Political Science 3000
Political Analysis
Final Exam Review
The material covered in this last half of the semester can be broken down into
sections based on 1) the purpose of inquiry and 2) the components for
establishing a causal relationship.
Purpose of Inquiry:
Description
Descriptive statistics
frequency distributions
measures of central tendency
measures of dispersion
Purpose of Inquiry:
Explanation
Establishing a causal relationship: 3 components
Time order: cause precedes effect
based on theory and observation
Correlation: cause and effect have a tendency to occur/not occur together
based on tabular analysis or regression
remember the four key questions to this particular aspect of establishing
correlation
Is there a relationship?
Changes in column percentages across categories of independent variable.
Is the relationship strong?
magnitude of percentage changes.
magnitude of measures of association. (know which ones to use when).
What is the direction of the relationship? (ordinal or interval/ratio only)
distibution of cases within the table
sign of measure of association
Is the relationship statistically significant?
to what extent is this relationship due to sampling error?
probability values: how are they interpreted?
You should also be familiar with Type I and Type II errors.
Accounting for other possible explanations
Elaboration-the introduction of a control variable
Two questions:
What happens to the relationship when the control variable is introduced?
Is the control variable antecendent or intervening?
Regression: Analysis of Interval/ratio level variables
Allows one to simultaneously test multiple hypotheses (i.e. do steps B and C
above at the same time.)
Know the interpretation of:
R
2
b and BETA
t and its associated probability
NOTE: In each instance it will be important for you to not just understand the
technical terminology of these analyses but also the
plain english interpretations
of them.