External validity is raising and answering the question how well the results are useful outside the setting of the research.
The relation with validity
Validity in general is seen as how (in)valid the research has been performed. If it’s well performed it is very useful. Several aspects can be to look at to evaluate a study. Content and construct validity have to do with the correct use of the terms. Predictive validity has to do with how well a test can (post)predict a situation in the future or in the past. Internal validity is raising and answering the question how well the research has been performed. External validity is raising and answering the question how well the results of a research are useful for daily live.
Examples of external validity
For instance, you might wonder if the results of a research on youngsters will give the same results as on elderly people. This was a good question some decades ago when a lot of research was done with students at the university.
Or, if the research about the influence of stress is done with rats that were specially bred for doing laboratory experiments, would this show similar hormonal reactions in human beings?
If you are researching the social interactions of people living in the mountains of Morocco, would you find the same results for the people in the cities of China?
And what to think about this one: will the influence of a medicine on young healthy adults be the same as on the sick people for whom the medicine is made?
All these questions are examples of external validity. As you might have noticed, they all have to do with the generalisability of the outcome of a research. Personally I prefer the term generalisability over external validity, but it’s okay for me to use both terms as equal.
Grouping the questions about external validity
These questions about external validity can be grouped in three dimensions. Would the result be the same if there was a change in time, a change in space or a change in respondents. Now, it is easy to create your own examples. But to help you a little, I will give a few more examples:
- Will humans react the same in a hundred years as they do now? (time)
- Will people react the same in their daily lives as in the experiment in a laboratory? (space)
- Will we get the same results if another sample would have been used? (respondents)
Well this may or may not be the case. But because we do not know, it is good to raise the questions and try to answer them.
For science it is also very well to repeat the research in a more or less equal way. If the results are (about) equal, the theory about this topic becomes more stable and will be more useful for daily life. Therefore, replication of a research is very important and should be highly valued in science.