true experimental research
Definition
A research design that involves the use of an intervention, plus random assignment of participants to groups. Recommended for testing cause and effect relationships.
Notes:
All three types (research designs) involve an experimental manipulation, but only true experimental designs use the random assignment that greatly facilitates ability to uncover cause-and-effect relationships. (Suter, 2012, Ch. 10, pg. 3)
true experimental design, therefore, is one that incorporates a true independent variable. You will also recall from Chapter 5 that a manipulation refers to the creation of group conditions by the researcher. In its most basic form, a manipulation would include a treatment group and a control group. (Suter. 2012, Ch.10)
true experimental research designs include the following:
randomized posttest control group design
randomized pretest-posttest control group design
randomized matched control group design
randomized factorial design (Suter. 2012, Ch. 10)
True experimental designs, especially those with clever control groups, are well suited to ferret out cause-and-effect relationships because of the power of random assignment of subjects to treatment and control groups. (Suter, 2012, Ch. 10)
True experimental designs, despite their potential, can still yield ambiguous findings due to many sources of bias and confounding, poor instrumentation (measurement), inadequate sampling, and so on. In other words, alternative hypotheses may be still be present, and usually are, within true experimental research. (Suter, 2012, Ch. 10, pg. 7)
Considerations:
- downsides of this design
- is it bias to prefer this design?
True experimental designs in education combine an independent variable (a manipulation) with random assignment. They are well suited for uncovering cause-andeffect relationships when a research question focuses on testing treatment or intervention effects. Many specific designs exist, yet they share a defining characteristic: a true independent variable. (Suter, 2012, Ch. 10, pg. 12)
True experimental designs incorporate a manipulation with random assignment to groups (e.g., the randomized posttest control group design). The strength of the true experimental design is highlighted by its contrast to “preexperimental design” that have very little control over threatening influences. True experimental factorial designs are especially useful, for they yield information about main (overall) effects as well as interactive (nonadditive, “compounding”) effects.