Research Methods



Lab: IV manipulated in controlled setting 

  • high control, cause and effect, relicable
  • low realism, demand characteristics

Field: IV is manipulated in natural setting

  • higher realism
  • lower control, ethical issues

Natural: IV has been manipulated naturally, effect on DV recorded

  • high ecological validity, unique research
  • low control, opportunities may be rare

Quasi: IV based on existing difference, between people, effec on DV recorded

  • high realsim
  • little to no control
1 of 10

Experimental Design

Independant Group Design: different ptps in each condition

  • no order effects 
  • ptp veriables not controlled, less economically valid

Repeated Measures Design: ptps tested in every condition

  • no ptp variables, more economically valid
  • order effects, demand characteristics

Matched Pairs Design: matches people who are similar through similar characteristic screening (monozygotic = share genetically same characterists)

  • no order effects
  • cannot be matched exactly, more time consuming
2 of 10


Extraneous: nuisance variables but randomly distributed

Confounding: Vary systematically within IV

Demand characteristics: ptps second guess at the aim so alter behaviour

Investigator effects: unconsious influence of researcher on situatio (ie gender, age race etc)

Standardisation: ensuring all ptps are subject to same experience 

3 of 10

Ethical Issues

Informed consent: advising ptps of what is invleved, may reveal research aims

  • get permisission
  • either presumptuive, prior, general, retrospective consent

Avoidance of Deception: telling the truth (rectified in debrief)

Protection from Harm: duty to protect ptpt from psychological and or physical risk/harm

Anonymity/ confidentaility: protecting personel data 

4 of 10

Sampling Methods

Random: all members of population have equal chance of selection

  • no researcher bias
  • time consuming, could end up with biased sample

Stratified: sample refelcts proprtions of people with different population strata

  • no reasercher bias, representative
  • cannot account for all sub groups

Opportunity: choosing whoever is availible

  • convienient
  • researcher bias, unrepresentative

Volunteer: ptps select them selves

  • less time- consuming 
  • attracts certain profile of people
5 of 10

Observation Methods

Naturalistic: behaviour observed where it would normally occur, no control from reseracher 

  • high realism 
  • low control

Controlled:some control over environment, including manipulation of veriables to observe effects

  • high realism (especially covert)
  • low controll, some extranous variables may be controlled

Covert/ Overt: oberving ptps with or without knowlege 

  • covert is ethically unsound, but good as low ptp effect 
  • overt can effect ptp behaviour 

Participant/ non-ptp: to join group or remain outsider

  • ptp obs. have increased realism, risk of 'going native'
  • non-ptp has more objectivity, less insight
6 of 10

Observational Design

Structured: resercher controlls whats recorded

  • may miss behavious 

Unstructured: reseacher records everything

  • more info
  • lots of qualitative data to analyse

Event Sampling: count events as they occure

  • useful for infrequent behaviour
  • may miss complexity

Time Sampling: count event at timed intervals

  • less effort
  • may not be representative of whole behaviour 
7 of 10

Self- report Techniques

Questionairs: pre-set list of written open/closed questions

  • can be easily distributed to many people, easy to analyse
  • social deirablitlty bias, low response rate

Structured interview: pre-set questions in fixed order, verbalised

  • same as questionair but fewer respondants

Unstructured interview: no set formula, just general topic, questions develop based on responses

  • flexible, social-desirability bias can be reduced by rapport
  • analysis more difficult

Semi-structured interviews: pre-set questions with flexibilty to ask follow ups 

  • advantages of both structured and unstructured
8 of 10



  • useful preliminary tool
  • quick and economic to carry out
  • using secondary data


  • cannot demonstrate cause and effect
  • third variable problem
  • miususe and misinterpretation
9 of 10

Pilot Studies

Checks procedure and materials, making modifications if nessesary

Single blind: ptps arent aware of aim until debrief

Double Blind: neither ptp or researcher knows aim

Control group: used as a comparision

10 of 10


No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Research methods and techniques resources »