PNB 3RM3 Lecture Notes - Lecture 24: Random Assignment, Observational Error
Document Summary
Establishing a baseline (bl) against which to compare the experimental condition. If performance deviates from baseline, experimenter can say a certain condition caused that deviation in performance. You don"t always need one - don"t need it if your question doesn"t require you to have one (i. e. Advantages and disadvantages of between and within groups designs. Within group - randomly assigned to participate in all levels of independent variables independent variables. Choosing a between-groups design - randomly assigned to participate in one level of independent variable. Three times to use it - use when. Giving a placebo vs. real thing to someone can"t both be done with one person. When treatment will have changed participant in some way and you can"t go back. Your experiment is too demanding to do within-groups. Must recruit larger n than when using a within-groups design. Can never be as equivalent as in within-groups design.