Skip to main content icon/video/no-internet
icon backReturn to Entries

Unitary View of Validity

Validity is a fundamental aspect of evaluation and testing, especially for high-stakes testing for promotion, graduation, and competency. The concept of validity has been the subject of inquiry and debate over the past century. Many authors have studied this complex concept and attempted to define its components. The discussions and controversies about the meaning of validity and its components have continued, and our understanding about validity has substantially evolved. The question in defining validity is whether a test measures the variable for which it was intended.

The early approach toward validity was criterion based, which assumes there is a definitive value for each variable of interest and the goal of assessment is to measure this variable as accurately as possible. Therefore, criterion-based validity is used to determine how well the test scores predict the criterion scores. The main challenge with this approach is that one has to come up with a well-defined and clearly valid criterion measure. In many situations, however, these criterion measures are not readily available. Moreover, the validity of a criterion measure can always be questioned.

In 1954, Paul Meehl and Robert Challman coined the term construct validity, which was later expanded by Lee Cronbach and Meehl. At that point in time, the concept of construct validity was an addition to the criterion and content models but not an alternative option. Over the next couple of decades, construct validity became a common approach to validity. Samuel Messick built upon this approach into what is now considered the unitary framework of validity.

Unitary Framework of Validity

The unitary concept of validity evolved from the traditional framework of validity, which consisted of three separate types of validity: content, criterion (including concurrent and predictive validity), and construct. Messick’s main argument for abandoning the traditional view was that the old framework was fragmented and incomplete, and it did not incorporate the value of score meaning and social consequences of score use into the definition of validity. He asserted any validity type in its traditional form, individually or combined with other types, could not address all upcoming questions and queries regarding the various aspects of validity when a test is used in different contexts or situations.

Messick proposed a unifying concept of validity that brings all these types under one umbrella and integrates content, criterion, and consequences to create a comprehensive theory of construct validity to address both score meaning and social values in test interpretation. In this view, validity is an evaluation of the degree to which the theoretical and empirical evidence as well as theoretical rationales obtained through the validation process support the score interpretations for the intended use. It combines scientific evidence with logical argument to justify test interpretation and use. Validation is a work in progress, and this continuous and open-ended process is based on the accumulation of evidence to support certain meanings that the user aims to associate with scores from an assessment or test. The construct validation of the test score is not validation of the construct itself and it does not attempt to define the construct. In other words, validity only applies to the scores or their interpretation in a specific context. Therefore, the commonly used terms valid instrument and valid test are inaccurate. In addition, because validity is a property of inferences, not instruments or tests, validity must be established for each intended interpretation.

...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles

Read next

More like this

Sage Recommends

We found other relevant content for you on other Sage platforms.

close

Have you created a personal profile? Login or create a profile so that you can save clips, playlists and searches

close
close
close Signed In to profile You are signed in as:
close
Logged in Institution Institution