External Validity in Research

Written by Saira Naeem

There are two main types of validity internal validity and external validity.

While internal validity is concerned with how well the research is structured, its procedural processes (research design and methods) and how it is conducted so that a cause and effect relationship can be upheld.

External validity refers to the generalizability of the research findings or conclusions.

External validity thus is concerned with the correctness of the research findings by examining their applicability from one setting to another.

In other words, through external validity, the researcher wants to warrant the fact that the conclusions drawn hold true for other people, at other places, at other times as well……not however for a certainty but acting on the principle of proximal similarity or transferability in the least.

Two sides of the same coin but with a few differences

Basically, if you take the time to ponder over the definitions of the two types of validity you would realize that they are like two sides of the same coin.

There are a few key differences however:

• While internal validity is concerned with the research design and methods, external validity is concerned with the generalizability of research findings.
• Internal validity is concerned with exercising control to eliminate possible alternative explanations, external validity is concerned with naturalness.
• Internal validity aims to measure the strength and accuracy of the research methods, external validity focuses on the applicability of research findings in the real world at large.

How external validity can be improved?

External validity is an important characteristic of research. You can use different ways to improve the external validity of your research.

Inclusion-exclusion criteria

Clearly define the population you choose by putting the inclusion-exclusion criteria into practice. The researcher defines in advance what is inclusion criteria and what is exclusion criteria for the participants. This helps the researcher in selecting the right respondents.

Psychological realism

In some instances, if research participants get to know the aim of the study or what is expected of them they tend to behave differently. So make sure to tell them a “cover story” regarding the aim of the study instead.

Replication

Use different samples and different settings to replicate the study again, this way you would get to see whether you get the same results. This improves the external validity as well as the generalizability of the research.

Field experiments

Do your research out of the laboratory in a natural setting. In a qualitative research the experiments are conducted in natural settings.

Reprocessing or calibration

Make use of statistical methods if and when needed to make adjustments for any problems related to external validity, suppose for instance you encounter uneven groups in case of a particular characteristic such as age, you may make use of the reweighting principle to overcome the problem.

Threats to external validity

Just like internal validity, external validity is also threatened by a few factors that the researcher needs to keep in mind, these are;

Selection bias

Things like motivation, special demographics of a population are related to the independent variable. These can result in problems like differences in groups and this threatens not only external but also internal validity of a study.

Pre–test and post–test effects

External validity may be threatened if a cause and effect relationship is related to the effect seen in a study produced by using the pre-test/post-test principle and these added tests are not applied.

Situational factors

Believe it or not, critics may come around and argue that the findings of a particular research are not applicable or generalizable to other people, places or settings only because of certain situational factors such as researcher characteristics, time, noise or location etc.

Sample features

Sometimes sample features or characteristics also hamper the generalizability of the research findings.

References

• “Understanding internal and external validity”.https://www.verywellmind.com/internal-and-external-validity-4584479
• “Difference between internal and external validity”. https://keydifferences.com/difference-between-internal-and-external-validity.html
• “External validity”. https://socialresearchmethods.net/kb/external.php

Comments

comments