CRIM 320 Lecture Notes - Lecture 10: Null Hypothesis, Scatter Plot, Partial Correlation
Document Summary
Scientists primarily interested in explaining phenomenon using a variety of theoretically relevant variables in order to do so. Primary difference from correlation is just looking at associations, specifying a dependent and independent variable outcome (dependent variable) One as an outcome, one to predict/explain outcome. Causation: determine what variables are able to account for or predict the variability in. Cannot create true experimental conditions (unethical to take people out and make thing commit crimes) Best explain outcome using theoretically relevant variables while trying to control for the effects of external variables. What are you interested in knowing about: gun homicide rate (dependent variable) What variable would help explain the outcome: gun ownership (independent) Regression line can be used to make predictions about an outcome using independent variables. Look at (cid:449)a(cid:455)s of dete(cid:396)(cid:373)i(cid:374)i(cid:374)g (cid:449)hethe(cid:396) the p(cid:396)edi(cid:272)tio(cid:374)s a(cid:396)e (cid:862)good(cid:863) or not. Comparing effect of independent variable vs. just guessing the mean of the dependent variable.