What are the differences between experiments, quasi-experiments, and observational studies?
Experiments are research conducted under specific and controlled conditions to discover, verify, or illustrate a theory, hypothesis, or relationship. Essential elements of an experiment require that at least one variable is manipulated, and random assignment is used to assign experimental units to different levels or categories of the manipulated variable. In a quasi-experiment, the researcher manipulates at least one variable; however, experimental units are not randomly assigned to different levels or categories of the manipulated variable. In an observational study, the analyst might lack the ability to assign levels of the manipulated variable. In other words, the experimenter might not have the ability to control the levels of the manipulated variable, unlike a quasi-experiment. An observational study, furthermore, might have a large number of non-controlled random variables that are thought to affect the dependent variable.
