What Is External Validity?

External validity is one of many types of validity that researchers try to achieve to maximize the accuracy and minimize the shortcomings of their study or experiment. External validity is a term that scientific researchers use to describe how likely it is that the results they have obtained from a sample group would apply to the whole population across various situations and times. Scientists strive to achieve a high amount of external validity for every experiment, because if the results of the experiment do not apply to the population outside of the sample group, then the experiment did not really find any useful results that could be used to predict future outcomes.

An example of a situation where external validity must be evaluated might be in a study done by a college psychology class evaluating the connection between the hours that college students spend working in a job and those students' grades. It may seem like a good idea to use all of the data from the students in the psychology class, or even all of the students involved in the psychology department, to get a quick and easy sample to test. This, however, would harm the external validity of the study, because it assumes several things that may not be true of the general population. For example, do psychology students have the same work and study habits as students in other majors? Additionally, do students at that particular school have the same study habits as students from schools across the country or the world?

Unfortunately, because a large number of experiments take place in a laboratory setting instead of in the subjects' daily lives, external validity can be somewhat hard to achieve. Usually, the researchers conducting the study or experiment would sum up what they believe to be "threats to external validity" in their written report of the experiment in an attempt to explain what may have gone wrong and what can be improved upon in the future to get a higher level of prediction accuracy. For example, if the subjects of a study are told to perform a task under supervision of the researchers, they might behave and perform much differently than they would if they were at home with family and other influences surrounding them. If the study does not take that into account, the external validity is flawed and the results will probably not very accurately predict future outcomes because they were found under special circumstances. The researchers should make note of that and try to improve future studies to minimize the difference.