define independent variable
manipulated variable using different conditions (levels of an IV)
define dependent variable
measured, outcome variable
define control variable
variable that researcher holds constant
3 criteria for supporting causal claim
covariance: is X related to Y?
temporal precedence: does X come before Y?
internal validity: are there confounds?
how do experiments establish covariance?
different levels of the IV serve as comparison conditions, allowing us to see any relationships between IV and DV
control, treatment, and placebo groups
how do experiments establish temporal precedence?
cause variable (IV) precedes the effect variable (DV)
- participants complete manipulation of IV
- then, participants complete measures of DV
define design confound
when a second variable varies systematically along with the IV, providing an alternative explanation for the results
internal validity problems to address in experiments
design confounds (systematic variability)
selection effect
define systematic variability
confound. levels of 3rd variable coincide in a consistent pattern with the experimental group membership
define unsystematic variability
levels of variable fluctuate randomly, doesn’t coincide with membership of an experimental group (i.e. natural variability, not a cause for concern)
define selection effect
when participants in one condition are systematically different than participants in other conditions
how to avoid selection effects
random assignment
matched groups: pseudo-random assignment using small-group Latin square
between-subjects design
different groups of participants placed in different conditions (i.e. independent-groups)
within-subjects design
each participant is presented with all conditions of IV (i.e. within-groups)
types of between-subjects designs
posttest only, pretest/posttest
posttest only design
participants only tested on DV once
pretest/posttest design
participants measured before application of IV condition, then after application of IV condition
types of within-subjects design
repeated-measures design, concurrent-measures design
repeated-measure design
participants measured on DV after exposure to one condition, then measured after exposure to another condition
possible problems in within-subjects design
order effects, demand characteristics
types of order effects
practice effects, carryover effects
define practice effects
participants get better at a task (due to practice) or get worse at a task (due to fatigue)
define carryover effect
contamination from initial condition carries over to the next condition
ex) taste of chocolate lingering during the next measure
how to avoid order effects
counterbalancing: presenting different levels of IV to participants in different orders
