CH 11

• Within-Subjects Experiment


  Subjects are assigned to more than one treatment condition
• Within-Subjects Factorial Design


  Assigns subjects to all levels of two or more independent variables
• Mixed Design


  An experiment with at least one between-subjects and one within-subjects variable
• Advantages of Within-Subjects Designs

  • Use fewer subjects
  • Save time on training
  • Greater statistical power
  • More complete record of subjects' performance
• Disadvantages of Within-Subjects Designs

  • Subjects participate longer
  • Resetting equipment may consume time
  • Treatment conditions may interfere with each other
  • Treatment order may confound results
• We can't use a within-subjects design when one treatment condition precludes another due to interference
• Order Effects


  Positive (practice) and negative (fatigue) performance changes due to a condition's position in a series of treatments
• Progressive Error


  Encompasses both positive and negative order effects
• Counterbalancing


  A method of controlling order effects by distributing progressive error across different treatment conditions
• Subject-by-Subject Counterbalancing


  Controls progressive error for each subject by presenting all treatment conditions more than once
• Reverse Counterbalancing


  Administers treatments twice in a mirror-image sequence (e.g. ABBA)
• Nonlinear Progressive Error


  Curvilinear (inverted-U) or nonomonotonic (changes direction) progressive error that cannot be graphed as a straight line
• Reverse counterbalancing only controls for linear progressive error
• Block Randomization


  A subject-by-subject counterbalancing technique where each subject is assigned to several complete blocks of treatments, with each block presenting the treatments in a different random order
• Subject-by-subject counterbalancing can result in long-duration, expensive, or boring procedures, especially as the number of treatments increases
• Across-Subjects Counterbalancing


  Techniques that present each treatment once and control progressive error by distributing it across all subjects
• Complete Counterbalancing


  Uses all possible treatment sequences an equal number of times, with subjects randomly assigned to one of these sequences
• Partial Counterbalancing


  A form of across-subjects counterbalancing where only some of the possible (N!) orders are presented
• A within-subjects design is usually preferable when you need to control large individual differences or have a small number of subjects, but may not be feasible if the experiment is long or there is a risk of asymmetrical carryover