Write a one page essay defending a causal political theory. Your theory should be falsifiable (it should be able to be wrong) and there should be a way to test that theory empirically. You should defend this causal theory against potential counterarguments.
Note that there are examples of
political theories in the reading. Also note to stay away from normative issues
about whether people are bad or good - or whether it is a good thing or a bad
thing that the causal relationship exists.
For example, if the theory is that people vote in their best interest,
you can say that wealthier people are more likely to vote Republican. But be careful not to judge people for doing
so. Stick to defending and arguing the
causal reasoning in your theory.
An example of a political
theory is here:
(Note that the simple causal
statement is the first sentence of the paragraph).
Political alienation from one's government causes people to
participate politically in the context of post-Soviet
You should provide a one page interpretation of Rousseau’s First
Discourse. Students must read Rousseau’s
discourse and make a one-two page argument about what he means when he says
what he says. This is not a critique;
this is not your own normative analysis.
Instead, students should be making an argument (considering potential
arguments) about what Rousseau is arguing.
Interpretation is difficult work and requires deeply thinking about the
meaning of the text. What does the
author really mean when he says “…”
It is not a critique or a response to what is being said. You won’t get credit for a critique or an
obvious summary. Interpretation is
thinking about the meaning of language – and what alternative interpretations might
be. Think of all the people who disagree
about what the text of the bible means.
What does it mean to say “thou shalt not kill?” Is it just murder or all
killing? Does it include only human beings?
We rely on other sections of the bible to enlighten us about the meaning
of certain passages. This is what you
have to do with the Rousseau assignment: read the Excerpt of the First
Discourse and make a single page summary of what he is trying to argue, paying
close attention to alternative interpretations and why they are weaker than
your interpretation. The strength of
your assignment will have to do with how strong your alternative
interpretations are. In other words,
even though there is a strong alternative interpretation, your interpretation
is nevertheless more convincing. So,
this means you have to identify at least two logically distinct interpretations
of the First Discourse – and then argue which one is most convincing.
Invent a concept and design an operationalization of that concept. Concepts must be unidimensional and you must design a decision rule for how to assign numbers to your units that vary along this dimension. Students should be explicit about what the high end and low end of this concept. The name of the concept should indicate the high and low end. Students should take into consideration that the concept selected may have more than one dimension. In this case, students will have to break it down into their component parts.
Pitfalls:
Things that are not concepts: political economy, voting interests, rational choice, socialization, culture, post-Soviet Russia, institutions, psychology, costs and benefits.
Things that are concepts: wealth, voting (whether people vote or not), party of vote decision (which party a person votes for), mother’s party identification (instead of socialization), perceived acceptability of violence (instead of culture), rule of having graded homework everyday (instead of institution), self esteem (instead of psychology).
Operationalizing a concept would be:
My concept
is acceptability of ethnic violence. I
will ask the following question in a survey.
I will now read some specific
strategies that some people recommend to resolve problems in the
Then, I
will assign each person to a number based on their answer. 4 is high acceptability of ethnic
violence. 1 is completely disapproving
of violence, the lowest acceptability.
fully approve = 4, somewhat approve = 3,
somewhat disapprove = 2, or completely disapprove = 1
Use Stata to open data and run a frequency analysis – and a central tendency analysis of a single variable (mean, mode, median). Include your SPSS syntax and output as well as your interpretation of the variation and central tendency of the variable.
http://socsci.colorado.edu/~bairdv/NES.dta
Come up with an original political example (that we did not talk about in class or in the reading) of two variables that are correlated, but spuriously correlated. Students should identify a common cause of those two variables and make the causal arguments in both cases.
Here is an example of a spurious correlation that is sociological:
Playing
violent video games is correlated with an increased tendency toward violence. These two variables are spuriously correlated
and have the common cause of lack of parental supervision. The two causal arguments are 1) lack of
parental supervision causes children to play more violent video games and 2)
lack of parental supervision causes children to be more violent. When controlling for parental supervision
(high supervision and low supervision) the correlation between playing violent
video games and being violent would decrease or disappear.
Conduct a crosstab and correlation analysis for two variables that are causally related. Make sure you include a causal argument, paying close attention to criticisms of this causal theory. Interpret the correlation between the two variables. How much confidence do we have in the estimation of this correlation? Have you rejected the null hypothesis?
Conduct a bivariate regression analysis for two variables that are causally related. Make sure you include a causal argument, paying close attention to criticisms of this causal theory. How much confidence do we have in the estimation of this bivariate relationship? Have you rejected the null hypothesis? Interpret the magnitude of the linear impact (from the bivariate regression) as well as the correlation between the two variables (from the crosstab and correlation analysis).
