RC 225: Experimental Psychology

• Within-subjects design


  In a within-subjects experiment, subjects are assigned to more than one treatment condition

  
• Within-subjects design


  Researchers measure subjects on the dependent variable after each treatment

  
• Within-subjects design


  Subjects participate in more than one treatment condition and serve as their own control
• Within-subjects design


  We compare their performance on the dependent variable across conditions to determine whether there is a treatment effect
• Two major sources of confounding variables in within-subjects designs

  • Confounding from environmental variables
  • Confounding from time-related factors
• Confounding from time-related factors


  • Participants are measured over time
  • Many events can affect a participant over time, so observed changes may be due to time-related effects rather than treatment differences
• Time-related effects


  • Maturation
  • Instrumentation
  • History
  • Statistical regression
  • Order effects
• Maturation


  A third grader may read better after three months because she is maturing rather than because of a reading intervention
• Instrumentation


  Measuring instruments, including researchers themselves, may affect scores
• History


  An unrelated event may influence the outcomes
• Statistical regression


  Extreme scores tend to become less extreme over time due to statistical regression
• Order effects


  Positive and negative performance changes due to a condition's position in a series of treatments
• Progressive error


  Encompasses both positive and negative order effects
• Order effects


  • Participation in one condition can directly affect participation in another treatment condition
  • Holding order constant would confound the experiment
• Practice effects


  Outcome measure improves on each administration due to the effects of practice taking the test
• Fatigue


  A person might perform worse on each administration due to fatigue
• Carryover effects


  The effects of participation on the first treatment carry over into the second treatment, making the outcome better or worse than if it had been the only treatment
• Contrast effects


  A type of carryover effect where the perception of one treatment is affected by its contrast with another
• Dealing with time-related and order effects

  • Time-related effects like history and maturation are less of a problem in shorter experiments
  • If order effects are likely to be a major problem, a between-subjects design may be better
• Counterbalancing


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


  • Subject-by-subject counterbalancing
  • Across subjects counterbalancing
• Counterbalancing and time-related effects


  • Half the participants undergo treatment 1 then 2, the other half undergo 2 then 1
  • Still a within-subjects design, just with different orders
• Counterbalancing and order effects


  • Counterbalancing does not eliminate order effects, but it eliminates them as a confound by spreading them equally across conditions
• Counterbalancing and the number of treatments

  • As the number of treatment conditions increases, the number of possible counterbalancing sequences increases dramatically (n!)
• Subject-by-subject counterbalancing


  Controls progressive error for each subject by presenting all treatment conditions more than once
• Subject-by-subject counterbalancing techniques


  • Reverse counterbalancing
  • Block randomization
• Reverse counterbalancing


  Only controls for linear progressive error, can confound the experiment
• Block randomization


  Assigns each subject to several complete blocks of treatment, with each block presenting the treatments in a different random order
• Advantages of within-subjects designs


  • Use fewer subjects
  • Save time on training
  • Greater statistical power
  • More completed record of subject's performance
• Disadvantages of within-subjects designs

  • Subjects participate longer
  • Resetting equipment may take time
  • Treatment conditions may interfere with each other
  • Treatment order may confound results