Assessment Validity: An Exploration of Construct Validity


In any type of quantitative research, it is vital to analyse the reliability and validity of your methods and measurements. Validity demonstrates if and how accurately a method measures something. Suppose a method measures what it claims to measure. In that case, it can be considered valid. There are three main types of validity in the traditional model of validity, including:

  • Content validity
  • Criterion validity
  • Construct validity

Construct validity is one of the most critical measures of validity. It is one of the most important concepts in all behavioural science and the heart of most studies and tests. High construct validity is cruical to any HR tool or assessment we use in the workplace, it ensures the tests we use, such as a cognitive reasoning assessment or DISC personality assessment, measure what they claim to measure. Let’s explore what construct validity is and why it’s important for our HR assessments.

What is Construct Validity?

There are numerous measures to assess the validity of tests and assessments. One of the most widely used measures is Construct validity. During the early 1950s, the APA Committee for Psychological tests were increasingly concerned with the quality and validity of all the new theories being generated. The committee decided to assemble an effort to set standards for psychological measures. The idea of construct validity was born from this effort.

Construct validity is now used across many domains such as education, social sciences and psychology to determine whether a test measures what it is supposed to measure. The definition of construct validity is the extent to which a test measures what it states to be measuring. Construct validity should demonstrate that scores on a test predict the outcomes it says it does.

What is Construct Validity in Assessment?

Construct validity refers to if, and how well, an assessment measures the behavioural or psychological concepts that the assessment is designed to measure. Essentially, is the test constructed in a way that it successfully tests what it claims to test. For example, suppose the construct being measured is cognitive aptitude. In that case, the test should show evidence of measuring the concept of aptitude ability.

Why Construct Validity is Important?

Cognitive, behavioural and personality assessments are beneficial to make confident hiring decisions and improve workplace performance. There are two fundamental requirements for any reputable workplace assessment, reliability and validity. One of the most significant concerns when creating a workplace assessment is whether or not it actually measures what we believe it is measuring. This is also known as the validity of the assessment. Construct validity is important as it tells us whether our assessments are measuring what we think they are. Tests that demonstrate a high construct validity allows us to trust and be confident in the tests we are using. 

Construct validity is exceptionally important in social sciences and psychology, where there is much subjectivity of concepts. Often, there is no universally accepted unit of measurement for constructs. Even fairly well-known constructs, such as IQ, are keenly debated. Construct validity ensures we have a strong knowledge base about a construct and provide more confidence in the interpretation and reliability of the scores. 

How do you Describe Construct Validity?

Construct validity describes how well a test or experiment measures up to what it claims. An example is a measurement of constructs related to the human brain, such as intelligence, emotion, or ability. Some specific examples of these could be maths proficiency or displayed aggression, such as the Bobo Doll Experiment. These concepts are abstract and theoretical but have been observed in practice. When describing construct validity, we can explain it has high construct validity if it assesses aggression instead of assertiveness or social dominance. 

How to Determine Construct Validity

Researchers usually determine the construct validity of their theory or test before advancing with their studies to ensure they can obtain results and interpret the findings. Researchers determine the construct validity of a measure by establishing its relationship to several other measures with which it should associate either positively, negatively or not at all. Comparing the test to other tests that measure similar qualities and observing a correlation will determine the construct validity of your test.

How to Measure Construct Validity of a Questionnaire?

Construct validity refers to the extent to which a questionnaire adequately assesses the theoretical construct it purports to assess. The two most popular ways to measure the construct validity of a questionnaire are:

  • Multitrait-multimethod matrix (MTMM): The MTMM is simply a matrix or table of correlations arranged to compare how a measure relates to other measures. A MTMM displays correlations between two or more measurement techniques used to assess two or more constructs. 
  • Structural Equation Modeling (SEM): SEM is a multivariate statistical analysis technique combining factor analysis and multiple regression analysis to analyse the structural relationship between measured variables and latent constructs

For example, if you are measuring the construct of self-confidence you might ask questions such as ‘I like being the centre of attention’ or ‘I feel comfortable talking to people I do not know.’ We can predict that people with high levels of self-confidence might aggree with these statements whereas people with lower levels of self-confidence might disagree with these statements. This kind of model is known as factor analysis and contributes to the SEM measurement of a questionnaire. 

What is Good Construct Validity?

Construct validity is typically determined by comparing a test to other tests that measure similar qualities or traits to see how highly correlated the two measures are. If the qualities or attributes are highly correlated, we can presume that the test has good construct validity. Suppose we use an assessment in our hiring process and use the insights to help inform our hiring decisions. If the assessment validity is low or non-existant, how can we trust the credibility of the assessment results? Construct validity ensures researchers can measure their construct or concept. Ultimately proving a test has construct validity means that we can trust the assessments claim is true.