Ch. 9 - Independent Groups Flashcards
randomized experiments
the golden standard design.
the only difference between groups is the variable we THINK causes the outcome of another. (2 total).
correlational research
is there an association between variables? NOT causation!
positive: same direction.
negative: opposing directions.
zero: no correlation.
formal (random) experiments
establish causality!
independent groups design
participants DO NOT OVERLAP between groups (stick to one)
experimental vs controlled conditions
experimental: intervention or treatment (INDEPENDENT/MANIPULATED)
controlled: absence of any treatment
dependent variables
measured by the researcher
random assignment
each participant has 100% equal chance of being assigned to a group
how many participants are needed?
depends on:
- your expected effect size
- power (probability of detecting an effect)
** under-powered studies may not replicate
representative sample
characteristics that are representative of your population
important to make sure results can be GENERALIZED.
confounding variables
3rd effect that causes an effect
could go undetected
DECREASE internal validity
can be:
- environmental
- individuals based
- researcher based
matched groups design
matched on some variable of importance like age/gender
ensures groups are equal to make causal claims
INCREASE internal validity
internal vs. external validity
internal: experimental control (is the IV the actual cause?)
external: generalizability (can it be applied?)
