This paper examines experimental research design as a methodological approach, defining its core characteristics and explaining how controlled variables and environmental management enhance internal validity. The paper traces the advantages of experimental design in laboratory settings while analyzing why true experimental methods are impractical in complex fields such as political science and global studies. Through discussion of confounding factors, inherent limitations, and real-world constraints, the paper argues that many real-world research contexts require quasi-experimental or non-experimental designs due to the impossibility of controlling all relevant independent variables.
Experimental designs are often considered preferable compared to non-experimental approaches. This preference reflects the underlying practices and controls associated with different research methods. To understand this distinction, it is first necessary to define experimental research and identify where and when it may be applied, then consider alternatives.
Experimental design is a research process where the independent and dependent variables can all be identified, and the researcher designs a research process that will control the independent variables, as well as the environment (which itself is usually a variable) and the units of analysis or measurement. For example, if a researcher wants to assess whether sunlight has an impact on the growth of a plant, they may create an experimental design where two identical plants are kept in identical conditions with the expectation of exposure to sunlight (or simulated sunlight). The researcher may then control for the number of hours the plant may be exposed to sunlight. The process is one where there is control of the independent variable (the sunlight), and the conditions of the study eliminate the potential for other independent variables from having an influence. This is a true experimental design, and the outcome (the dependent variable) observed is the result of the controlled conditions.
To implement an experimental design, it is necessary for there to be the potential to control the independent variables which may impact the outcome. This process of control means that there is a much higher level of internal validity, as it is able to control different influences and significantly reduce the potential for confounding factors to be present. Confounding factors are influences that may impact the outcome of the experiment but are outside of the experiment, including unrecognized influences. If a study has a high level of internal validity, there is a lower chance that the outcome can be explained by confounding factors other than the variables being studied.
Non-experimental research designs are likely to be used in contexts where there are many different influencing factors present or where it is not possible to control all of the independent variables. In such situations, the researcher must rely on alternative methodologies that account for variables that cannot be isolated or manipulated.
The problem with experimental design is that it is not always possible to control all the potential influencing variables, so it is not a practical approach to all issues. Research undertaken in a laboratory may be well suited to experimental design, but in areas such as social and political research, there are many influences which may not be controlled. Even in research where there is an apparently experimental design, such as Milgram's experiments where obedience to authority was assessed, the outcome is impacted not by the experimental conditions, but by issues that remain unseen, such as the attitudes, views, and experiences that the test subject brings into the experiment with them.
"Why political science resists true experimental methodology"
You’re 68% through this paper. Sign up to read the remaining 1 section.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.