CRIM 320 Lecture Notes - Lecture 5: Chi-Squared Distribution, Level Of Measurement, Dependent And Independent Variables
9 Apr 2020
School
Department
Course
Professor
Document Summary
Analyzing nominal data using a proportional reduction in error (pre) technique. Like chi-square, is measure of association for nominal data. Instead of association, lambda treats 1 variable as dependent and the other as independent variable. Figure out to the extent of one variable can help us better understand the other variable, specifying one has dependent and the other as independent. Dependent = that which we are trying to explain. Independent = that which we are trying to explain it. Association = simply that, no cause/effect relationship that is hypothesized at all, simply go together. Regression = attended to use variables to predict other variables. In essence, what we are going to use in lambda. Unlike chi-square (and its derivates), it is a pre measure. Point of doing lambda is to determine whether your proportional reduction of error is significant. More useful techniques cuz it tells us more things but harder to use ly significant.
