4.2.3. - Research Methods Flashcards
Define lab experiment?
Experiment taking place in controlled environment, within which researcher manipulates IV and records effects on DV
Participants go to researcher
Name 2 benefits of lab experiments
- EVs and CVs can be controlled (high internal validity)
- Repeatable (standardised) (enhances validity)
Name 2 problems with lab experiments
- May lack generalisability (low external validity due to low mundane realism)
- Demand characteristics (less natural, more cues)
Define generalisation?
Extent to which findings can be broadly applied to population
Define mundane realism?
Extent to which experiment mirrors the real world
Define field experiment?
Experiment that takes place in natural, everyday setting, within which researcher manipulated IV to measure DV
Researcher goes to participants
Name 2 benefits of field experiment?
- Natural environment means high external validity (higher mundane realism)
- Less demand characteristics (unaware of being studied)
Name 2 problems with field experiments?
- Harder to control EVs and CVs (lower internal validity)
- Ethical issue of no informed consent if not aware of study
Define natural experiment?
Experiment where the IV has been manipulated but not by the researcher and the DV is naturally occurring or measured
Name 2 benefits of natural experiment?
- May be only ethical option for that research (would be unethical to manipulate IV in that way)
- High external validity (real life issues)
Name 2 problems with natural experiments?
- Situation may arise rarely (can limit comparisons to similar experiments, therefore lack generalisability)
- Participants not randomly allocated (can have differences in groups)
Define quasi experiment?
Study (not technically experiment) where IV ‘exists’ and is not manipulated by anyone, DV may be naturally occurring or measured
Why is a quasi experiment not technically an experiment?
No manipulation of IV
Name 3 benefits of quasi experiments?
(Share with lab experiments because often under controlled conditions)
- EVs and CVs controlled (high internal validity)
- Repeatable
- Comparisons can be made between different typed of people
Name 2 problems with quasi experiments?
- Participants not randomly allocated (can be differences in groups)
- Causal relationships not demonstrated (no IV manipulation so no certainty IV is reason for DV change)
Define general population?
Everyone in an area
Define population/ target population?
Group of people who are the focus of the researcher’s interest
Define sample?
Smaller group of people drawn from target population for the investigation.
Thought to be representative of target population.
Define bias (in context of sampling)?
When certain groups are under/ over represented in sample
Name the 5 sampling techniques?
- Random sample
- Systematic sample
- Stratified sample
- Opportunity sample
- Volunteer sample
Define random sample?
All members of a target population have equal chance of being selected
How is a random sample made?
1) Complete named list of target population
2) Use lottery method to randomly draw members of population
Name a benefit of random sample?
No researcher bias
Name a problem with random sample?
Not guaranteed to be representative (hard to generalise)
Define systematic sample?
Participants selected using set ‘pattern’
How is systematic sample made?
1) Sampling frame produced (organised list of target population)
2) Every nth person selected
Benefit of systematic sample?
Fairly representative and unbiased
Problem with systematic sample?
Takes time and effort (using complete target population list)
Define stratified sample?
Composition of sample reflects proportion of people in certain subgroups (strata) within general population
How is stratified sample made?
1) Appropriate strata (sub groups) identified
2) Proportions of people in strata in population must match proportion in strata in sample
Benefit of stratified sample?
Representative sample (generalisation possible)
Problem with stratified sample?
Stratification is not perfectly representative - doesn’t show all ways people are different
Define opportunity sample?
Participants are the people most available (e.g. easiest to obtain)
How is opportunity sample made?
Researcher goes to location and asks people nearby to participate
Benefit of opportunity sample?
Cheap, quick method
Problem with opportunity sample?
Unrepresentative of whole target population (only certain area) so hard to generalise
Define volunteer sample?
Self-selection by participants
How is volunteer sample made?
Ads
Benefit of volunteer sample?
Participants are willing with minimal researcher input (so are more likely to engage)
Problem with volunteer sample?
Unrepresentative as can attract people with certain traits (E.g. curious) so hard to generalise
Define ethical issues?
Issues that arise due to the conflict between the rights of participants in studies and the goals of the research to produce valid, authentic data
What is the BPS code of ethics?
Quasi-legal document instructing researchers in the UK about what is acceptable when dealing with participants
What is the goal of the BPS code of ethics?
To protect participants based on principles:
1) Respect
2) Competence
3) Responsibility
4) Integrity
How are ethics assessed?
Ethics committees weigh up costs and benefits before approving research
What are the 4 main ethical issues?
1) Informed consent
2) Deception
3) Protection from harm
4) Privacy and confidentiality
Define informed consent?
Participants should be made aware of: aim, procedures, right to withdraw and right to control data use so they can make an informed judgement on whether to take part
Why is informed consent criticised by researchers?
Demand characteristics can cause unnatural behaviour
How is informed consent dealt with?
- Consent form
- Alternative form of consent:
1) Presumptive - ask similar group
2) Prior general - consent to many studies including one with deception
3) Retrospective - consent after study
Define deception?
Deliberately misleading or withholding info so consent is not informed
How is deception dealt with?
Debrief after research ( told aim and anything withheld, including what their data will be used for and their right to withhold it)
Define protection from harm?
Participants should be at no more risk to physical/ psychological harm than in their daily lives
How is protection from harm dealt with?
- Right to withdraw
- Reassured behaviour is normal in debrief
- Counselling offer
Define privacy?
Participant’s right to control info about themselves
Define confidentiality?
Participant’s right to have any personal data they expose protected by the researcher under the ‘Data Protection Act’
How are privacy and confidentiality dealt with?
- Reminded of right to withhold data
- Anonymity in data recorded
Define pilot study?
Small-scale version does an investigation that takes place before the real investigation is conducted
How is a pilot study different to a regular study?
Uses less participants
What are the aims of a pilot study?
Check procedures and materials work so that researchers have time to identify and modify any design faults before large scale research is completed.
Name 2 benefits of pilot study?
- Removes research faults
- Saves money and time
When would you use pilot studies?
- Experiments
- Self reports (e.g. questionnaire questions checked)
- Observational studies (e.g. observers can be trained)
Define single-blind trial?
Participants not told the aim of the research
Benefit of single-blind trial?
Reduces demand characteristics
Problem with single-blind and double-blind trials?
Ethical issue of deceiving participants and not getting informed consent
Define double-blind trial?
Participant and investigator not told aim of research
Benefit of double-blind trial?
Reduces demand characteristics and investigator effects
Give example of when double-blind procedures are used?
Drug trials
Neither participant nor investigator knows if it is drug or placebo
Give example of single-blind measure procedure?
Most experiments (e.g. observing effects of fizzy drink/ water on sport performance)
Define experimental condition?
Group containing the manipulation of the independent variable
Define control condition?
Group not containing the manipulation of the independent variable
Why is control group/ condition used?
Acts as a baseline for comparison to confirm change in DV is due to IV manipulation
