experimental method

• experimental method involves manipulation of IV to measure the effect on the DV. It has a casual relationship.
• independent variable = variable that is manipulated

  Dependent variable = what is being measured (caused by IV)
• Operationalising = making the IV testable and the DV measurable
• • extraneous variables = factors other than the IV that might affect the DV if it's not controlled.unwanted variables that add error to the experiment.
  • Situational variables = features of a research situation that may influence participants behaviour if it's different for different conditions.
  • Participant variables = any characteristic of individual participants that may influence the study (age, gender, etc) Only acts as an EV when an independent groups design is used.
• • Directional (one-tailed) hypothesis = states that there will be a difference between results and predicts the direction of the difference
  • Non-directional (two - tailed) hypothesis = states that there will be a difference between results but not what the difference will be.
  • null hypothesis = states that there will be no difference between results - the IV will not have an effect on the DV. Any difference will be due to chance alone.
• types of experiment: lab

  takes place in a special environment where all variables are controlled

  + high degree of control

  + replicable

  - low ecological validity

  - experimenter bias
• types of experiment: field

  natural environment but variables still controlled

  + controlled IV

  + high ecological validity

  - loss of control over extraneous variables

  - lack of consent
• types of experiment: quasi

  IV has not been determined by the researcher, naturally exists instead eg. gender difference studies.

  + controlled conditions

  - can't randomly allocate pp
• types of experiment: natural

  IV not brought about by the researcher, would have happened anyway eg. reactions to earthquakes

  + provides opportunities that would usually be unethical

  + high external validity

  - not replicable

  - difficult to randomise
• sampling method: opportunity

  pp happen to be available at the time

  + time saving & less costly

  - not representative of whole population

  - researcher bias, choose who they want
• sampling method: random

  all equal chances of being selected, random number generator

  + no researcher bias of who picked

  - time consuming

  - volunteer bias, pp can refuse to take part
• sampling method: systematic

  every nth member selected from sampling frame

  + respresentative & no researcher bias

  - not truly unbiased
• sampling method: stratified

  subgroups within the population (age groups or genders). pps obtained from each group in proportion to their occurrence in the population using random selection.

  + no researcher bias

  - time consuming
• sampling method: volunteer sampling

  self selection by advert or when asked to

  + easy & not time consuming

  - volunteer bias, may attract a certain type of person

  - motivations like money could drive pp so may not take study seriously
• experimental design: independent groups

  pp only perform in one condition of the IV

  + no order effects

  + less demand characterisitics

  - individual differences

  - need more pps

  fixed by: random allocation of pps
• experimental design: repeated measures

  same pps take part in all conditions

  + eliminates individual differences

  + not as time consuming

  - order effects

  fixed by: counterbalancing - one half does one condition first and the other half second
• experimental design: matched pairs

  pairs of pps first matched on some variable, one member of each pair does different condition

  + no order effects

  + less demand characteristics

  - time consuming & expensive to match

  - large pool of pp needed