CRIM 320 Lecture Notes - Lecture 10: Explained Variation, Total Variation, Observational Error
Document Summary
Whether a relationship exists between 2 variables. The direction and strength of that relationship. Crim320 lecture 10: bivariate statistics at the interval/ratio level. When we examine associations between variables, we are interested in how the categories of one variable relates to categories of another. With interval/ratio data, we seek to determine: We can also determine the approximate rate of change in the dv across values of the iv. With interval-ratio variables, association is often called correlation . The primary measure of association for interval-ratio variables is the correlation coefficient, or pearson"s r. Pearson"s r measures the amount of change in y (dv) produced by a unit change in x (iv), where units are expressed in sds. With interval/ratio data, it is possible to create a scatterplot. Dot is placed at the intersection of the obs"s score on x and y.