Pre-experimental Designs

Weakly controlled, exploratory designs

Pre-experimental designs encompass three basic configurations: the one-shot case study, the one-group pretest–posttest design, and the static-group comparison. Because they lack randomization and adequate control groups, many threats to internal validity remain uncontrolled. Although straightforward to conduct and sometimes useful for piloting ideas, causal conclusions drawn from these designs are methodologically unsafe.

Defining the Concept

Pre-experimental designs represent the most basic and least controlled forms of experimental research. Campbell and Stanley (1963) classified these designs as precursors to true experiments and highlighted their serious internal validity problems. Core weaknesses include the absence of a control group, lack of random assignment of participants to conditions, and failure to control for potential confounding variables. Consequently, while findings may carry descriptive value, they are insufficient for establishing causal relationships.

Main Design Types and How They Work

Three core pre-experimental designs are distinguished. (1) The one-shot case study applies a treatment to a group and then measures the outcome only once; there is no baseline and no comparison group. (2) The one-group pretest–posttest design measures the same group before and after a treatment; change is observed, but whether that change stems from the treatment or from other factors cannot be determined. (3) The static-group comparison contrasts a treated group with an untreated group; however, the groups are not equivalent because participants were not randomly assigned.

Concrete Example and Application

A university wishes to pilot a newly developed online academic writing course. The researcher delivers the course to a single volunteer group, measures writing scores before and after, and finds that average scores increased. This one-group pretest–posttest application does not rule out rival explanations such as maturation, history, or testing effects. Had the course been run concurrently in a second group with random assignment, the findings would carry considerably stronger validity.

Common Pitfalls and Recommendations for Good Practice

The most common mistake is reporting results from a pre-experimental design using causal language. Stating that "an increase in Y was observed in the group that received intervention X" is more honest than claiming "intervention X improved Y." These designs have a legitimate role in feasibility assessment, instrument development, and as preparation for larger studies. Researchers should explicitly report threats to internal validity, characterize findings as exploratory, and plan transitions to true or quasi-experimental designs whenever possible.

Key terms

Internal Validity: The degree of confidence that the observed change is genuinely caused by the treatment.
Random Assignment: Distributing participants to groups by chance to ensure group equivalence before treatment.
Maturation Effect: Natural changes in participants over time that may be mistaken for a treatment effect.
History Effect: External events occurring during the study that affect the outcome independently of the treatment.
Static-Group Comparison: A pre-experimental design comparing a treated group with an untreated group formed without random assignment.