Treatment Groups

Experimental Research

The primary purpose of experimental research is to establish a causal connection between an independent variable (IV), the cause, and a dependent variable (DV), the effect. (This stands in contrast to survey research, which typically measures correlational relationships between variables.) This means that researchers using treatment groups are interested in demonstrating how changes in DVs are a direct result of—and caused only by—changes in the IV.

[Page 1780]In order to establish causality in experiments, an IV must be manipulable, or able to be varied by the researcher. This is not to say that all IVs must be manipulable; many causes of changes to DVs are due to things that researchers cannot control. For example, a tornado within a small town cannot be controlled by researchers, but can certainly have an effect on the structures of the town or the well-being of its residents. However, in experimental research, researchers must have some logical way of varying the IV in order to measure its effects on the DV in order to claim causality; if they cannot make changes, they cannot claim to have produced the results. While researchers cannot manipulate an individual’s genetic predisposition to a disease, they can experimentally test different levels of a drug in preventing the disease from surfacing. By manipulating the IV, researchers can measure how changes in the DV are related to the manipulations of the IV.

It is important that researchers using experimental design focus on increasing the internal validity of their studies, or the assurance that any changes in the DV are a direct result of changes in the IV and not due to other causes. One protection against threats to internal validity is the use of groups.

The Use of Groups in Experimental Research

Some research examines how changes in an IV affect one group of individuals. Imagine if a class of students is given a questionnaire at the beginning of a semester to assess their fear of public speaking, and then administered the same questionnaire at the end of the semester. Are any changes in the level of fear (the DV) due to the curriculum of the class (the IV)? While that is one possible explanation, there may be several others: the students may have matured over time, they may have spent time out of class working on their public speaking skills, or they may have taken a new internship which required them to engage in more public speaking, thus lessening their fear. Researchers using only one group know what change occurred across the span of the semester, but they do not necessarily know the cause of that change. The internal validity of the study is low, because the researchers cannot support their claim that the class curriculum was the cause of decreased levels of fear. An improvement to this design would be to find a way to compare the results of those who took the class to the results of similar students who had not.

...