This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Limits of Statistical Methods-treatment is not usually randomly assigned-treatment and control groups could systematically differ in ways related to outcome of interest. -this limits our ability to say the treatment is the cause of the effect Endogeneity: reverse causality; a common issue (Y causes X instead of X causing Y) Spurious effect: third variable is responsible for the association between others.-random sampling is not always possible-collection of reliable data is not always possible-may only be a few cases of phenomenon of interest (ex: revolution)...
View Full Document
This note was uploaded on 02/09/2012 for the course SOC 110 taught by Professor Moskos during the Spring '08 term at Northwestern.
- Spring '08