Psychology

Experimental and Control Groups Explained



Tweet
Janet Grischy's image for:
"Experimental and Control Groups Explained"
Caption: 
Location: 
Image by: 
©  

Much of scientific research is about accurate, reliable, repeatable comparison. In controlled experiments in fields like psychology and medicine, scientists sometimes separate their subjects into two groups in order to compare them. One group is designated as the control and one as the experimental group.

Next, a variable is introduced into the experimental group, something that treats its members differently than the controls. They might walk a mile each day, or eat a celery stalk at 9 AM, or take a certain medication in a certain way. The other group, the control group, does not make this change.

The fundamental concept is to make two very similar groups, with the inevitable differences among subjects randomized. When differences are randomized, they are as likely to appear in one group as another. Scientists randomize in some ways that seem counterintuitive.

For example, when choosing experimental rats, scientists might randomly reach into the cage and draw out the first rat they grasp. These rats have been bred and raised to be as similar as possible. However, they are naturally not identical. Grabbing them randomly and assigning them to each group in turn seeks to ensure that any differences among the rats will be randomly distributed in the control and experimental groups.

Scientists seeking random groups of people who have or do not have a certain disease might in some cases start at the home of a child with a certain disorder, and then knock on neighborhood doors until they find a similar child who does not have the disease and who can participate in the experiment.

Researchers investigating the use of a new drug might, at one stage, seek patients who have similar characteristics, like being at a certain stage of a disease, or having received certain treatments in the past. This helps assure that any results are more likely to pertain to the drug under study, and not to something else about some of the patients.

So scientists try to make the control and experimental groups differ in only one important way. The experimental group will have its behavior, environment, or chemistry changed somehow. The control group will not. This way, scientists will be able to more clearly see the effect of the change they introduce.

Scientists use additional methods to ensure that their results are not confounded. They often keep the people who are actually running the experiment from knowing which subjects are in the control group and which are in the experimental group. This is to avoid the unconscious biases of the experimenters from affecting the outcome.

For similar reasons, researchers try to keep human subjects from knowing which group they belong to. Expectations can have strong effects on a subject’s reactions. For reasons like these it can be hard to participate in an experiment, even though scientists often compensate subjects amply for their time and travel.

Anyone who acts as a subject is making a praiseworthy contribution to society and broadening human knowledge, whether or not he or she personally benefits. Finally, at the end of the experiment, subjects will almost always get to find out if they were in the experimental group or were one of the controls.


Tweet
More about this author: Janet Grischy

From Around the Web




ARTICLE SOURCES AND CITATIONS
  • InfoBoxCallToAction ActionArrowhttp://www.montana.edu/wwwarc/Micro%20501/ILAR%20practical%20exp%20design.pdf
  • InfoBoxCallToAction ActionArrowhttp://www.socialresearchmethods.net/kb/index.php
  • InfoBoxCallToAction ActionArrowhttp://www.fas.org/ota/reports/8310.pdf