Validity and reliability: how are they different?

Validity and reliability are two key constructs when analyzing many of the results produced by studies carried out in psychology. Using more or less valid, more or less reliable instruments can subtract or add value to the study itself. Now, validity and reliability, how are they different?

To understand a little about the distinction between validity and reliability, it is necessary that we know what they are, so that, having a firm foundation at a conceptual level, we can develop the aspects that differentiate them and make them unique when comparing them.

The validity

Although there are different types of validity, when we refer to it in psychometrics, we refer to the degree to which an instrument measures the variable it is intended to measure (Hernández-Sampieri, Fernández and Baptista, 2014).

For example, if a scale allows us to know the weight of an object, we can affirm that it is valid to the extent that it fulfills the function for which it was designed; If we could not perform such a measurement, then the scale would not be valid for that specific task. If a test of anxiety that purports to measure it does not do so, then it is invalid.

On the other hand, in research the concept of validity “refers to to what is true or what is close to the truth. In general, it is considered that the results of a research will be valid when the study is free of errors” (Villasís-Keever, 2018).

Determining validity is essential in all research, as it is the previous step to be able to trust the data obtained from that research design, data acquisition, analysis of data processing, interpretation and its resulting judgment (Cohen, Manion and Morrison, 2018).

The reliability

Reliability or trustworthiness is understood as the consistency or stability of the measurements when a measurement process is repeated (Prieto and Delgado, 2010). It is the degree to which an assessment tool produces consistent results when measuring the same thing under the same conditions.

Thus, a common procedure to obtain the reliability of an instrument is to have it measure the same non-changing variable several times. Since this is complicated – it is complicated, for example, for the anxiety experienced by a subject to not vary – what is usually done is to compare the measurement we obtain with the evaluated instrument with that given to us by an instrument that we know is reliable.

Let’s go back to the scale example. If we only weigh the object once, we cannot affirm that the measurement is reliable, that is why we carry out several measurements and if we observe that we obtain very different values ​​in each one, we can affirm that it is not reliable.

The same happens in research and with all the instruments from which we obtain measurements, such as psychotechnical tests.

Differences between validity and reliability

Validity and reliability are closely related concepts, but they are different, have different purposes and focus on different aspects that provide different information about the quality of the research or measurement instrument. Let’s look at some of the characteristics that make them different:

1. Measurement

One of the main differences between validity and reliability is that the former measures the degree to which a research instrument measures what it is intended to measure. While the second measures the degree to which the measurement or research instrument produces consistent results, when repeated measurements are carried out.

2. What are they associated with?

Validity relates to the correct or accurate applicability of the instrument or research in a situation. Reliability is associated with the stability of the results of a test.

3. The objective

Another difference between validity and reliability is that validity is focused on the result and reliability on the consistency of these, regardless of whether they are correct or not. That is, there can be reliability without validity. In an investigation or test we can obtain consistent results, although that does not imply that they are correct. But we cannot have correct results without them being reliable.

4. The research question

Validity asks “does this instrument measure what it should be measured?” Reliability questions “how representative is this measurement?” Can we trust her?” Each question refers to a different object of study, that of validity is precision, that of reliability, consistency or stability.

5. The evaluation

Validity and reliability differ in the way they are evaluated. Reliability is evaluated checking that the results are consistent over time, between different observers, and between parts of the test itself. While validity is evaluated by checking how well the results correspond with established theories and other measures of the same concept.

The validity and reliability of an instrument are two different properties, but as a whole they give us an idea of ​​the quality of that instrument. The problem is that obtaining reliable and valid measurement instruments in psychology is especially expensive due to the abstract nature of the medium; This cost makes it difficult to develop an instrument that does not add a significant amount of error to the results of the research in which it is used.

The good news is that psychology has been on this battle front for years and notable progress has been made in this regard.

You might be interested…