Research Design
The Basics of Social Research
by Earl Babbie
Presented by Karl F. MacDorman
Ch t Outline
Chapter
O tli
z
z
z
z
z
z
z
Three Purposes of Research
The Logic of Nomothetic Explanation
Necessary and Sufficient Causes
Units of Analysis
The Time Dimension
How to Design
g a Research Project
j
The Research Proposal
I t d ti
Introduction
According to Babbie what is scientific inquiry?
z
z
Observation
Interpretation
“The process by which scientists ask questions,
develop and carry out investigations
investigations, make
predictions, gather evidence, and propose
explanations.”
p
- NIH
Th
Three
Purposes
P
off Research
R
h
According to Babbie
Babbie, what are the three
purposes of research?
Th
Three
Purposes
P
off Research
R
h
1
1.
2.
3.
Exploration
Description
Explanation
Th
Three
Purposes
P
off Research
R
h
Two points Babbie does not mention
mention, but
which are purposes of research are
z
z
Prediction and
Control
Purpose of Exploratory
St di
Studies
What is the purpose of an exploratory study?
Purpose of Exploratory
St di
Studies
z
z
z
To satisfy researchers
researchers’ curiosity and desire
for better understanding.
T test
To
t t the
th feasibility
f
ibilit off undertaking
d t ki a more
extensive study.
T develop
To
d
l methods
th d to
t b
be employed
l
d iin a
subsequent study.
What is the main drawback of an exploratory
study?
Purpose of Exploratory
St di
Studies
Exploratory studies do not provide satisfactory
answers to research questions, because the
participants in the study may not be
representative of the larger population you
are trying to generalize about
about.
Criteria for Nomothetic
C
Causality
lit
1
1.
2.
3.
Correlation. A statistical correlation
Correlation
between at least two variables.
Ti
Time
order.
d
Th cause ttakes
The
k place
l
b
before
f
the effect.
N
Nonspurious.
i
Th
There
is
i no third
thi d variable
i bl
that can explain away the observed
correlation as spurious
spurious.
What is an example that does not meet
each criteria?
Criteria for Nomothetic
C
Causality
lit
Examples of unmet criteria:
z Uncorrelated: If equal percentages of
Democrats and Republicans supported an
issue, party affiliation could not be identified
as a cause for supporting the issue.
z Reversed time order: A child’s religion is not
generally the cause of the parents’ religion.
z Spurious: Ice cream sales do not cause
death by drowning, although they may be
correlated.
correlated
False Criteria for Nomothetic
C
Causality
lit
z
z
z
Research can determine some causes
causes, but it
cannot determine complete causation.
Exceptions do not disprove a causal
relationship.
Causal relationships can be true even if they
don’t apply in a majority of cases.
What are necessary causes?
What are sufficient causes?
Necessary and Sufficient
Causes
z
z
Necessary cause - a condition that must be
present for the effect to follow.
S ffi i t cause - condition
Sufficient
diti th
that,
t if present,
t
guarantees the effect will follow.
Necessary and Sufficient
C
Causes
Necessary Cause
Sufficient Cause
Male
Took Exam
Female
F
F
F
PREGNANT
A
Did Not Take
F
FAILED
C B
C B A
B A A
F
U it off A
Units
Analysis
l i
What are units of analysis?
Things we examine to create summary
d
descriptions
i ti
off allll such
h units
it and
d tto explain
l i
differences among them.
What are some examples?
U it off A
Units
Analysis
l i
What or whom to study:
z Individuals
z Groups
z Organizations
z Social artifacts
What is a social artifact?
What are some examples of them?
S i l Artifact
Social
A tif t
Any product of a social beings or his or her
behavior is a social artifact.
z A social artifact implies a set of all objects of
the same kind: all children’s books, all
biographies, all press conferences.
z Social interactions such as weddings,
friendship choices, court cases, traffic
accidents,
id t and
d so on, are also
l social
i l artifacts.
tif t
What is the ecological fallacy?
Units of Analysis and Faulty
R
Reasoning
i
Ecological fallacy – assuming something
learned about an ecological unit says
something about the individuals in that unit
unit.
What’s an example it?
z To
T assume younger voters
t
tend
t d to
t vote
t for
f
female candidates, because precincts with
younger voters gave female candidates a
greater proportion of the vote.
What is misguided reductionism?
z
Units of Analysis and Faulty
R
Reasoning
i
z
Misguided Reductionism – Reducing
something to a simple explanation when in
reality it is complex
complex.
Wh t are cross-sectional
What
ti
l studies?
t di ?
Th Time
The
Ti
Dimension
Di
i
z
Cross-sectional
Cross
sectional Studies
Observations of a sample or cross-section of
a population or phenomena that are made at
one point in time. (1980 U.S. Census)
What are longitudinal studies?
Th Time
The
Ti
Dimension
Di
i
Longitudinal Studies
Observations of the same phenomenon over
an extended period
period.
What would be an example?
z Field-research
Fi ld
h projects
j t
z
What are trend studies?
L
Longitudinal
it di l Studies
St di
Trend Studies
A type of longitudinal study that examines
g within a p
population
p
over time
change
For example?
z Comparison of U.S. Census data over a
period of decades (1980, 1990, 2000)
z
What are cohort studies?
panel studies?
What are p
L
Longitudinal
it di l Studies
St di
z
Cohort Study
Examines specific subpopulations, or
cohorts as they change over time
cohorts,
time.
z
Panel Study
Examines the same set of people each time
(e g interviewing the same sample of voters
(e.g.,
every month during an election campaign).
Th Research
The
R
h Process
P
Interest
Idea
Conceptualization
Research
M th d
Method
Operationalization
p
Observations
Theory
Data Processing
Analysis
Application,
Dissemination
Population, Sampling
How to Design a Research
P j t
Project
What are the steps involved in designing a
research project?
How to Design a Research
P j t
Project
1
1.
2.
3.
4.
5.
6.
7.
8.
9.
Define the purpose of your project
project.
Specify exact meanings for the concepts you
want to study.
y
Choose a research method.
Decide how to measure the results.
Decide whom or what to study.
Collect empirical data.
Process the data.
Analyze the data.
Report your findings.
R
Research
h Design
D i
iin Review
R i
What is triangulation?
z
Triangulation – The use of several different
research methods to test the same finding.
What are the elements of a research proposal?
Elements of a Research
P
Proposal
l
1
1.
2.
3.
4.
5.
6.
7.
8
8.
Problem or objective
Literature review
Participants for study
Measurement
Data-collection methods
Analysis
y
Schedule
Budget