Research Methods Flashcards
(82 cards)
Outline 4 experimental methods
Lab: controlled environment, IV manipulated, DV measured
Field: real world setting, IV manipulated, DV measured
Natural: IV not manipulated by researcher, DV measured in real life setting
Quasi: IV is naturally occurring and not manipulated
evaluate lab exp
+ high control and replicable
- low ecological validity
- demand characteristics
evaluate field exp
+ high ecological validity
- less control
- ethical issues e.g. consent
- hard to replicate
evaluate natural exp
+ can study variables that cant be manipulated for ehtical reasons
- no control over Iv
- hard to replicate
evaluate quasi exp
+ allows comparisons between pre-existing groups
- p variables may confound resutls
outline types of experimental design
independant groups: diff P in each conditions
repeated measures: same P in all conditions
matched pairs: P matched on key variables and then split into groups
evaluate independant groups exp design
+ no order effects
- P variables effect results
eval repeated measures exp design
+ control P variables as its the same people
- order effects
- demand characteristics more likely
eval matched pairs exp design
+ less P variability than independant design
- time consuming
- difficult to match people perfectly
how can you control variables in independant groups and repeated measures
IG: random allocation
RM: counterbalancing to mange order effects
define operationalisation + eval
defining variables in measurable, testable terms
e.g. memory as num of words recalled
+ essential for reliability and validity > if variables arent clearly defined its hard to replicate
define extraneous variables and eval
unwanted variables that could effect DV and vary systematically with the IV
- can reduce internal valdity (IV causing change in DV)
define confounding variables
extraneous variables that DO effect the DV and vary systematically with the IV
define demand characteristics
P guess the aim of study and change behaviour
causing reduction in the internal validity
define investigator effects
researcher unconsciously influences P
define standardisation + eval
ensuring all P treated the same so reliabiltiy and controls are increased
define randomisation
using change to redcue bias and order effect (RM)
O and E random sampling
every member of population has equal chances of being selected
+ unbiased, more representative
- could be unrepresentative by chance
- time consuming
O and E systematic sampling
every nth person is selected from a list
+ reduces researcher bias
+ straight forward
- may not be truly unbiased depending on the list
OandE stratified sampling
population divided into subgroups and P selected randomly in proportion
+ more representative
- requires prior knowledge of population structure
OandE opportunity sample
uses P who are readily available
+ quick and convenient
- biased
- not representative
- lacks generalisability
OandE volunteer sample
P self select by responding to advert or request
+ easy and ethical
- volunteer bias (more helpful/motivated people take part)
whats informed consent and how do you deal with it
P must be told what the study involves and agree to take part
- can use consent forms or presumptive consent (asking a similar group if they would consent)
whats deception and how do you deal with it
P must not be misled unless justified; must be debriefed
- debrief after + right to withdraw