Transcript 5.+The+Merits+of+Methodology_Slides

1
MERITS OF
METHODOLOGY
How to Conduct Systematic Political Research
Lecture Structure
2





Aims
Methods of Comparison
Single Case Study
Using Quantitative Literature
Applying a Theory vs. Discursive Method
3
Methods of Comparison
Methodology
4
Definition:
 A body of methods, rules, and postulates employed
by a discipline: a particular procedure or set of
procedures.
 The analysis of the principles of inquiry in a
particular field.
(Merriam-Webster Dictionary)
Questions addressed in this course
5






How should we study conflict?
What causes conflict?
Who participates in conflict?
How do people behave during conflict?
What are the consequences of conflict?
How do we evaluate conflict?
Comparing Literature
6
What is the key similarity between Mamdani
(2001) and Collier & Hoeffler (2004)?
Comparing Literature
7

Mamdani (2001)
 Rwandan
genocide
 1994
 Qualitative

evidence
Collier and Hoeffler (2004)
 Civil
conflicts
 1960-1999
 Quantitative evidence
Methods of Comparison
8
Three basic approaches to answering a research
question:
 Large-N: > 50 cases.
 Small-N: 2-20 cases.
 Single case study: 1 case.
N = number of cases/observations.
Small-n (aka ‘comparative method’)
9


Cases are intentionally selected in order to establish
‘control.’
Most Similar Systems Design (MSSD).
 Similar
cases.
 Different outcome.

Most Different Systems Design (MDSD).
 Different
cases.
 Same outcome.
Most Similar Systems Design (MSSD)
10
Which cases would you use to investigate the link
between ethnic tensions and civil war, using this
research design?
Most Different Systems Design (MDSD)
11
Which cases would you use to investigate the link
between ethnic tensions and civil war, using this
research design?
12
Single Case Study
What is a Single-Country Study?
13


“any study in which a single country forms the
basic unit of analysis, but which may also be
broken down into smaller units across time and
space...” (Landman, 2000: 86)
N=1
What is a Case Study?
14

Single observation/case:
 Time
 Space
 Level

N=1
of Analysis
Functions of a Case Study
15




Classification.
Contextual description.
Hypothesis generation.
Theory testing:
 Most
likely study.
 Least likely study.


Outliers.
Process tracing.
Limitations of Single Case Studies
16

Limited scope/generalisability.

Selection bias.
Further Reading
17
Landman, Todd. 2007. Issues and Methods in
Comparative Politics: An Introduction. London:
Routledge.
18
Quantitative Literature
Regression Analysis (1/2)
19
4.5
4
3.5
Conflict
3
2.5
2
1.5
1
0.5
0
0
0.5
1
1.5
2
2.5
Natural Resources/GDP
3
3.5
Characteristics of Quantitative Literature
20

Large-N

Empirical data

Statistical techniques
Large-n
21
How many observations are examined in
Collier and Hoeffler (2004)?
Collier & Hoeffler’s dataset by the numbers
22

98 countries

79 civil conflicts

1960-1999 divided into 5-year periods

750 observations
Concepts in this Course
23










Authority and coercion;
Structural violence and deadly force;
Conflict and war;
Positive and negative peace;
Peacemaking;
Peacekeeping;
Peace enforcement;
Peacebuilding;
The military; and
Militarism.
Concept v. Variable
24

Concept = abstract.

Variable = observable/measurable.
How do we measure the following concepts?
25

Conflict

Democracy

Ethnic tension

Inequality
Statistical Techniques
26

Used to look for patterns between variables.

Correlation: relationship between two variables.

Correlation is not the same as causation!
 Example:
ice cream and summer.
Natural Resources and Conflict
27
8
7
Conflict
6
5
4
3
2
1
0
0
1
2
3
4
Natural Resources/GDP
5
6
7
Regression
28


Adds a line of best fit to the observed result.
Regression analysis therefore measures the extent to
which independent variables predict the dependent
variable.
Regression Analysis (1/2)
29
4.5
4
3.5
Conflict
3
2.5
2
1.5
1
0.5
0
0
0.5
1
1.5
2
2.5
Natural Resources/GDP
3
3.5
Regression Analysis (2/2)
30
8
7
Conflict
6
5
4
3
2
1
0
0
1
2
3
4
Natural Resources/GDP
5
6
7
Terminology
31

Significance Levels

Regressor/Predictor/Independent Variable

P-values

Confidence intervals
Limitations
32



Cannot prove causation.
Limited interpretive ability (i.e. explaining why these
relationships exist).
Use of proxies.
Analysing Collier & Hoeffler (2004)
33
Q: Does natural resource prevalence explain the onset of civil conflict?
Steps to follow:
 How did they operationalise and measure these concepts.
 Do these proxies make sense? Is there correlation? In other words –
what do we see happening to the likelihood of conflict variable
when we look at different natural resource/GDP values.
 Ask how much we should read into the correlation. Ask the students
here what might make us more or less likely to find one of the
correlations important or suggestive of a causal relationship.
34
Theory and Your Research
Applying Theory Systematically
35

Logic of theory

Language that a theory demands

Units of analysis

Consistency through the paper

Acknowledging limitations
Referring to Theory as Evidence
36

Not constrained by the logic of the theory

Used to add to understanding, or to hold theory up
to criticism or praise

Can compare explanations offered by different
theories