PSYCH RESEARCH METHODS Flashcards
(30 cards)
controlled variable
anything you keep the same
example:
- Participant differences
- Noise levels in the experiment
- Caffeine consumption
extranous variable
anything that is not the IV but may affect the DV (reduces internal validity)
confounding variable
variable other than the IV that has affected the results (either an EV that was not controlled or could not be controlled) (reduces internal validity)
opertaionalising variables
refers to specifying exactly how the variables will be manipulated or measured in a particular controlled experiment.
sample representation
sample population would be representative of the differences presented within the target population.
convenience sampling
readily available
- volunteer
convenience sampling - pros
- quick
- cheap
convenience sampling - cons
- biased, the sample might not represent the whole population.
- not random, increases the chance of inaccurate results.
random sampling
- every member of the population has the same chance of being selected
random sampling - pros
- sample generated can be more rep than conveince sampling.
- reduces experimenter bias in selecting partipcants
random sampling - cons
- time consuming, to ensure every participant has an equal chance of being selected.
- may not create a rep sample when sample is small.
stratified sampling
selecting people from the population in a way that ensures that its (subgroups) are fairly represented in the sample
stratified sampling - pros
- produce rep sample
stratified sampling - cons
- time consuming and expensive
within subjects
participants complete every experimental condition.
within subjects - pros
- no participant variables (more likely that results are due to manipulation of IV and not any EVs)
- less people required
within subjects - cons
- can produce order effects;
completing one experimental condition may influence how participants perform in the latter condition/s (e.g. due to fatigue, practice, participants’ expectations) - participant dropping out of this experiment = greater impact as the experimenter loses two data points instead of one.
prevent control for order effects (within subjects)
counterbalancing - changing the order of conditions for different participants to cancel out order effects.
between subjects
participants are divided into different groups and complete only one experimental condition.
between subjects - pros
- less time consuming than within-subjects design
- does not create order effects.
between subjects - cons
- may require more participants than a within -subjects design.
- participant differences
mixed design
combination of elements from within- & between-subjects.
mixed design - pros
- can compare results across experimental conditions AND individuals
- multiple experimental conditions can be compared to a baseline control group.
mixed design - cons
- costly and time consuming