repeated measures factorial anova

repeated measures factorial anova

1. week 10

2. advantage

   • Reduced variability due to individual differences
     • As the participant is the same at each measurement, it follows that individual differences will be the same, thus error from this source is reduced ie individual acts as own controll
   • more observations per participant
     • More observations will help balance out random error
   • increased efficency
     • fewer particiapnts are needed for the same power as an independant group design. repeated meassures allows for me data points
       • In the example, to obtain power = .56 an independent groups design needed N = 74, whereas the repeated measures design needed only N = 20
   • Many variables in psychology can only be studied in a repeated measures design
     • longtitudinal or sequenced based research
       • practice/trainin
       • age related changes
       • treatment effect overtime

3. disadvantage

   • order effects can threaten internal validity The participant’s behaviour may change over time, independently of the level of the independent variable
     • Practice effects
       • Improved performance in later conditions due to being more familiar with task and having done task before. difficult to avoid, can use, shorter task,give practice before to get to post practice level of preformance
     • fatigue effect
       • Reduced performance in later conditions due to lowered motivation or reduced energy. May avoid with rest periods; shorten testing time; increase interest in participant
     • Other psychological effects
       • Others, sometimes difficult to predict, psychological consequences of doing a task again. E.g., reduced anxiety on later testing conditions due to familiarity with the task, experiment, etc.
     • treatment carryover effect
       • the effect of earlier conditions will affect performance in later conditions. Can test for lingering effects; give ample time between conditions; use a filler task
     • Sensitiszation effects
       • participants behaviour changes because they learn (correctly or not ) the hypotheses . can produce participant bias and affect internal and construct validity . avoid with placebo and subtle variations of the conditions
     • CONTROLING FOR ORDER EFFECTSReducing the number of levels of the IV can reduce order effects  Fewer levels gives fewer opportunities for order effects to exert their effect  Downside is not all levels of the IV may be used
       • randomisation
         • Participants are randomly allocated to different orders of conditions  Ideally, randomisation will balance out the order of treatments, so each condition comes first about the same number of times it comes last. this may mean n group size may vary
           • counterbalancing
             • The counterbalanced condition that the participant completed might even be included as a factor in the design  It effectively acts as an independent groups factor to give a mixed factorial design  Will tell you whether there were any effects as a result of the order of presentation in the experiment
               • can highlight 1-main effects order AB vs BA (ie is it better to give meds first then other treatment) and 2- Interaction of treatment x order (clue you to order effects in your study)
             • complete counterbalanicing
               • Complete counterbalancing - An equal number of participants receive all possible orders of conditions. AB,BA counterbalancing works well with 2 levels
             • Latin square
               • latin square Common when there are 4 or more levels of the IV  Does not aim to provide a complete counterbalancing  However, it will ensure that each condition is tested in each relevant position the same number of times and each condition precedes and follows each other condition an equal number of times  This can be done with an even number of conditions, but not with an odd number of conditions  For an odd number of conditions, two Latin Squares can be used
                 • like a sudko in abcd must be in the row and collumn