2.4c Reliability and Validity

Reliability and Validity

Researchers also strive to produce research that is reliable and valid. Reliability is the degree to which the results are consistent. A study’s results are considered reliable if the results can be replicated. Validity refers to the accuracy of the measurements, that is, that the researcher is measuring what they intend to measure.

How Can Research Be Reliable?

Calarco et al. (2021) showed that their research is reliable because they achieved intercoder reliability. Intercoder reliability means that at least two research team members independently analyzed (or coded) the same data and then came together to compare their results. If both researchers (or coders) found similar themes in the data, this established intercoder reliability. Sometimes, researchers produce different themes. When this occurs, they analyze the data together until they reach a consensus on coding criteria that makes sense to all research team members. They can then independently analyze and compare their coding on a subset of data to be sure their results are reliable.

How Can Research Have Validity?

Collins et al. (2020) used a survey created by the U.S. Census Bureau and the U.S. Bureau of Labor Statistics. The federal government has conducted this survey, the U.S. Current Population Survey, since 1940. Questions on it have been repeatedly tested for validity and to ensure that participants understand what is being asked so they can answer appropriately.