factors
 Case study design
Useful if your objective is to find out how and why something
occurs in a comprehensive way (A case could be an individual, a
group, a community, an event, a subgroup of a population, or a
city. )

The case you select becomes the basis of a thorough, holistic and
in-depth exploration of the aspect(s) that you want to find out
about.

E.g. Why are crime rates higher in low-income communities?

Why do IT companies have longer working hours?
Establishing the context

Historical Social

Phenomenon
Economic
Legal

Political
 Limitations of case study design

• Difficult to justify case selection

• Difficult to generalize to other contexts (e.g. other

neighborhoods, other companies, other districts,

other cities)
 Research design data
collection
• Research design is NOT the same as method of data

collection

• Data for any design can be collected with any data

collection method

• Mixed methods
Things to note
Conceptualisation: specify the meanings of the concepts
and variables to be studied.
Operationalisation: how will we actually measure the
variables under study?
Reliability: are the results repeatable?
Replication: can others replicate the results?
Validity: are the results a true reflection of the world?
Internal (the underlying phenomenon is
measured)/external (generalize to the population)
Generalizability and transferability
Purpose statement sample (Quant)
Purpose statement sample (mixed)
Measurement
Why measure?
• Some concepts are easy to see (e.g.
gender)
• Many concepts cannot be directly
observed (e.g. political beliefs)
• Concepts in the social world are often
abstract and vague
Examples

• “Sustainability”
• “Satisfaction”
• “Happiness”
Measurement and concepts
Measurement is the link between data and
concepts.

• Concept -> How can you observe it

empirically (in the real world)?

• Measurement -> data

 Measurement process

•Conceptualization
•Operationalization

source
Conceptualization
The process of taking a vague idea and
giving it a conceptual/theoretical
definition.

Vague -> systematically organized ideas

How?
1) Personal experience (e.g. observing
directly)
2) Discussions with other people
3) Reading which others have said
(literature review)

E.g. online game addiction

Strategies for defining a concept

•Types/dimensions of an idea
•Possible units of analysis
•Differentiate from similar ideas
Operationalization
Linking a conceptual definition to a
concrete set of measurement techniques.

Measurement techniques include survey

questions, interview questions, and a
system for observing events/analyzing
documents.
Four levels of measurement
Nominal: categories (e.g. “employer”, “city”,
“female”)
Ordinal: ranked categories (e.g. “low-
income”, “high temperature”, “XXL”)
Interval: distance between categories (e.g.
“28 °C”, pH, IQ)
Ratio (e.g. “180cm”, “15000 HKD”, “500
people”)
Nominal measures
Example: Gender
 €
Female Male

Ordinal measures
Example：Is religion important to you?

€   €
not important slightly important important very important

Low High
Interval measures
Example: IQ

€   € 
80 90 100 110 120

Ratio measures
Example: Income

€   € €
$0 $10,000 $ 20,000 $ 30,000 $ 40,000 $ 50,000
Special measures: Scale & Index
Scale: measuring intensity/direction of a
variable
E.g. from “strongly dislike” to “strongly like”

Index: combination of multiple indicators

E.g. Consumer Price Index (CPI)
Research Proposal
 Learning objectives

• To appreciate the nature and the

importance of the research proposal
• To know how to put together the
contents of the research proposal
• To understand the important issues
involved in the research
 The research proposal

The research proposal is the key “gateway”

document to your research project. It sets out
the main details of how your research will be
conducted.

It is also a key document in formulating and

crystallizing your topic ideas into a “do-able”
research project.
 Purpose

• Organizing and developing your topic ideas

• Testing the scope of the research
• Identifying an appropriate supervisor
• Convincing others of the merit of your idea
• Initiating the research process
• Obtaining support and early access rights
• Being a basis on which to develop your research
 Contents
1 Title
2 Research background
3 Research goal / objectives
4 Research questions
5 Literature review
6 Methodology
7 Ethical issues
8 Time table
9 Reference list
 Research goal
A common way to express the research problem.
The goal (or aim) of this research is to …
E.g. “to understand the impact on quality of life by
the West Island Line.”

Note: The goal can be “tightened” by specifying:

• the guiding theory for the research
• the type of study (quantitative or qualitative)
• the method (e.g. a comparative study, a case
study, etc.)
 Research questions
A more specific way to state the research problem.
The normal arrangement is to list a number of questions that the
research will answer (the impact on quality of life by the West
Island Line):
RQ1: What aspects of life are most affected?
RQ2: Who are most affected?
RQ3: Why certain people are more satisfied than others?
RQ4: ???

Using this method expresses your research problem in a much

“tighter” and “shaper” manner from the start, and is suitable
where you have a clear set of issues arising from the literature
that addresses the research problem.
Literature review
• focus on relevant scholarly efforts and may cover
relevant:

International Empirical
theories concepts
cases analyses

• Attention may also be paid to...

Empirical vs
Theoretical Research
theoretical
debates gaps
studies
Literature review
• Critically appraising what other people have written
on a topic

• Included in academic books and journal articles

• Indicates what questions people have asked and

what theories people have used. Do you
support/disagree with these theories/explanations?
Literature review
• A critical review of the literature
• Not just a summary
• Link ideas by pointing out development of theories, the
direction of the empirical research, debates and
contradictions.

• Highlight gaps
• Discuss the strengths and weaknesses
• Different literature are compared and contrasted
Why do one?
To…

…demonstrate …identify what is

…find out what is
awareness of past NOT known
currently known
research
Literature review
• Identify clear research topic

• Homelessness

• Homeless teenagers in Hong Kong

• The link between unemployment and homeless

women in Hong Kong
How to write a literature review?
• Resource Step by step with examples:

• https://www.sjsu.edu/writingcenter/docs/handout
s/Literature%20Reviews.pdf
Theoretical review
Which concepts are crucial to our understanding of
this topic/problem?

Clear definition and explanation of the meaning and

dimensions of all concepts which you use
E.g. unemployment; migrants; health; happiness
Developing a theoretical
framework
• What theories have other people developed to
explain a phenomenon/problem?

• What are the debates/competing theories in the

literature?
 Empirical review
Which place(s) are you focusing on?
E.g. Public housing in Hong Kong; urbanization in Africa;
Earthquake response strategies in multiple cities

How are you narrowing this down?

What empirical studies have been carried out in this

place?

Do not assume that everyone understands the context

in your study area.
 Narrowing down your search
You should make your review manageable and relevant.
Some ways you may do this by:

Time
Geography Language
(since 2015 etc.)

Subject
Subject area
materials By research
(only housing
(only books and approach
studies)
articles)
To sum up, a good literature
review…
• Identifies the importance/significance of a topic

• Reviews the key concepts and theories relevant to

the topic

• Identifies a gap in the literature – important

questions which have not been asked/answered

• Is NOT a descriptive summary

• Is NOT a list of authors and their ideas

Your research project
• A strong theoretical framework is a key criterion by
which you will be assessed

• You cannot simply write down your personal

opinions

• Your work must be connected to an existing body

of research
 Ethical practice
Have I given full consideration to the way I behave as
a researcher with respect to the other human beings
who are involved in my research?
(deception; protection of participants; right to
Withdrawal)

Academic integrity (the moral code or ethical policy

of academia): avoidance of cheating or plagiarism;
maintenance of academic standards; honesty and
rigor in research and academic publishing.