Paper 2 Research methods

• Research aim


  A general statement of the purpose of the study- should make clear what the study intends to investigate
• independent variable


  Variable that is changed (experimental/control)
• Dependent variable


  Variable that is measured
• Operationalisation


  Being precise and clear about what is being manipulated or measured. Make is testable and repeatable.

  (instead of how many people do this, say the number of people that do this)
• Hypothesis


  A clear precise testable statement that states the relationship between variables.

  Nul = No relationship

  Experimental/alternative = change
• 1. Extraneous variables

  2. Confounding variables


  1. Another variable other than IV which may affect DV if not controlled.

  Situational (temp lighting) or participant (age, intelligence)

  2. Any variable other than IV that may affect DV, can't be sure of true source.
• Demand characteristics


  Sues in an experiment that tell the participant what behaviour is expected
• Investigator effect


  Effects of the investigators behaviour on the outcome of the research (selection of participants)
• 1. Directional hypothesis

  2. Non - directional hypothesis


  1. States direction (more/less , quicker/slower)

  2. No direction (affects , changes)
• Experimental method : Lab experiment + AO3


  An experiment that takes place in controlled environment when investigator manipulates IV, and controls DV

  + High control

  + Easy replication

  - Lack ecological validity

  - Demand characteristics
• Experimental method : Field experiment + AO3


  An experiment which takes place in natural setting within which the investigator manipulates IV. (Hofling hospital study on obedience)

  + High ecological validity

  + Genuine behaviour

  - Lack of control -> hard to replicate

  - Ethical issues
• Experimental method : Natural experiment + AO3


  An experiment where IV change is brought by researcher, would have happened if researcher wasn't there (natural events - tsunami)

  + Ecological validity

  + Very ethical

  - Not controlled

  - Ps may know they are being watched
• Experimental design : Independent groups + AO3


  Ps are allocated into different groups each does one condition

  + Order effects are not a problem

  - Participants are not the same
• Experimental design : Repeated measures + AO3


  All Ps take part in all conditions of the experiment

  + Participant variables are controlled

  - Order effects
• Experimental design : Matched pairs + AO3


  Ps are matched on variables that may affect DV. The pair is split and assigned to conditions

  + Demand characteristics and order effects are less of a problem

  - Can't be matched easily
• Counter balancing


  Involves systematically changing the order of tasks for participants in a 'balanced' way to 'counter' the unwanted effects on performance in any one order.
• Internal validity


  Look 'in' at experiment (what goes on in a study)

  Is what experiment measured what was intended?
• External validity


  Looking 'outwards' to everyday life.

  Can you generalise results to a wider setting?
• Standardised procedure


  Procedure that is the same for all participants except where a variation is introduced to test a hypothesis
• Target population


  A group of people who are the researchers focus
• Random sampling + AO3


  Collect list of members from target, all names assigned a number (lottery method)

  + Free from researcher bias, no influence

  - Time consuming

  - Selected Ps may reject
• Opportunity sampling + AO3


  Use people from target at the time and willing to take part. Ask members if interested

  + Convenient, less cost, less time

  - Unrepresentative

  - Researcher bias
• Volunteer sampling + AO3


  Advertise an investigation and Ps select themselves to take part

  + Easy and quick, less time

  - May attract certain profiles

  - Volunteer sabotage
• Systematic sampling + AO3


  Choose Ps in systematic way, like every Nth participant

  + avoid researcher bias

  - Possible to get all male for example
• Stratified sampling + AO3


  Classify population into subgroups (strata) and choose sample which consist of people from each strata

  e.g. Simpsons cast 60% male 40% female so the sample size has to be 60% male and 40% female

  + Random selection

  + Generalisable

  - Cannot effect all ways people are different
• Pilot study


  A trial run before the actual experiment
• Social desirability bias


  A tendency to give socially approved answers to questions about oneself.
• Questionnaires : Psychometric testing


  Assessments to measure psychological attributes e.g. personality, intelligence
• Questionnaires : Open question


  Respondent writes response in their own words

  e.g. How is your week?
• Questionnaires : Closed question


  Respondent chooses response from fixed responses

  e.g. scale from 1-10
• Questionnaires : Quantitative data

  Number based, closed questions

  + Simple to analyse

  - Less external validity
• Questionnaires : Qualitative data


  Letter based, open questions

  + More detail

  + High external validity

  - Difficult to analyse
• Interviews : Structured


  All applicants are asked the same set of standardized questions
• Interviews : Unstructured


  interviewers are free to ask the applicants anything they want
• Interviews : Semi - structured


  A list of questions have been worked out in advance, but interviewers are free to ask follow up Qs
• Correlation coefficient


  a statistical index of the relationship between two things (from -1 to +1)
• Observational studies: Covert or overt + AO3


  Covert = Observer is not seen by participants

  + Remove demand characteristics, increase internal validity

  - Ethical issues = lack consent

  Overt = Observer is not seen by participants

  + More ethically acceptable

  - Might influence behaviour as Ps know
• Observational studies: Participant or non-participant + AO3


  Participant = Observer gets involved in behaviour (Stanford prison)

  + Increase external validity

  - Lose objectivity

  Non - participant = Observer does not get involved

  + Maintain objectivity

  - Lose insight
• Observational studies: Structured + AO3


  Before researcher begins they determine behaviour to be observed and create standardised checklist and record frequency of behaviour.
• Observational studies: Controlled or naturalistic + AO3


  Controlled = Regulated by researcher, reduce naturalness NOT LAB EXPERIMENT

  e.g. strange situation

  + Replication is easier, less extraneous variables

  - May produce findings unapplicable to life

  
  

  Naturalistic = Watching and recording behaviour in a setting which it would normally occur. NOT NATURAL EXPERIMENT

  + High external validity

  - Lack of control hard to replicate

  - Uncontrolled extraneous variables