experimental methods

• laboratory experiments: defined by the high level of control the researcher has over all the variables in the study
• in lab experiments the experimenter will control environmental factors e.g noise and temperature as well as the experience each participant has by using standardised procedures
• strengths of lab experiments

  • by controlling variables outside IV and DV, confidence in establishing cause and effect relationship between changes in IV and observed difference in DV
  • high internal validity, observed change in DV is due to change in IV
  • lab studies easily replicated due to standardised procedures
• weaknesses of lab experiments

  • lack ecological validity, can't be applied to real world situations
  • lack mundane realism, lowering external validity
  • pps may alter behaviour due to demand characteristics in order to match the aim
• field experiments: defined by conducting experiment in naturalistic settings. this change in location is an attempt to avoid artificial nature of lab studies
• locations field studies are conducted in can include shops, work and school, or anywhere the participants would be expected to behave naturally
• strengths of field studies

  • pps behave more naturally in normal environment - higher ecological validity, can be applied to other naturalistic settings
  • has mundane realism
  • if pps are unaware they won't show demand characterisitics
• weaknesses of field studies

  • lack control over possible extraneous variables that could influence measurement of dependent variable
  • difficult to randomly assign pps to separate conditions resulting in a change in DV that may be due to participant variables, reducing internal validity
• natural experiments: two levels of independent variables have occurred naturally in the real world without influence of researcher. researcher simply records change in DV between two levels of IV
• natural experiments can happen whenever an event causes people to form into levels of the independent variable. e.g this could be a natural disaster that only impact some members of a community
• strength of natural experiments

  • allows research in areas that couldn't happen in controlled experiments due to ethical or costs reasons
  • high in external validity as natural experiments are an example of real behaviour occurring in the real world free of demand characteristics
• weaknesses of natural experiments

  • researcher has no influence so extraneous variables can't be controlled, so researcher shouldn't claim they have found a cause and effect relationship
  • these are rare events that can't be replicated exactly to test for reliability
• quasi experiment - participants cannot be randomly assigned between levels of IV. often because the level of IV is an innate characterisitic of the participants
• lab experiments: IV is changed by researcher between conditions of experiment. any change in dependent variable will be measured while all possible variables that could change the DV are kept consistent between conditions
• field experiments - IV still changed by researcher between conditions of experiment and researcher measures difference in DV
• natural experiment - IV not changed by researcher between conditions of experiment. changes in DV still measured but other possible variables that could change the DV cannot be controlled
• in quasi experiments there is a wide range of characteristics that psychologists would like to study, but these characteristics already exist in the sample e.g gender, age, income level, education level
• in quasi experiments IV already exists in pps so they cannot be randomised between conditions of experiment. difference in DV will be measured while other variables that could change DV are kept constant
• strength of quasi

  these experiments are the only way to experimentally study factors that are pre-existing characteristics of participants
• weaknesses of quasi

  there may be other factors related to the level of IV that cannot be controlled for, these change systematically between levels of IV and alter the measurement of the dependent variable. known as confounding variables
• independent groups design: diff pps complete in each of the two (or more) conditions of the experiment. pps are randomly allocated to each condition to avoid researcher bias when assigning to conditions
• independent groups design

  produces unrelated data, the individual data points in one condition cannot be paired with any of the data points in the other condition
• problem w independent groups designs

  if more pps with a particular characteristic are randomly assigned to one of the groups (e.g age) this can influence the measurement of the DV (extraneous variable)
• repeated measures design: the same participants complete in each of the two (or more) experimental conditions
• repeated measures design

  produces related data, each participants score (data point) in one condition can be paired with a data point (their own score) in the other condition
• problem with repeated measures design

  • order effects: taking part in first condition influences performance in second condition e.g worsen due to fatigue or boredom, or improve due to practice.
  • pps more likely to figure out the aim and alter behaviour due to demand characteristics
• repeated measures design

  counter balancing attempts to control for order effects. this uses ABBA format, half the pps complete condition A first and B second, the other half of the sample start with condition B then A
• strengths of independent groups design

  • pps less likely to work out the aim than repeated measures design as they only take part in one condition, meaning reduced demand characteristics
  • no order effects as participants only take part in one condition
• strengths of repeated measures design

  • needs half the pps compared for independent groups design for the same amount of data
  • pps variables between conditions is not a problem as pps take part in both conditions
• matched pair designs: diff pps complete in each of the 2 (or more) conditions of the experiment. pps are first assessed and ranked on a characteristic (e.g aggression) and then the top 2 pps (then each following 2) are randomly assigned to separate conditions
• matched pairs design produces related data, each pps score (data point) in one condition can be paired with a data point (the pp matched to them) in the other condition
• strength of matched pairs design

  • reduced pp variables as pps are matched on a relevant characteristic
  • no order effects as pps only take part in one condition
• weaknesses of matched pairs design

  • takes longer to set up than other experimental designs
  • needs twice as many pps as a repeated measures design
  • participants are similar but not identical so there may still be some participant variables between conditions that influence the dependent variable