Prospective study
observational study in which subjects are followed to observe future outcomes
Retrospective study
observational study in which subjects are selected and then their previous conditions or behaviors are determined
Experiment
uses random assignment to assign experimental units to treatments
Observational study
a study based on observation, where no treatments are imposed, can be either prospective or retrospective
random assignment
assigning experimental units to treatment groups at random
principles of experimental design
control, randomization, replication, and blocking (optional)
statistically significant
when the response is so different between the treatment groups that it was unlikely to have occurred by chance
confounding
in a poorly designed experiment, confounding occurs when it is impossible to separate out the effects of two or more factors
response variable
the data the experimenter keeps track of to compare across treatment groups
level
the specific values that the experimenter chooses for the factor
completely randomized design (CRD)
an experimental design where all experimental units are assigned at random to treatments.
blinding
when either the subjects or the evaluator doesn't know what treatments they received
placebo
a fake drug used in the testing of medication; a tool for blinding
blocked design
an experimental design used to make the treatment groups as similar as possible. first, subjects are grouped by some common characteristic, then randomly assigned to the treatment groups
matched pairs design
an experimental design where each subject or pair gets both treatments (with the order of treatments being randomized)