Skip to main content icon/video/no-internet

External validity refers to the extent to which research findings from one study generalize to or across groups of people, settings, treatments, and time periods. In other words, to what extent does the size or direction of a researched relationship remain stable in other contexts and among different samples? In an effort to measure precise effect sizes and control for confounding variables, many scholars use survey methods featuring hypothetical or retrospective reports, whereas others examine communication phenomena in sterile research labs. In addition, the findings of many social scientific studies are based on the responses of convenience samples consisting of college students and volunteers. While these studies contribute to the growing body of communication scholarship, concerns about their results’ applicability to broader, more diverse populations of people in real-life contexts remain.

In recent years, scholars, practitioners, and editors have expressed a renewed commitment to conducting and supporting rigorous research with high levels of external validity. For example, researchers who examine the effectiveness of communication skills training programs or health interventions want to ensure that the findings of their studies are generalizable to target populations beyond their specific research studies’ samples. Similarly, communication scientists who study message effects aim to produce replicable, generalizable findings that would ring true outside of laboratories and simulated survey scenarios, and researchers seeking external funding must assure grant program officers and foundations that the results of their pilot data are not limited to their immediate research participants. Recognizing the importance of external validity, researchers are attending to the ways they design their studies, collect data, and report their findings. This entry focuses on four central issues to external validity: sampling, ecological validity, replication, and reporting. Each of these issues is explored in this entry.

Key Issues for External Validity

Sampling

The results of a well-designed social scientific communication study should be generalizable beyond its own participants. William R. Shadish, Thomas D. Cook, and Donald T. Campbell explained that researchers may seek to (a) generalize their results to a wider group of people, (b) apply findings to a smaller group of people, or (c) make predictions about a single person or dyad. For example, imagine a researcher tested the efficacy of an intervention program that aimed to significantly reduce young teachers’ communication apprehension. In its inaugural edition, 70 high school teachers from a wealthy school district participated in the study. Half of the sample completed the intervention program, and the other half did not. The results indicated that the communication apprehension of those who participated in the intervention program decreased significantly more than those in the control group. While the findings from the study are promising, are they replicable among young teachers who work in an impoverished school district? Will the intervention program be effective for older teachers? When a researcher begins to design a communication study, important methodological decisions must be made about the project’s desired sample and sampling techniques, since they will affect its external validity.

While good in theory, it is very difficult to collect data from every single person in a population. Due to limited resources, time, and availability, most researchers opt to collect data strategically from a sample, which is a subset of a larger population. For example, if a communication scholar wanted to study the effectiveness of a new government policy on a local community of 15,000 people, it would be unrealistic to expect the scholar to obtain data from all 15,000 people.

...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles

Sage Recommends

We found other relevant content for you on other Sage platforms.

Loading