Upgrade to remove ads
Only $35.99/year

AP Statistics Experimental Design

Terms in this set (22)

observational study
studies in which researchers don't assign choices, they observe them; observing "in the wild", recording the choices they made and the outcome
retrospective study
study where researchers first identified subjects
prospective study
identifying subjects in advance and collecting data as events unfold
Experiment
study design to prove a cause and effect relationship
random assignment
an experiment requires a random assignment of subjects to treatments
Subjects
humans who are experimented on
Participants
humans who are experimented on
experimental units
other individuals who are experimented on (rats, days, petri dishes of bacteria)
Factors
explanatory variable that an experimenter must identify to manipulate
levels
specific values that the experimenter chooses for a factor
Treatment
combination of specific levels from all factors that an experimental unit receives is known as its treatment
Block
grouping of similar individuals together and then randomization within each of the blocks to remove variability due to difference among the blocks (done because certain attributes of experimental units affect the outcomes of an experiment)
completely randomized experiment
ideal simple design, simplest kind of experiment
statistically significant
when the differences observed are bigger than from randomization; because they're bigger, we can attribute the differences to the treatments and say the differences are statistically significant
Control
baseline measurement, used to compare; level of factor to compare treatment results to a situation in which "nothing happens"
control group
experimental units to whom it is applied
Single-blind
when all individuals in either one of the classes is blinded
Double-blind
when everyone in both classes is blinded, we call the experimenter double-blind
Placebo
"fake" treatment that looks just like the treatments being tested is placebo
placebo effect
highlights both importance of effective blinding and importance of comparing treatments with a control
Matching
subjects are paired because they are similar in ways not under the study, used to reduce variation like blocking
confounding
levels of one factor are associated with the levels of another factor
Sets with similar terms