SOCIOL 2P06 Lecture Notes - Confounding, Randomized Experiment, Regression Analysis
27 Sep 2011
School
Department
Course
Professor
Document Summary
September 16, 2011 can be part of randomized experiment some topics: classroom observation. Usually done in elementary school: useful student interaction teacher-student interaction richness of direct observation self reports are suspect. Manipulating x if we manipulate the cause, the outcome will change: we often can"t manipulate x. Eg how much does education increase wage? cannot manipulate: examine existing relationship so if we see x or y vary together without our having influenced x. But other factors may influence this such as where people are being surveyed from. Problem: confounding variable: education is related to class background. Two categories in this example: upper & lower class: class background is related to wage. Confounding variable: education and class background vary at the same time: maybe class background not education increases wage, we want to hold class background constant , solution: consider upper- and lower-class background as two separate groups.
