Experimental design

• Experimental design= the method you use to allocate your pps to complete conditions within your study
• Repeated measures:

  • All pps are tested in each of the IV conditions & results from each condition are compared
  • Can lead to order effects from the pps (changes in performance because both tasks are being carried out)
  • PPs may guess the aim of the study- demand characteristics
  • Don't need to recruit as many pps- minimised participant variables
• Independent groups:

  • PPs are divided into 2 groups & tested on one condition, results from each group are then compared
  • You need twice as many pps- more costly & time-consuming for the researcher
  • No order effects & less likely to be demand characteristics
  • PP effects: they are different pps so you have no control over natural variations between participants (one group could be more intelligent than the other)
• Matched pairs design:

  • Recruit one group of pps and then get pps who match them on relevant characteristics. Each group tested in 1 condition & compare results (twins are ideal pps; they're already matched)
  • Very time-consuming (need to find lots of matching pps)
  • Can rarely find exact matches & can never be sure you have matched all the relevant variables.
• Most scientific experiments meet the criteria of:

  • researcher manipulates the IV
  • researcher randomly assigns pps to conditions
  • researcher has control over extraneous variables

  Having these features allows you to establish cause & effect.
• Laboratory experiment:

  • Experiment conducted in a specific environment, where variables can be carefully controlled.
  • Involves manipulation of an IV & measurement of DV- therefore enables us to establish cause & effect
  • High level of control of extraneous variables
  • PPs randomly allocated
  • Artificial setting so reduces ecological validity
  • Higher chance of demand characteristics
  • IV manipulated by researcher- systematic & replicable so more likely to produce reliable results.
• Field experiment:

  • Conducted in a more natural environment & IV manipulated by researcher.
  • Higher levels of ecological validity & reduced demand characteristics.
  • Involves manipulation of IV & measurement of DV- therefore enables us to establish cause & effect.
  • Cannot control all EVs to same extent as lab.
  • PPs are not necessarily randomly allocated.
• Natural experiment:

  • Changes in the IV are natural rather than manipulated by the researcher, as it would be unethical or impractical to set conditions.
  • Can be higher in ecological validity if in natural environment
  • No random allocation
  • Difficult to replicate- may have low reliability
  • Not all extraneous variables are controlled
  • Not a 'true' experiment due to lack of manipulation of an IV & so cannot therefore state cause & effect.
  • Can allow for research that otherwise couldn't take place.
• Quasi experiment:

  • The IV is not manipulated by the researcher as it has naturally occurring differences between people.
  • Can allow research into individual differences
  • Low control of extraneous variables
  • Cannot randomly allocate pps
  • Not a 'true' experiment due to lack of manipulation of an IV & so cannot therefore state cause & effect.
• The purpose of experiments is to compare the difference(s) between 2 variables.
• A pilot study= a small scale version of your main study (fewer pps & cheaper resources).
• Pilot studies are used as a trial for all aspects of the method so that improvements can be made & they help to test the validity & reliability of the method selected- to help you identify any potential flaws.