(3) Experimental Designs

• to know that the IV affects the DV, we need a comparison condition (something to compare with) - a different level of the independent variable
• Experimental Design :

  • refers to the way in which participants are used in the experiment
  • used : how participants are arranged in relation to the different experimental conditions (levels of IV)
• The three different types of experimental designs :

  • independent group
  • repeated measures
  • matched pairs
• Independent Group :

  • different participants are used in each condition of the experiment
  • random allocation used to decide which condition each participant should be allocated to - reduces bias
  • random allocation ensures each participant has an equal chance of being assigned to one group or another
  • one group of participants in the experimental condition - group 1
  • a different group in a control condition - group 2
  • performance of two groups are compared
• EVALUATION of Independent Groups

  Negatives :

  • Participants who occupy the different groups are not the same : participant variable (individual differences) may effect the IV therefore the DV, so invalid and unreliable results — to deal with is researcher uses random allocation
  • less economical than repeated measures group : each participant contributes single results only - twice as many participants needed to produce equivalent data that is collected in a repeated measures design
• EVALUATION of Independent Groups

  Positives :

  • order effects are not a problem - participants only experience the experiment once
  • participants are less likely to guess the aims - lower chances of demand characteristics (eg please-U or Screw-U effect)
• Random Allocation :

  • used to address the problem of participant variables (usually in independent group design) - participants randomly allocated to different experimental conditions
  • Attempts to evenly distribute participant characteristic’s across the conditions of the experiment - uses random techniques
  • random techniques : picking pieces of paper out of a hat labelled A or B (the different conditions) / or use a random number generator
• Random allocation techniques :

  • pick a piece of paper out of the hat labelled A or B for the different conditions
  • random number generator
• Repeated Measures :

  • all participants experience both conditions of the experiment
  • all participants would first experience the experimental condition and then the same participants would experience the control condition
  • two sets of data from both conditions are compared to see if theres differences
• EVALUATION of Repeated Measures

  Negatives :

  • each participant has to do at least two tasks and order of task is significant (order effect) : could have continuing effects - researcher uses counterbalancing to deal with this
  • Order effects : repeating two tasks could create boredom/ fatigue which may cause deterioration on the second task -so order matters
  • Order effects : participant performance may improve through effects of practice so would perform better on second task - order acts as a confounding variable
  • demand characteristics -work out aim of study when they experience all cond.
• EVALUATION of Repeated Measures

  Positives :

  • participant variables are controlled - individual differences
  • fewer participants needed - so can collect more data
  • guarantees that you are comparing 'like with like' - all participants have done both conditions so no individual differences between the group
• Matched Pairs :

  • participants are paired together on a variable(s) relevant to the experiment
  • one from the pair does experimental condition the other does control (or the other different conditions)
  • attempts to control for the confounding variable of participant variables - individual differences between participants reduces the chances of randomly getting a 'better' group
• EVALUATION of Matched Pairs

  Positives :

  • participants only take part in a single condition so order effect and demand characteristics are less of a problem
• EVALUATION of Matched Pairs

  Negatives :

  • participant can never be matched exactly - there will important differences between partners that may affect the DV
  • matching is time consuming and expensive (especially is a pre-test is required eg IQ test or physical test) - less economical than other designs