experimental techniques

• repeated measures design 


  all participants take part in all conditions - one group of participants complete one condition then the same group completes the next condition - e.g. if you were testing the effect of noise on memory, group of participants could be tested in noisy conditions (condition A) in the morning and then the same group would be tested in silent conditions (condition B) in the afternoon
• repeated measures advantages


  no participant variables - no individual differences between participants in each condition - e.g. participants in condition A won't differ in any way from participants in condition B because they're the same participants - increases internal validity

  requires half number of participants - less participants needed compared to independent groups or matched pairs design - e.g. 10 participants in each condition only 10 needed compared to 20 for other designs - cheaper and less time consuming
• repeated measures disadvantages


  suffers from order effects - order participants complete conditions may affect performance - e.g. participants may do better in second test because they have already had practice at the task (practice effect) or do worse as they're bored (boredom effect) - lowers internal validity

  high demand characteristics - participants have more of a chance of guessing aim of experiment which might affect behaviour - e.g. completing memory task in noisy conditions then silence participants may guess aim and change behaviour - lower internal validity
• independent groups design


  participants placed into separate independent groups - each group completes one condition of the experiment only - then compare the DV of each group - e.g. if testing effects of noise on memory you would have one set of participants being studied in a noisy condition (condition A) and different group of participants being tested under the silent conditions (condition B)
• independent groups advantages


  doesn't suffer from order effects - order participants complete different conditions doesn't affect performance because participants only complete one condition - e.g. won't suffer from boredom or practise effects - increases internal validity

  low demand characteristics - participants have less chance of guessing aim - e.g. as participants only take part in one condition they aren't necessarily aware of research expectations so behaviour less likely to change to fit in with expectations - increases internal validity
• independent groups disadvantages


  low degree of control over participant variables - might be individual differences between participants in each condition that haven't been controlled and could influence DV - e.g. participants in one condition may be more intelligent - lower internal validity

  requires twice as many participants - more participants needed to achieve same amount of data - e.g. to achieve 10 participants in each condition 20 participants needed unlike 10 in repeated measures - more expensive and more time consuming
• matched pair designs 


  different participants used in each condition - participants matched into pairs based on important characteristics that may influence results - one of each pair then takes part in condition A and one in condition B - e.g. for every young male in condition A there is a young male in condition B, for every older female in condition A there is an older female in condition B
• matched pairs advantages


  doesn't suffer from order effects - order participants complete different conditions doesn't effect performance as participants only complete 1 condition - practise and boredom effects don't have an effect as they only do one task - higher internal validity

  low demand characteristics - participants have less chance of guessing aim - e.g. participants only take part in 1 condition so aren't aware of expectations so behaviour less likely to change to fit in - increases internal validity
• matched pairs disadvantages


  less control over participant variables - still more chance of individual differences between participants in each condition than repeated measures - e.g. participants in condition A may be on average much more intelligent than those in B - decreases internal validity

  requires twice as many participants - more participants needed for same amount of data - to achieve 10 participants in each condition 20 needed compared to 10 for repeated measures - more expensive and time consuming
• random allocation


  one way of controlling researcher bias from process of allocating participants to conditions in an independent groups design - involves assigning all participants a number and putting them on slips of paper and into a hat - numbers drawn out so that every other number goes into condition A - should help reduce issues of bias and help control variables
• counterbalancing 


  one way of controlling order effects in a repeated measures design - involves half of participants starting in condition A followed by condition B and half starting in condition B followed by condition A (ABBA) - balances out problem of order effects as it means its not always the same condition performing better due to practice or due to boredom