experimental design

• There are three different experimental designs:

  • Repeated measures
  • Independent groups
  • Matched pairs
• repeated measures design is when one single group of people perform all of the conditions of the experiment.

  A comparison is made between a single participants score on one condition to their score on other condition.
• repeated measures example

  in a memory study they would have their memory tested in the morning and then again in the afternoon.
• repeated measures strengths

  • (+) No participant variables as same participants used in both conditions
  • (+) Fewer participants are needed which makes the design more economical
• repeated measures weaknesses

  • (-) Demand characteristics likely as both conditions are seen which affects validity
  • (-) Potentially could be order effects, confounding the results and making them invalid (e.g. may do better on the second condition due to the practice effect or being less anxious OR may do worse on the second condition due to the boredom effect)
• Order effects can occur in a repeated measures design and refers to how the positioning of tasks influences the outcome e.g. practice effect or boredom effect on second task
• order effects can be controlled by counterbalancing. One way is the ABBA method- AB or BA

  Divide participants into two groups:

  Group 1: each participant does A and then B

  Group 2: each participant does B and then A

  So any order effects will cancel each other out
• An independent groups design is when there are different groups of people who each perform only one condition of the IV.

  The scores of the group who complete one condition are compared to the other group in the other condition
• independent groups strengths

  • (+) Reduction in demand characteristics- because participants only see one condition so cannot guess aim as easily.
  • (+) No order effects – there is no practice or fatigue because participants are only tested once.
• independent groups weaknesses

  • (-) There are more likely to be participant variables – differences in people that can explain results other than IV.
  • (-) More participants are needed than in a repeated measures design which means it is less economical
• how are participant variables dealt with in an independent groups design?

  random allocation
• If using an independent groups you should use random allocation. It means putting your participants into which condition they will be doing RANDOMLY.
• random allocation is important in an independent groups design because

  • It minimises the chance of having any bias in deliberately picking people to do a particular condition.
  • It minimises the chance of participant variables affecting the study as it (theoretically) distributes participant variables evenly.
• matched pairs design is when you pair your participants up on a certain quality (e.g putting people with similar intelligence or age together) that is believed to affect the performance on the DV.

  Each person from the pair does one condition and results are compared to their partners in the other condition.
• matched pairs example

  For example, in the memory study you would pair people on age (as this could affect memory) and have one person from the pair (of the same age) do the one condition and the other one do the other condition.
• matched pairs strengths

  • (+) Reduces participant variables as participants matched on certain characteristics that could affect the DV.
  • (+) Reduces demand characteristics because the participants only take part in one experimental condition so are therefore less likely to guess the aim of the study.
  • (+) No order effects because participants are only tested once.
• matched pairs weaknesses

  • (-) More participants are needed to take part in the study as they each take part in only one condition. Additionally, the researcher probably has to start with a large group of participants to ensure they can obtain matched pairs on key variables.
  • (-) Matching participants on the relevant variables can be difficult and time-consuming
  • Participants are similar but not identical so there may still be some participant variables that affect the DV