The goal of experimental research is to explain effects and determine a causal relation between two variables. Experimental researchers attempt to answer a research question that asks what effects one variable has on another variable. For example, experimental research may be used to answer the questions "Does this reading intervention increase fluency?" or "Does this intervention increase on-task behavior?"
Experimental Research: What It Is
Experimental research involves active manipulation of the independent variable, or intervention. A population sample must be chosen at random, and participants are randomly assigned to different groups. An intervention or treatment is implemented, and the effects on the participants, the dependent variable, are recorded. The researcher attempts to determine whether the intervention does in fact affect that population; if it does, these changes may be analyzed to determine how and why these effects occurred. Experimental research is used to study anything that can be counted or measured, such as behaviors, scores or times.
Examples of Experimental Research
The goal of experimental research is to answer a research question. To answer this question using a fundamental research design, researchers randomly assign participants to at least two different groups: an experimental and a control group. Participants are tested before any interventions are introduced to provide a baseline from which effects are measured. The experimental group then receives the intervention while the control group receives a placebo. While effects may be measured in a variety of ways as the experiment progresses, both groups also receive a post-test to measure change pre-test results.
There are a variety of experimental research designs. For example, in a between subjects design, two different groups of participants receive the intervention, and the effects are compared between the two. In a within subject design, the effects are measured as participants take part in two different conditions, one with the intervention and one without. In all experimental studies, the researcher manipulates the intervention, and the effects are measured and analyzed.
When to Use It
Experimental designs are intended to determine causation -- whether or not the independent variable causes the dependent variable. Experimental designs should be used when the cause of an effect can be manipulated. For example, an experimental design would be ideal for studying the effects of a specific reading intervention on the fluency rate of a specific grade or grouping of students. In this case, the effects can be operationally defined, measured and counted. If these measurements can be counted, researchers can statistically analyzed them and determine a relation.
When Not to Use It
Experimental design would not be ideal in situations where the researcher was attempting to more fully understand a culture, practice, group, emotions or beliefs. The research question would be broad, without situational constraints. For example, a research question answered by a descriptive design may be, "How does a student with autism respond to an inclusion environment after transitioning to middle school?" Individual case studies are an example of a nonexperimental, descriptive design in which the subject is observed over an extended period of time in an attempt to gain a better overall understanding of a particular focus.