Research Methods Flashcards - A Level Psychology AQA

• Ethics

  • informed consent
  • deception
  • debriefing
  • right to withdraw
  • protection from harm
  • privacy
  • confidentiality
• Informed Consent

  • should be given at the start of the study

  Contains:

  • tells aims of experiments
  • details + procedures + what is required
  • ethics: can withdraw any time + kept anonymous + receive full debrief
  • sign + date
• Deception

  • so don't lie, but sometimes necessary to protect them
• Debriefing

  • after study, told the true aims, further information, returns to originate state
• Right to withdraw

  • where everyone should be aware they can leave at any point
• Protection from harm

  • like physical, psychological on animals
• Privacy

  • as people don't expect to be studied + observed without knowledge
• Confidentiality

  • so no one can identify participants, anonymous
• Way to go around consent

  • parental consent for anyone under 16
  • presumptive consent, ask similar demographic
  • prior consent
• Sampling

  • random sampling
  • systematic sampling
  • volunteer sampling
  • stratified sampling
  • opportunity sampling
• Random sampling

  • write list of total target population
  • names on paper
  • all names into hat
  • pick out name
• Random sampling

  + unbiased

  + quick and easy

  - unrepresentative sample by chance
• Systematic sampling

  • use a system
  • every nth person
• Systematic sampling

  + no researcher bias as no control

  - unrepresentative sample by chance
• Volunteer sampling

  • ask for any volunteers
• Volunteer sampling

  + quick and easy

  + less dropouts

  - demand characteristics

  - volunteer bias
• Stratified sampling

  • identify subgroups in total population
  • work out proportions
  • randomly select each condition based on proportions
• Stratified sampling

  + representative

  - more time consuming
• Opportunity sampling

  • whoever is available
  • most convenient
• Opportunity sampling

  + quickest and easiest method

  - unrepresentative sample

  - researcher bias
• Independent variable

  • is event that is directly manipulated by an experimenter in order to test its effect on another variable
• Dependent variable

  • is a measurable outcome of the action of the independent variable in an experiment
• Operationalise

  • is ensuring the variables are in a form that can be easily tested
• Hypotheses

  • non-directional
  • directional
  • null
• Non-directional hypothesis

  • There will be a correlation between covariable 1 and covariable 2
  • There will be a difference between the results in condition 1 and condition 2
• Directional hypothesis

  • There will be a positive/negative correlation between covariable 1 and covariable 2
  • Condition 1 will have a higher/lower result in (_?) than condition 2
• Null hypothesis

  • There will be no correlation between covariable 1 and covariable 2
  • There will be no difference between the results in condition 1 and condition 2
• Control variables

  • extraneous variable
  • cofounding variable
  • investigator effect
• Extraneous variable

  • is any variable other than the IV that could affect the results (DV) that we should attempt to control
• Cofounding variable

  • are any variables other than the IV that has affected the results (DV)
• Investigator effect

  • is anything that an investigator does that affects a participant's performance in a study, other than what was intended (maybe cofounding/extraneous variable)
• Experimental design

  • repeated measures design
  • independent groups design
  • matched pairs design
• Repeated measures design

  + less participants needed

  + no participant variables

  - order effects (practice gets better, boredom)

  - demand characteristics
• Counterbalancing

  • experimental technique used to overcome order effects
  • half the participants do the first condition then the condition, half the participants do the second condition then the first
• Independent groups design

  + less demand characteristics

  + no order effects

  - double participants needed

  - researcher cannot control the effects of participant variables
• Matched pairs design

  + no participant variables

  + avoids orders effects

  - very time consuming

  - not possible to match everyone
• Observations

  • participant observation
  • non-participant observation
  • covert observation
  • overt observation
  • controlled observation
  • naturalistic observation
• Participant observation

  • where researcher takes part/acts as participant in own research
• Participant observation

  + first hand/richer observations

  - harder to record while trying to fit it

  - researcher bias
• Non-participant observation

  • where researcher is not part of it