CCJ 355 Lecture Notes - Lecture 6: Ex Post Facto Law, Ceteris Paribus, Random Assignment
Document Summary
True experiments: components of true experiments, two comparison groups (more or less equal, through randomization, random assignment to the two (or more) comparison groups, assessment of change in the dependent variable for both groups (posttest) Nonequivalent control group design: two selection methods used for a comparison group: Interaction of testing and treatment: solomon four-group design subjects randomly assigned to at least two experimental groups and at least two comparison groups, one experimental group and one comparison group have pretest, other two will not. Generalizability and time in research: sample generalizability, ability to apply findings to clearly defined, larger population, cross-sectional vs. longitudinal research designs, cross-sectional research design, data collected at one point in time. Longitudinal research designs: data collected at two or more points in time. Other issues: fairly easy to show empirical association in nonexperimental designs; time order & Nonspuriousness more difficult to show: use of statistical control to reduce risk of nonspuriousness.