Experimental design

• The types of experimental designs include:

  • repeated measure design
  • independent groups design
  • matched pairs design
• In a repeated measures design the same participants take part in all conditions of the experiment.
• Strengths of repeated measures designs:

  • controls for participants variables (as the same participants are used in each condition)
  • fewer participants are needed
• Weakness of repeated measures designs:

  • order effects (e.g. practice or fatigue) can influence results
  • demand Characteristics: participants may guess the aim of the experiment and alter their behaviour.
• An independent groups design is when different participants are used in each condition of the experiment.
• A strength of independent groups design is that there is no order effects as each participant only does one condition.
• A weakness of independent groups design is that participant variables may differ between groups, which can affect the results. It also requires more participants.
• Matched pairs design is when pairs of participants are matched on key characteristics (e.g. age, IQ), and one member of each pair is assigned to each condition.
• Strengths of matched pair design:

  • reduces participant variables: matching ensures that key differences (e.g., memory ability) are less likely to confound results.
  • no order effects: each participant only takes part in one condition, avoiding practice or fatigue effects.
• Weaknesses of matched pairs design:

  • time-consuming: finding closely matched pairs can take a lot of time and effort.
  • matching may be imperfect: it’s difficult to match participants on every relevant variable, which may still lead to variability.
• To control for order effects in a repeated measures design, counterbalancing is used. This is when half the participants do condition A, the other half start with condition B.
• To control for participant variables in independent groups design, random allocation is used to balance participant variables across groups. It attempts to evenly distribute participant characteristics across the conditions of the experiment.