COMM 88 Lecture Notes - Lecture 13: Selection Bias, Random Assignment, Ice Cream Van
5/17/18
● Factorial designs
●Interaction effect
○The unique effect of the combination of IVs
■The effect of one IV depends on the levels of the other IV(s)
○Example for a music X Task interaction:
■Music reduces learning when studying reading. But enhances it when studying
math
○To test for an interaction effect, graph the cell means
●Music no music
■Math M=60 M=60 /60
■reading M=50 M=70 /60
●55 65
●There is an interaction effect if the lines are NOT parallel
○We do have an interaction here
■“Although music lowered scores overall, the effect was
really only when studying reading. Music had no effect
when studying math.”
●EX:
○Music No Music
●Math M=90 M=70 mean:80
●Reading M=70 M=50 mean: 60
○80 60
■There is a Main effect for task
●And Main effect for music
○But no interaction (bcuz parallel)
●A word about factors (IVs)
○In one design, can have IVs/factors:
■Manipulated variable(s)
●Ex: music exposure, study task
■Subject variable(s)
●Ex: gender, personality trait, TV use (high/low)
○Can only make causal conclusions about manipulated IVs (Not Ss variables)
○If no manipulated variables at all, than it is not an experiment (it’s a survey, with
factorial-type set up)
●Experimental research (continued)
○Key elements to a true experiment
■Manipulation/control
■+Random assignment
■=internal validity
●Posttest only control group design
Document Summary
The unique effect of the combination of ivs. The effect of one iv depends on the levels of the other iv(s) Example for a music x task interaction: To test for an interaction effect, graph the cell means. There is an interaction effect if the lines are not parallel. Although music lowered scores overall, the effect was really only when studying reading. There is a main effect for task. Ex: gender, personality trait, tv use (high/low) Can only make causal conclusions about manipulated ivs (not ss variables) If no manipulated variables at all, than it is not an experiment (it"s a survey, with factorial-type set up) Note: the term pretest can also be used to refer to the trial run of some aspect of the study (pilot test) Examine effect of two or more ivs simultaneously. If not a true experiment of if you do experiment improperly, than . Alternative explanations for results become possible (i. e. threaten internal validity)