Experimental designs

• Experimental design


  • Refers to the way in which participants are 'used' experiments.
  • In order to find out if the IV did effect the DV we need a comparison so we need at least 2 groups

  
• Repeated measures design


  • All participants take part in each condition of the experiment

  
• Independent group design


  • Different groups of participants take part in each of the conditions level of the IV

  
• Matched pairs design


  • Pairs of participants are first matched on some key variables that may affect the DV. Then one member is assigned to each condition

  
• Repeated measures design strengths

  • Need less participants
  • Participant variables do not have an effect on variables

  
• Repeated measures design weaknesses

  • The differences in the DV result could be due to extraneous variables such as differences in the test/ task given to participants. Dealing with this: Make sure that the tests are equivalent or half do test 1 and half do test 2
  • Demand characteristics can affect results as participants may guess the aim of the study. Dealing with this: Single blind, use a cover story about the aim of the test to stop them guessing what the study is

  
• Repeated measures design weaknesses

  • The order of the conditions may affect performance or results because of order effects. Participants may do better or worse due to practice, fatigue or boredom effect. Dealing with this: Counter balancing ABBA (1/2 do condition A first and 1/2 do the condition B)

  
• Independent group design strengths


  • Order effects do not have an effect on the results
  • Different tests or tasks are not needed for each condition or level of the IV
  • Demand characteristics are not going to have an effect on the results

  
• Independent group design weaknesses

  • There is no control of participant variables. Dealing with this: Randomly allocate participants to conditions to distribute them more evenly
  • You need to have a lot of participants which can be time consuming to recruit. Dealing with this: Be prepared to spend more time and money

  
• Matched pairs design strengths

  • Neither participant variables or order effects have an effect on the results
  • Demand characteristics are not going to have an effect on the results
  • Different tests or tasks are not needed for each condition or level of the IV

  
• Matched pairs design weaknesses

  • You need to have a lot of participants which can be time consuming to recruit. Dealing with this: Be prepared to spend more time and money
  • Matching can take a long time to complete. Dealing with this: Restrict matching pairs
  • You may actually control all participant variables as you might have missed important ones out. Dealing with this weakness: Conduct a pilot study considering key variables