Experimental Design

• Independent Groups Design


  Groups/Conditions are completely separate of each other. Pps are randomly allocated to each group/condition,

  • Each group does only one condition of the experiment
• Repeated Measures


  Each pp completes all conditions of the experiment
• Matched Pairs 


  The pps in each group are carefully matched on important variables/characteristics relevant to the study. Then for each matched pair, one goes into condition A & one into condition B. This is repeated for each pair.
• A strength of Independent group design 


  Order effects not a problem, as they only take part in one condition.
• Weakness of Independent groups

  Individual differences between pps conditions & may be a confounding variable.

  • Less economical due to more pps needed, therefore takes more time
• Strength of Matched pairs

  Individual differences and pps variables are controlled because the same pps do both conditions. Fewer pps do both conditions. More cost-effective
• Weakness of matched pairs

  Order effects

  • Bored, Fatigue & practise effects
  • Counterbalancing can be used to try and over come - using the ABBA technique
• Strength of Matched Pairs

  Order effects are not a problem due to only taking part in the condition once
• Weakness of Matched Pairs

  Time-consuming to match participants, less economical than other designs

  • Even when matched carefully, can not control every single variable
• Strength of Natural experiment

  Provides opportunities for studies that would otherwise not be possible to research, mainly due to ethical reasons

  High external validity
• Weakness of Natural Experiment

  Natrually occurring events happen rarely

  Rare events mean its difficult to generalise findings to other situations

  Pps cannot be randomly allocated - harder to establish cause & effect
• Quasi expeirment have an IV based on existing difference between people. E.G: Age, gender, social class
• Stregnths of Quasi experiement

  Same as lab experiement, due to also being able to be done in a lab
• Weakness of quasi-experiment 


  Cannot randomly allocate pps to conditions - confounding variable

  IV is not deliberately changed by the researcher