MULT30018 Lecture Notes - Lecture 7: Internal Validity, Confounding, Dependent And Independent VariablesPremium
2 pages26 viewsSpring 2018
This preview shows half of the first page. to view the full 2 pages of the document.
Limitations of Regression Analysis: Experiments as a solution
• Independent variables causing dependent variable
o Confounding factors 'third factor'
o Unmeasured 'third factor(s)'
How do experiments work?
• Intervention group vs. control group
• Complete control of environment (very high internal validity)
• Does the lab replicate a 'real world' environment? (lower external validity)
Example of lab experiment: Darley and Latane (1968)
• Why do bystanders often fail to help in an emergency?
• Participants randomly assigned to 1 of 3 conditions
• Percent responding by end of fit across conditions
• High external validity
• Less control of environment, leading to lower internal validity
Do 'get out the vote' efforts work?
• What effects does mobilization have on turnout?
• Field experiments allow researchers to contact some people and not others and then see vote
• = mobilization REALLY works
• Sometimes called quasi experiments because researcher does not manipulate Ivs
o TV violence = increase in crime
o Scholarships improve academic performance
• Experiments help us establish causation
• Many experiments are unethical
• Experimental research has real policy implications
7.1 Examples of experiments
• Behavrioural insights, behavioural economics
o Experiments based on psychology and hedonics
• Pension defaults experiments
You're Reading a Preview
Unlock to view full version
Loved by over 2.2 million students
Over 90% improved by at least one letter grade.