Durable Leadership Library

The most reliable source of pragmatic leadership insights.

SubStack / Medium

Research in the social sciences is complicated and often hard to measure. This difficulty comes from the fact that social processes are often hard to describe clearly in terms of numbers. Reliability and validity assessments are two concepts in qualitative research that are used to add rigor the discipline. Assessing three types of validity—content, construct, and criterion—shows what a certain method can and can't do when it comes to measuring a theoretical concept qualitatively. Also, these evaluations help find important quality issues that might be affecting the results of the research. Because of this, assessing reliability and validity are important ways to make sure the quality of social science research.

Content validity is how well a qualitative measure covers all the parts of a theoretical concept being measured and leaves out nothing important to the concept. Content validity is the degree to which a qualitative measure takes into account enough of a theoretical concept's characteristics. Sampling validity is one type of content validity. For content validity, qualitative researchers often do a thorough review of all the literature on the topic they are studying to find all the possible signs of the concept they are looking into. The next step is to decide which indicators should be part of the qualitative measure. This choice was made after looking at the indicator's "face validity," or how well it seems to measure the concept. Once the qualitative measure is made, it can be used to measure the content validity of future qualitative research studies.

Construct validity is how well a measure of a variable matches a concept's general theoretical framework. This is important because it helps researchers figure out if their measurements are actually measuring what they want to measure. For a measurement to have a high level of construct validity, it must be both reliable and valid. If a measurement isn't accurate, the results will be different every time it's used. If a measurement isn't correct, it won't give a true picture of the thing it's supposed to measure. Because of this, construct validity is important to make sure that research measurements are correct and give the same results every time.

Criterion validity is about how a quantitative measure relates to a real-world analysis that uses the measure. This includes predictive validity, which shows how well the results predicted by a certain numeric measure match the results from measuring other variables that are related. Criterion validity is a key part of figuring out how reliable and valid qualitative research methods are. Without this evaluation, it would be hard to judge how true the research findings are. Also, it would be hard to compare and contrast different qualitative research methods without criterion validity. Criterion validity is an important part of any qualitative research study for this reason.