Methods Flashcards
what is external validity and the two types
the extent to which the results are valid outside of the research
ecological validity - how well results can be generalised to real life setting
population validity - how well results can generalise from the sample to the wider population
what is triangulation
when 2+ experiments point to the same results
the strengths of one study make up for the faults of another
3 key parts to designing a study
- aim
- hypothesis (explain the IV and DV)
- method and how to overcome problems
what are demand characteristics?
Any cue from the researcher that may be interpreted by the ppts as revealing the purpose of the investigation. This often leads to the ppt changing their behaviour for two reasons:
screw you effect - ppt underperforms to sabotage results
please you effect - ppt overperforms to please researcher
types of observation (6) and their pros and cons
Naturalistic - experiment takes place in a natural environment.
pro: high ecological validity - can be applied to real life situations
con: lower internal validity - confounding variables not controlled, experiments not replicable
Controlled - experiment takes place in a controlled environment
pro: high internal validity - controls confounding variables, experiment is replicable
con: low mundane realism
Covert - ppts are unaware of being in an experiment
pro: no change in ppts behaviour
con: could be unethical
Overt: ppts are aware and have consented to being part of the experiment
pro: ethical
con: ppt may change their behaviour
Participant: researcher becomes a part of the group in the experiment
pro: first hand experience
con: blurred lines between ppt and researcher
becomes subjective (going ‘native’)
Non-participant: researcher stays removed from the experiment
pro: researcher has a clear role
remains objective
con: researcher doesn’t get first hand experience
what is single blind and double blind, and what do each control for?
single blind: where the ppt is not aware of the research aims. controls for demand characteristics because the ppt cannot change their behaviour according to the aim.
double blind: neither the ppt or the researcher are aware of the research aims. controls for the investigator effect - prevents leading questions or biased behaviour that could have an affect on the ppt
what are investigator effects?
any effect of the investigator’s behaviour (conscious or unconscious) on the outcome of the DV.
what is internal validity
degree to which the study measures what it is supposed to including:
- how much ppts believe the experimental set up
- how well conclusions can be said about the IVs affect on the DV
4 types of experiment and their pros and cons
lab experiment: - controlled experiment - IV is manipulated by researcher - record IV's affect on DV - extraneous variables controlled pros: -high internal validity - can be replicated cons: - low external validity - demand characteristics
field experiment: - natural setting - IV is manipulated by researcher - record the effect of the IV on the DV pros: - high external validity - no demand characteristics cons: -confounding variables are not controlled -cannot be replicated - covert experiments raise ethical issues
natural experiment: - IV changes without researchers input - record the effect of IV on DV - setting is not necessarily natural pros: -high external validity cons: -opportunities for research rare - participants are not randomly allocated -reliability is reduced
quasi experiment - IV is not determined - 'variables' naturally exist pros: -high internal validity -replication possible cons: -ppts not randomly allocated -variables not controlled
4 parts of the ethical guidelines
informed consent: ppts should be aware of the aims of the research, the procedures and their rights. this is not necessary if ppts are being observed in a public place. can be dealt with by:
- presumptive consent
- prior general consent
- retrospective consent
deception: you cannot deliberately mislead or withhold information from ppts
protection from harm: ppts should not be at anymore risk than daily life and should be protected from physical and psychological harm
privacy and confidentiality: ppts have the right to control information about themselves and information they give should be confidential.
what is there to consider in cost/benefit analysis
costs to ppts and the reputation of psychology, vs benefits of knowledge gained
two types of hypothesis
directional hypothesis
- 1 tailed
- predicts the direction of an effort
- eg group A will score more than group B
non-directional hypothesis
- 2 tailed
- predicts an effort but no direction
- eg there will be a difference between group A and group B’s scores
what are the 4 types of variables?
independent variable - the one that is manipulated
dependent variable - the one that is measured
extraneous variable - any variable that should be controlled
confounding variable - affects the DV
4 experimental designs
independent groups - ppts split into groups which each take part in a different condition
problem: individual differences may affect results
solution: - random allocation
- stratified sampling
- increase group size
repeated measures - all ppts take part in all conditions
problem: order effects due to demand characteristics and improved / reduced performance
solution: counterbalancing
randomisation - mix series’ of tasks so ppt is always switching between conditions. this reduces demand characteristics
matched pairs - ppts are matched according to variables that may affect the DV, ie IQ, age, gender. the matched pair splits between conditions
5 sampling techniques
random sample: all members of population have an equal chance of being picked
systematic sample: organise population into an order. pick every nth person (predecided)
stratified sample: the composition of the sample reflects the proportions of the population
opportunity sample: researcher asks the people nearby
volunteer sample: ppts self select themselves after seeing the experiment advertised
what is aquiescence response bias
when people score a certain way on a questionnaire due to the tendency to agree with everything
what is metaanalysis
combining results from a number of studies to get an overall view of a topic
3 types of correlation
positive - as one co variable increases, so does the other
negative - as one covariable increases, the other decreases
no correlation - there is no relationship between covariables
strengths and weakness of correlations
strengths:
- can determine whether there’s a relationship between covariables
- cheap and not time consuming
- forms basis to design an experiment
weakness:
- no IV is manipulated so you cannot draw casual conclusions
- doesn’t show a possible third variable - can be misleading
- only measure linear relationships
closed questions vs open questions
Closed
- have a fixed range of responses
- leads to quantitive data
Open
- response written however ppt wants
- qualitative data
3 types of scale (questionnaires)
likert scale - symmetrical scale
strongly agree –> strongly disagree
rating scale - rates how strong peoples feelings are
very entertaining –> not entertaining at all
fixed choice - no sense of scale tick one: entertaining funny boring sad
bad techniques to use in a questionnaire
- overuse of Jargon
- technical terminology
- emotive language
- leading questions
- double barrelled / double negatives
3 types of interview and definitions
structured interview - predetermined set of questions in a fixed order. interviewee doesn’t elaborate
semi structured - pre set questions but interviewer is free to ask follow up qs. interviewee can elaborate
unstructured - no set qs so interaction is free flowing. interviewee encouraged to elaborate
cherry picking problem
file drawer problem definitions
CP - picking only the results that fit the hypothesis
FD - more significant results are more likely to be published than non significant results
What is a family study
1) take an index case with the illness / disorder etc
2) measure concordance rates with family members
3) compare to concordance rates with a control family
- control index does not have illness / disorder
- this is the base rate
4) expect 1) to be greater than 3) if genetics has an effect
What is the independent variable in this study?
Whether X or y ….
Mention both variables
What is the dependent variable in this study?
State what is measured
Experimental design
Design is different from methods
E.g.
Repeated measures
Matched pairs etc
What should you mention when evaluating a therapy
Effectiveness
- does it work
- for how long
- for what % of people
- how strong is the evidence
Appropriateness
- does it have side effects
- how long does it take
- how much does it cost
- are there risks
- is it suitable for everyone
- are there ethical issues
How to answer scenario / stem questions
- apply knowledge in different contexts
- answer must address stem
Advantages and disadvantages of independent groups
+ no order effects
- individual differences
Repeated measures
+ and -
+ no individual differences
- order effects (Practise / fatigue )
+ and - of matched pairs
+ no order effects
+ few individual differences
- time consuming
- expensive
Process of clinical trials
- type of field expt
- way of testing whether a treatment works
- random allocation
- control group with placebo
- blind procedures
What is peer review
- other researchers assess quality of research before publishing
- usually anonymous but may not be to prevent criticising rival researchers
- may slow down rate of change
What is a pilot study
A small scale version of the study to test whether the procedure works