stats3 - Monday, August 30...

Info iconThis preview shows pages 1–11. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Monday, August 30 http://emwilk.myweb.uga.edu/POLS4150.html Course Webpage: POLS4150 RESEARCH METHODS IN POLITICAL SCIENCE (90595) Units of Analysis / Levels of Analysis. Types of Variables / Levels of Measurement. Units of Analysis Units of Analysis: who / what are our observations? Level of Analysis: individual (micro) or aggregate (macro)? may depend on Many hypotheses (H) can be tested with different units depending upon the level of analysis: Ex: General H: Education increases turnout Levels of Analysis individual (micro) level Behavior of the individual is observed Unit of Analysis John 14 Yes Jack 22 Yes Kate 16 No Sawyer 11 No Juliet 20 Yes Turnout General H: Education increases turnout Specific H: Likelihood to turnout increases as yrs of education increases Education (yrs) Levels of Analysis Units of Analysis: Individuals Independent variable: yrs of education Dependent variable: Voter turnout Levels of Analysis aggregated (macro) level Individual level behavior is aggregated & studied at the macro level Unit of Analysis GA 48% 61% NY 58% 58% VA 63% 68% MN 60% 78% AZ 43% 56% Turnout % General H: Education increases turnout Specific H: States higher PCT of college grads have higher turnout % College grads Levels of Analysis Units of Analysis: states Independent variable - % college graduates Dependent variable: Voter turnout Units of Analysis / observation Ecological fallacy drawing conclusions about individual behavior based on aggregate data Hypothesis: strength of partisanship increases the likelihood to vote. Observable implication: counties with a higher % of strong partisans should have higher levels of turnout. This approach can result in an ecological fallacy Units of Analysis / observation County 1 Strong partisans Others # in C 1 20 80 # voting 20 40 % turnout 100% 50% N = 100 60% County 2 Strong partisans Others # in C 2 40 60 # voting 30 20 % turnout 75% 33% N = 100 50% Units of Analysis Handout 1. As years of formal education increases, so does income Dependent Variable: Independent Variable: Unit of Analysis: Income Education Depends on level of measurement Individual individuals Aggregate states, counties, countries, ect. Units of Analysis Handout 2. An incumbents expected vote share decreases as challenger spending increases. Dependent Variable: Independent Variable: Unit of Analysis: Incumbents vote share Challenger spending Incumbents running for office Units of Analysis Handout 3. Periods of divided government lead to legislative gridlock. Dependent Variable: Independent Variable: Unit of Analysis: Gridlock (how would you measure this?) Partisan control of government Time (years, sessions of Congress) Units of Analysis Handout 4. International disputes over territory tend to be more severe than other disputes Dependent Variable: Independent Variable: Unit of Analysis: Severity of dispute Reason for dispute International disputes (India vs. Pakistan; Somalia vs. Ethiopia) Units of Analysis Handout...
View Full Document

Page1 / 50

stats3 - Monday, August 30...

This preview shows document pages 1 - 11. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online