STA130H1 Lecture Notes - Lecture 10: Confidence Interval, Randomized Experiment, Marginal Distribution
STA130H1 verified notes
10/13View all
Document Summary
This is called an interaction term: including the term bty_avg*gender on the right-side of the model specification in lm includes the interaction term plus both of the variables in the model. Vocabulary: a marginal distribution is the distribution of only one of the variables in a contingency table, a conditional distribution is the distribution of a variable within a fixed value of a second variable. Examples of situations that may have confounding variables: a 2012 study showed that heavy use of marijuana in adolescence can negatively affect iq. Is it possible that there are other variables, such as socioeconomic status, that is associated with both marijuana use and iq: another 2012 study showed that coffee drinking was inversely related to mortality. How can confounding be avoided: data can be collected through experiments or observational studies. In observational studies, data are collected without intervention. The data are measurements of existing characteristics of the individuals being measured.
