preexperimental design

Definition
Preexperimental design: A weak research design involving a treatment but no control features (e.g., a one-group pretest-posttest design).

Notes
As suggested by its name, this design involves merely pretesting a selected group of participants, administering a treatment, then posttesting the group to determine whether a shift in scores occurred from pretest to posttest. Notice that this design has virtually no control for biasing influences in the form of threats to internal validity, such as those described in Chapter 7 (extraneous events, etc.). (Suter 2012, Ch.10, pg. 11)

Let's consider one more weak preexperimental design to highlight the value of strong randomized true experimental designs. It is often referred to as the static-group comparison design and is shown below:
T Post
C Post
Notice the absence of the symbol R, which you will recall refers to random assignment. That is why the design is referred to as static, meaning “inactive, passive, or intact. ” There is no active movement of subjects into groups, as the random assignment procedure would require. With this design, typically one group is located (already intact) and given some type of treatment, while another similar group is located to function as an intact comparison. (Suter 2012, Ch. 10, pg. 12)

Weak preexperimental designs can be greatly improved by the technique of matching, and as such would be called quasi-experimental research designs, as explained later in this chapter.