The validity of a research tool tells us how accurately it measures what it is supposed to measure. There are two main types of validity: internal validity and external validity.¹

The internal validity of a study is important when assessing causation. Internal validity refers to whether the observed effects of a variable under investigation can be attributed to the hypothesized cause. There are many threats to internal validity which are generally related to whether there are other plausible alternatives/explanations for the observed effects. Confounding or extraneous factors, such as a measurement error, may explain the results of the study.

For example, if the scales are not correctly calibrated when measuring the weight of study participants, then recorded weight will be inaccurate (measurement error), and the lack of internal validity from the measurement errors results in the questioning of conclusion drawn from the study. A further example of lack of internal validity would be failing to control for an important confounding variable. To continue the weight example, if the study was examining the possible relationship between weight and diabetes, researchers would need to account for other factors that influence both weight and diabetes development such as physical activity. If a confounder, such as physical activity, is not accounted for in the study, the study has a lack of internal validity.

External Validity

External validity is related to whether the results of research can be generalized to a larger population. In other words, can study findings be applied to a broader context? External validity is related to sampling design and assumptions made about the relationship between the sample and general population. For example, the results from a study that only includes individuals 18 to 25 years should not be applied to the entire United States population, which includes people from a much broader age distribution.

To continue the previous example from this section, when studying the relationship between diabetes and weight, researchers may choose to only select people who are 18 to 25 years old and who have similar levels of physical activity because they suspect that age and physical activity may be confounders for weight. While controlling for these confounders may increase the internal validity but the generalizability (external validity) of the study results is decreased.

1. Martyn Shuttleworth (Jul 5, 2009). Types of Validity. Retrieved Aug 19, 2022 from Explorable.com: https://explorable.com/types-of-validity