CRCJ 3001 Lecture Notes - Lecture 12: Random Assignment, Lactose Intolerance, Confounding
Document Summary
Quasi-experiments: natural setting quantitative experiments & other nonexperimental. Quasi-experimental designs: quasi-experiment almost but not quite an experiment, cannot reach same causal conclusions as would in experiments, have some characteristics of experiments, don"t have as much control as full experiments. Include at least two levels of the iv (but cannot always manipulate iv: state causal hypotheses, accept already existing groups (usually cannot assign participants to groups) Non-equivalent control-group design: aka quasi-experimental between-groups, two major problems with non-equivalent groups, the groups may differ on the dv at the start of the study, solution, measure the experimental & control groups on the dv both before. & after the manipulation (i. e. pretest & post-test measurement: difference btwn the pre-manipulation & post-manipulation scores can then be calculated (difference score, measure of how much each group changed as a result of the manipulation. 1: there may be other diffs btwn the groups that controlled by random assignment, solution, try to rule out each potential confounding variable.