It is essential that your chosen research method measures what you intend it to, which is determined by testing your data’s reliability and validity.
Reliability refers to how consistent a measure is.
Using a research method that provides stable and consistent data is essential for the accuracy of your results. Research methods with immense variability–for example, an intelligence test that deviates by 50 points every time the same participant takes it–do not provide reliable results. To trust your data, you must ensure your research method is reliable.
When a survey instrument is reliable, researchers can more confidently state that it accurately measures what it claims. So how is that different from validity?
Validity is the extent to which a measure’s scores represent the variable they are intended to measure. Validity tests can help us make sure that the conclusions we draw from our data are reasonable based on what the data measures.
A measure could be highly reliable but not valid. For example, your bathroom scale could report your weight as 150 pounds week after week. However, if your scale is broken, and consistently off by 20 pounds, the weight it reports is not valid because it is not accurately measuring the variable it is supposed to (in this case, weight). It also cannot accurately be used to draw conclusions about what you weigh in a given week.
Survey data can also be reliable but not valid. Contrarily, survey data cannot be valid unless it is reliable. In this section, you’ll learn how to test for both reliability and validity.
