Measuring and Manipulating Variables Flashcards
variables
“environmental” factors that can have at least two levels
what are 4 types of independent variables?
physiological
experience
stimulus/environmental
participant/subject
physiological
IV: manipulation of a participant’s normal biological state
experience
IV: manipulation in the amount or type of training or learning that has long term effects on the person
stimulus/environmental
IV: manipulation of some aspect of the environment, short term change that lasts only as long as the study
participant/subject
IV: aspects of the participant treated as if they are IV’s
to select your dependent variable, consider:
- should relate back to your hypothesis
- be operationally definable
- be both valid (relates to measure of interest) and reliable (consistent measurement)
what are 4 types of dependent variables?
correctness
rate/frequency
degree/amount
latency/duration
correctness
DV: how many were right?
rate/frequency
DV: how often did it occur within a certain time interval
degree/amount
DV: how much of it was there?
latency/duration
DV: how fast? how long did it last?
types of extraneous variables:
nuisance variables
nuisance variables
unwanted variables that vary randomly across participants and increases the variability of scores within groups. affects all groups (thus not a confound).
allows us to generalize our results, harder to determine effect but conclusion will still be valid.
confounding variables
unintended variables that create a systematic difference between the groups on the DV, renders results meaningless
what can one do to ‘control’ against extraneous variables?
randomization, elimination, constancy, balancing
randomization
equal chance of being assigned to any group in the experiment, makes sure there are no pre-existing differences between groups, needs a large sample size, removes confounds but has no effect on nuisance variables
elimination
specific extraneous variables completely removed from the experiment.
-you must know what the potential extraneous variables are, not always easy in practice (ex. time)
constancy
when it is impossible to completely remove an extraneous variable, a researcher may try to minimize its effects by having it remain constant for all participants
-often refers to testing conditions
balancing
when it is impossible to completely remove an extraneous variable, a researcher may try to minimize its effects by distributing it to all groups equally
-basically, all EXPERIMENTERS test all CONDITIONS equally
what are two effects that may result from balancing?
(sequence of order effects and carry-over effects may result)
order effects
when position in a series affects how participants respond, doesn’t depend on the event but the position
carryover effects
when the effects of one event influence responses to the next event, specifically depends on the event previous (not position)
how can one reduce the ‘side effects’ of balancing?
counter-balancing (varying the order in which items are present)
-i.e. within-subject/within-group and complete/incomplete
within-subject
every subject tests all orders and conditions
ex. (AB) and (BA)
within-group
each participant tests one order
ex. (AB) or (BA)
complete
all conditions are tested in one subject (A and B both tested)
incomplete
random testing of a single portion of the condition (A or B tested)
what are the requisites for complete counterbalancing?
each event must be presented an equal number of times to other events, and be tested in a position/order of a sequence an equal number of times and precede some other event an equal number of times
