POLSCI 1G06 Lecture Notes - Lecture 8: Dependent And Independent Variables

25 views4 pages
of 4
Political Science 1G06 2014 Lecture 8a Comparative Politics and the
Comparative Method
- Can Political Science be scientific? How?
- If finding objective laws of Politics is the goal of Political Science
(and not every Political Scientist agrees that this is the goal…), how
do we achieve that goal
- The type of experimental method used in the natural sciences simply
cannot be replicated in the social sciences
- We can develop an hypothesis
oA statement that one variable in combination with another will
cause a particular outcome x + y = z
- However, there is a problem with testing this hypothesis:
- We cannot simply isolate and combine variables
- Social variables are always parts of societies containing many other
- How can you be certain that x combined with y to produce z when
variables a-w are also present?
- Any one of the alternative variables, or perhaps some combination of
them could be responsible for the outcome
- So if we are interested in developing a science of politics:
oHow do we find laws?
oHow do we test our hypotheses?
- One possible answer (there are others) is to rely on either critical cases
or the method of Comparative Politics – they allow us to at least
approach scientific adequacy in our analysis of the social world
How to construct a Political Science:
Step 1 – Categorization/classification
- The first step in designing a scientific approach to politics is in
grouping the social world into parts that share similar characteristics
- In principle you can categorize along any line depending on what you
are investigating
- Note, however, that categorization is always a contestable process – it
smoothes out (and therefore hides) differences
- The objectivity and applicability of any political category can be
Step two – hypothesis generation
-Case study
- Find a society (usually a state) in which the phenomenon you are
seeking to understand is present
- Why did the phenomenon develop in that country?
- Note that a single case study is usually not sufficient for proving a
general social law
- You cannot use the same case from which the hypothesis was
generated to test that hypothesis
Step three - testing
- The objective is to control for as many variables as possible so that
you can be reasonably certain that the variables you have identified
are in fact causing the outcome you are interested in explaining
-1. Single Country Study: Crucial cases
-A) Most likely case
- A case where all the conditions the theory or hypothesis calls for are
present, but not the outcome
- This is useful for invalidating a theory, but less useful for confirming
- If the theory states that certain variables will produce a certain
outcome, can you find a case where the variables are present but the
outcome is not?
-B) Least likely case
- Find a case where the theory is least likely to be true
- If the theory holds true even where we would least expect it to, then
the theory might actually be capturing some kind of social law
-2. Multi-country comparative studies:
-A) Method of Difference (Most Similar Systems Study)
- Seek out two countries that are as similar as possible
oSimilar except for the fact that only one of the countries has the
independent variable contained in your hypothesis (the variable
that your hypothesis suggests causes the dependent variable)
- Did the dependent variable (the variable that you are interested in
explaining) occur in the country having the independent variable but
not in the country lacking it?
- If the only meaningful thing that is different between your cases are
the variables highlighted in your hypothesis (independent and
dependent variable)
oThen it is reasonable to infer that the independent variable
causes the dependent variable
-B) Method of Agreement (Most Different Systems Study)
- Seek out countries that are as different as possible
oDifferent except for the fact that every country selected has the
independent variable contained in your hypothesis (the one that
your hypothesis suggests causes the dependent variable)
- Did the dependent variable also occur in all of the countries?
- If the only meaningful thing that the cases share are the variables
highlighted in your hypothesis (independent and dependent)
oThen it is reasonable to infer that the independent variable
causes the dependent variable
-A Brief Guide to Argumentative Research Papers (the paper you
will be writing next semester):
- 1. Establish the research question and the context in which that
question emerged. What is the issue you will be exploring, and why
does it matter?
- 2. Clearly state a thesis (i.e. an answer to your research question)
- This is a basic statement that clearly establishes what your paper will
set out to prove
- This differs from a mere opinion
- A thesis sets out to establish a logical relationship between two or
more variables
- This thesis must be