You measure the magnitude of the relationship as x goes from its lowest value to its highest value with the slope coefficient or the B (NOT THE BETA).
Then, you have to see what the range is of x – your independent variable (NOT Y – NOT the dependent variable). If X ranges from 1-4, then the range is 4-1, which is 3. If X ranges from 1-10, then the range is 10-1 which is 9. Multiply the B (NOT THE BETA) by the range of x and you have how much y changes as x goes from its lowest value to its highest value.
Identify two variables that you think are correlated, but spuriously. Develop the theory that argues that they are correlated but have a common cause. Identify the common cause. Use crosstab analysis to show that the relationship is spurious, controlling for the common cause, and then showing that the relationship is mitigated. Interpret the crosstab and correlation results.
Example:
The correlation
I am going to examine is the relationship between Liking Obama and the approval
rating of Bush. It could be argued that the respondents liking of Obama causes
his approval of Bush. I claim that both
are dependent on the respondent’s ideology.
Identify two variables that you think are correlated, but there is an intervening effect. Develop the theory that argues that they are correlated but have an intervening effect. Identify the intervening variable. Use crosstab analysis to show that the relationship is intervening, controlling for the intervening variable, and then showing that the relationship is mitigated. Interpret the crosstab and correlation results.
Example:
Support for
the military causes support for
You should come to class with a research question: why does y vary? You should have a list of concepts that answer the question why does y vary. And you should have a list of at least 25 sources of previous work that has looked at the same or similar questions.
Conduct a multiple regression analysis with at least two independent variables. Provide a defense of the theory and causal inferences used in your analysis. Conduct a correlation analysis of all the variables in the model and compare the Pearson r’s to the beta’s. Are your relationships intervening, spurious or neither (just two alternative causes).
Provide the outline of your final research literature review. This outline should include the answers to your: Why does y vary? research question. Note there is an example on the syllabus. You should include one paragraph arguing for the causal relationships between one of your independent variables and one of your dependent variables.
Design an experimental research design.
You must have a theory and you must take special care to identify a
treatment. Discuss potential causal
inferences and implications of your study. What are the threats to internal and
external validity?
The final exam will be a take home exam in which they receive an article and they have to engage in a 3 page detailed critique of all the aspects of research design in the article. Your job is to evaluate the causal theory, measurement, statistical analysis, the selection bias, omitted variable bias, inferences and implications of the study.
You should find a research topic dealing with political science. Everyone’s research question is the same: What causes ______ to vary/change (across people, geographic units, time)? You should review the academic literature associated with previously found explanation for what makes your ____ vary/change (across people, geographic units, time)? From that literature, you should deduce your own theory from an idea that comes from a lacuna (something missing) in that literature. You should develop a research design that would answer the question and explain how you intend to go about doing this research and the implications of the findings that you expect.
I. Research Question – First Sentence or Two. What causes ______?
II. Why the Question is Important – 2 pages.
III. Literature Review – 6-8 pages. This is the bulk of the paper. The outline of this section is a list of causes of the political phenomenon you are interested in. You should highlight and resolve to the best of your ability any contradictions to the literature. Think of this part of the paper that is making the argument that YOUR study needs to be done. Each section will be an argument about the manner and extent to which that particular concept causes your political phenomenon of interest. Be sure to consider both sides of the argument. You will need to deal with all relevant counterarguments.
Here is an example of an outline of the literature review:
What causes the tendency to litigate to vary across people?
A. Political disadvantage
B. Perception that courts are fair
C. Social networks
D. Anger
E. Self esteem
F. Severity of the grievance
G. Attribution of blame of government
H. Resources (income, wealth, information, education)
For each bullet point, you will mention what the literature has to say about whether this concept causes a tendency to litigate. You will discuss many different scholars and their findings in terms of how they relate to your conclusions about the manner and extent to which each concept is a cause of your political phenomenon of interest.
IV. Summary of Literature Review: Implications for your Research. One or two paragraphs. The summary of the literature review should introduce your research design. How are you going to contribute to what we already know?
V. Research Design. 2-3 pages.
VI. Discussion of the Implications of Expected Findings: You should answer the question: What difference does it make for our understanding of politics, perhaps in terms of how minorities can be better protected, better representation, better economic growth, or other things that we care about … that we now know what we are able to conclude from your research? 2 pages.
You will be graded on the rigor of your discussion of the literature and on the internal consistency of your arguments. I will pay close attention to whether you have considered all potential counterarguments or inconsistencies presented by the literature in your discussion. You will also be graded on the creativity of the research design as well as your ability to answer the question: what difference does this research make in its contribution to previous literature? The answer to this question enables you to connect the first parts of the paper with the last.