What is concurrent validity?

Validity usually refers to how accurately a test or method measures something. For instance, if a test measures what it is supposed to measure, and the outcome closely resembles other valid tests or real-world values, it is considered valid. This is especially important because the conclusions you gather from research are only useful if they’re valid.

Ensuring research is done properly depends on several factors, such as determining whether what is being tested is actually what is intended to be tested. Remember to also consider how observations are made and how they are influenced by the circumstances in which they are made.

Consequently, several validity types have been created to ensure that research is done correctly and that appropriate tests are picked to establish real relationships.

In this article, we’ll briefly look at different validity types and go into detail about concurrent validity. We’ll look at some examples of concurrent validity, the advantages and disadvantages of this validity method, and when concurrent validity is used.

Concurrent validity measures the amount of agreement between two different assessments. Typically, a validated test will be classified as the "gold standard," and concurrent validity will measure how a new test compares to it.

This new test will measure the same or similar constructs and help certify new methods against the accepted ones. If the results from the two tests correlate, a level of concurrent validity will be established.

Several validity types, apart from concurrent validity, have become a part of the research methodology. They include the following:

Construct validity

Construct validity measures the ability of the assessment to evaluate the construct in question. Basically, this validity answers the question, "Does the test measure the concept that it is supposed to measure?"

Indicators and measurements must be developed based on relevant and existing knowledge to ensure that construct validity is achieved. While a construct is often a concept that cannot be directly observed, it measures validity by observing other associated indicators.

Content validity

Content validity is the ability to evaluate how well an instrument measures or evaluates all the relevant parts of the constructs it intends to measure. Generally, a construct is an idea, theme, or theoretical concept that cannot be measured directly.

Psychological constructs are states like anxiety, self-esteem, or confidence, so scales to measure these states need to fully account for all ranges of feelings and behaviors. Numerous exploratory factor analyses and other multivariate statistical procedures are used to support the validity of the content.

Predictive validity

Predictive validity refers to the ability of a measurement or test to predict a future result or outcome. This result can refer to performance, behavior, or even a disease that could develop in the future.

For instance, getting a good Grade Point Average (GPA) and SAT/ACT scores is a trusted predictor of a student’s success in higher education. 

Face validity

Face validity refers to whether a test appears to be valid from external appearances and whether it measures the required aspect. If the test measures what it is supposed to measure, the test is said to have face validity.

Criterion validity, also known as instrumental validity, measures the quality of the measurement methods. This quality is demonstrated by comparing a measurement with a measure that is already known to be valid in the real world.

There are two types of criterion validity. They include concurrent and predictive validity, and both are used to show how a test compares against the gold standard or the criterion. However, there are some differences between the two:

• To establish concurrent validity, the test scores and criterion variables have to be measured at the same time.

• To establish predictive validity, test scores need to be obtained at one point in time, and the criterion scores have to be measured at a later time.

Concurrent validity is especially important when a new measure or test indicates that it is better in some way than its predecessors. For instance, this new test may claim to be faster, cheaper, or more objective.

To see how this validity method works in the real world, consider the following examples:

Evaluating nursing competence

In our example, nursing competence is usually assessed by asking a supervisor about a specific nurse in question. However, management wants to try out a new method for testing competence that involves outside professionals observing nurses at work instead of relying on supervisor input.

These two tests are run simultaneously, and the results show that the supervisor and outside professionals come to different conclusions about the nurses’ competence. Consequently, the new test is found to not have concurrent validity.

Comparing tests and GPA

In this example, a new math test is created. After the test is administered, researchers compare the results to the student's current GPA in that class. If the GPA correlates with the test's result, concurrent validity is established.

Aptitude test and supervisor assessment

To confirm the validity of a leadership aptitude test, a business compares these test results to a supervisor’s assessment of the potential recruit.

In general, there are two main reasons why concurrent validity is checked for:

• To make sure that a test is measuring what it claims to measure

• To supersede the original test

If the new test has a high concurrent validity when compared with an already accepted test, it can usually be used as a substitute. This is incredibly beneficial if the new test is less expensive, easier to implement, or shorter than the previous gold standard.

To determine concurrent validity, you have to measure the correlation of results from an existing test and a new test and demonstrate that the two give similar results.

Since the term “concurrent” means simultaneous, both the new test and the existing, proven test have to be carried out at the same time.

After the tests are completed and the results are finalized, the stronger the correlation between the two, the better the concurrent validity. Typically, the correlation value should be between 0 and 1, and the closer to 1 the results are, the better.

Concurrent validity is an excellent research method because:

• It is a quick way to validate data

• It is an accepted way to confirm personal attributes, including strengths, weaknesses, and intelligence

Some issues with concurrent validity need to be recognized. For instance:

• Concurrent validity can be skewed if the gold standard is biased. This bias can impact a valid measure and ultimately result in the new measure failing to achieve concurrent validity.

• Concurrent validity can only be used when a gold standard exists, which can be challenging to find.

• Concurrent validity can only be applied to tests used to assess current attributes and are not suitable for measuring future performance.

Concurrent validity scores usually range between 0 and 1:

• Less than 0.25: small concurrence

• 0.25 to 0.50: moderate

• 0.50 to 0.75: good

• Over 0.75: excellent

While both convergent and concurrent validity are established by calculating the correlation between a test score and another variable, each represents a different validation method.

Convergent validity tends to demonstrate how much a measure of one construct aligns with other measures of related constructs. In contrast, concurrent validity is used to show how a measure matches up to some known gold standard.

Both reliability and validity refer to how well a method measures something:

• Reliability: the consistency of a measure and whether the results can be reproduced based on the same conditions.

• Validity: the accuracy of a measure and whether the results represent what they are supposed to test.

Both reliability and validity are the difference between good and bad research reports since quality research depends on a commitment to testing and improving the validity and reliability of the results.