Quasi-experimental research design

Quasi-experimental Research Designs

A quasi-experimental research design meets 2 out of 3 criteria for causality: empirical association and time order; when random assignment is not possible, a quasi experimental design may be used as it is the next-best choice to the experimental design

In most cases, subjects are not randomly assigned because you may not be able to do so; for example, in studying child abuse and subsequent delinquency, you cannot randomly assign certain children to receive abuse

Three types: nonequivalent-groups design, time-series designs, and ex post facto control group designs

Nonequivalent-groups design

- involves the experimental group and the comparison group (the comparison group is "matched" to characteristics of the experimental group except for of course the hypothesized cause/ the independent variable); because we are not using randomization, we cannot say that the two groups are equivalent, hence the name nonequivalent-groups design; as a general rule, the groups should be comparable as far as variables that may have an effect of the dependent variable

Time series design

- allows us to measure months, weeks, years, etc