Classification of Experimental Design
There are three basic types of experimental research designs. These include pre-experimental designs, true experimental designs, and quasi-experimental designs. The degree to which the researcher assigns subjects to conditions and groups distinguishes the type of experimental design. This module will focus on the different types of true experimental designs. True experimental designs are characterized by the random selection of participants and the random assignment of the participants to groups in the study. The researcher also has complete control over the extraneous variables. Therefore, it can be confidently determined that that effect on the dependent variable is directly due to the manipulation of the independent variable. For these reasons, true experimental designs are often considered the best type of research design. There are several types of true experimental designs and they are as follows:
- Post-test Only Design: This type of design has two randomly assigned groups: an experimental group and a control group. Neither group is pretested before the implementation of the treatment. The treatment is applied to the experimental group and the post-test is carried out on both groups to assess the effect of the treatment or manipulation. This type of design is common when it is not possible to pretest the subjects.
- Pretest-Post-test Only Design: The subjects are again randomly assigned to either the experimental or the control group. Both groups are pretested for the independent variable. The experimental group receives the treatment and both groups are post-tested to examine the effects of manipulating the independent variable on the dependent variable.
- Solomon Four Group Design: Subjects are randomly assigned into one of four groups. There are two experimental groups and two control groups. Only two groups are pretested. One pretested group and one unprotested group receive the treatment. All four groups will receive the post-test. The effects of the dependent variable originally observed are then compared to the effects of the independent variable on the dependent variable as seen in the post-test results. This method is really a combination of the previous two methods and is used to eliminate potential sources of error.
- Factorial Design: The researcher manipulates two or more independent variables (factors) simultaneously to observe their effects on the dependent variable. This design allows for the testing of two or more hypotheses in a single project. One example would be a researcher who wanted to test two different protocols for burn wounds with the frequency of the care being administered in 2, 4, and 6 hour increments.
- Randomized Block Design: This design is used when there are inherent differences between subjects and possible differences in experimental conditions. If there are a large number of experimental groups, the randomized block design may be used to bring some homogeneity to each group. For example, if a researcher wanted to examine the effects of three different kinds of cough medications on children ages 2-16, the research may want to create age groups (blocks) for the children, realizing that the effects of the medication may depend on age. This is a simple method for reducing the variability among treatment groups.
- Crossover Design (also known as Repeat Measures Design): Subjects in this design are exposed to more than one treatment and the subjects are randomly assigned to different orders of the treatment. The groups compared have an equal distribution of characteristics and there is a high level of similarity among subjects that are exposed to different conditions. Crossover designs are excellent research tools, however, there is some concern that the response to the second treatment or condition will be influenced by their experience with the first treatment. In this type of design, the subjects serve as their own control groups.
Once the design has been determined, there are four elements of true experimental research that must be considered:
- Manipulation: The researcher will purposefully change or manipulate the independent variable, which is the treatment or condition that will be applied to the experimental groups. It is important to establish clear procedural guidelines for application of the treatment to promote consistency and ensure that the manipulation itself does affect the dependent variable.
- Control: Control is used to prevent the influence of outside factors (extraneous variables) from influencing the outcome of the study. This ensures that outcome is caused by the manipulation of the independent variable. Therefore, a critical piece of experimental design is keeping all other potential variables constant. For example, if testing the effects of fertilizer on plant height, all other factors such as sunlight, soil type and water would have to be constant (controlled).
- Random Assignment: A key feature of true experimental design is the random assignment of subjects into groups. Participants should have an equal chance of being assigned into any group in the experiment. This further ensures that the outcome of the study is due to the manipulation of the independent variable and is not influenced by the composition of the test groups. Subjects can be randomly assigned in many ways, some of which are relatively easy, including flipping a coin, drawing names, using a random table, or utilizing a computer assisted random sequencing.
- Random selection: In addition to randomly assigning the test subjects in groups, it is also important to randomly select the test subjects from a larger target audience. For example, if a researcher wanted to look at the impact of sleep on the test scores of 5th graders in a particular city, a sample of 5th graders would need to be randomly selected from the city’s population in such a way that any 5th grader would have an equal chance of being selected for the study. This ensures that the sample population provides an accurate cross-sectional representation of the larger population including different socioeconomic backgrounds, races, intelligence levels, and so forth.