POLS 3650 Lecture Notes - Lecture 14: Statistical Inference, Point Estimation, Statistical Significance
Document Summary
Lecture 14: comparing a sample to the population. Bivariate analysis explores the relationship between two variables. This entails the investigation of covariation = as the values on one variable change, the values on the other variable change as well. If our variables have an ordinal level of measurement or higher, we can also specify direction: positive: as values on x increase, values on y increase, negative: as values on x increase, the values on y decrease. Additional criteria: theoretical plausibility, cause (independent variable) precedes effect (dependent variable, non-spuriousness. Every bivariate analysis in inferential statistics proceeds through five stages: formulate hypotheses, visualize data, calculate point estimate, conduct test of statistical significance, draw conclusions. When we compare two groups, we explore the relationship between a dichotomous iv and a dv. If the dv is nominal or ordinal: compare proportions. If the dv is interval/ratio: compare means.