two approaches or models of validity—trinitarian and unitary. In some ways, these two models are competing views of gathering evidence for a test’s validity. In other ways, the two approaches have an overlap of elements. For this discussion: Be sure to include citations from Guion’s 1980 article, “On Trinitarian Doctrines of Validity,” and Messick’s 1995 article, “Validity of Psychological Assessment: Validation of Inferences From Persons’ Responses and Performances As Scientific Inquiry Into Score Meaning.”
Validity is a crucial concept in the field of psychological assessment and is concerned with whether a test measures what it claims to measure. There are two main models of validity, namely the trinitarian and unitary models. These models propose different approaches to assessing the validity of psychological tests and have both unique elements and areas of overlap. In this discussion, we will explore the trinitarian and unitary models of validity, drawing upon the works of Guion (1980) and Messick (1995).
The trinitarian model of validity
The trinitarian model of validity, as described by Guion (1980), suggests that validity should be assessed through three distinct approaches: content, criterion-related, and construct validity. According to Guion, these three approaches together provide a comprehensive framework for evaluating the validity of a psychological test.
Content validity involves examining the content of the test to ensure that it adequately represents the construct it intends to measure. This approach focuses on the match between the test items and the construct being assessed. Guion (1980) emphasizes that content validity is particularly important for tests that claim to measure specific domains or content areas.
Criterion-related validity, another component of the trinitarian model, involves assessing the relationship between test scores and a criterion measure. The criterion measure can be an external standard or a measure of another construct that is theoretically related to the construct being measured by the test. Guion (1980) explains that criterion-related validity provides evidence for the extent to which the test scores predict or correlate with the criterion measure.
The final component of the trinitarian model is construct validity, which aims to evaluate the underlying theoretical construct that the test seeks to measure. Messick (1995) highlights that construct validity involves examining the relationships between test scores and measures of related constructs, as well as assessing the convergence and divergence of these relationships. It is concerned with the theoretical framework underlying the test and the extent to which the test aligns with that framework.
The unitary model of validity
In contrast to the trinitarian model, the unitary model posits that there is a single definition and approach to validity. Messick (1995) argues that the unitary model is based on the idea that validity is not separable into distinct components, but rather a global evaluation of the inferences drawn from test scores.
According to Messick (1995), the unitary model considers validity as a unitary concept that encompasses four aspects: content, response processes, internal structure, and relations to other variables. Content validity in the unitary model is similar to that in the trinitarian model, as it examines the match between the test items and the construct being measured.
However, the unitary model goes beyond content validity by including the assessment of response processes. Response processes refer to the cognitive and psychological processes that individuals engage in when responding to test items. Messick (1995) argues that understanding the response processes is vital for interpreting test scores and ensuring validity.
Internal structure validity, the third aspect in the unitary model, involves examining the internal consistency and factor structure of the test. It focuses on the reliability of the test and the extent to which the items of a test measure the same construct.
The fourth aspect of the unitary model is the relations to other variables. This component is akin to criterion-related and construct validity in the trinitarian model, as it encompasses the examination of the relationships between test scores and external criteria or measures of related constructs.
Overlap and divergence between the trinitarian and unitary models
While the trinitarian and unitary models propose different approaches to validity, there are also areas of overlap between these models. Both models emphasize the importance of content validity, which involves ensuring that the test items adequately represent the construct being measured. Additionally, both models consider the relationships between test scores and other variables or criteria, although they may use different terminology to describe this aspect.
Guion (1980) and Messick (1995) acknowledge the overlap between the models and propose that they can be integrated to provide a more comprehensive assessment of validity. Guion (1980) suggests integrating the trinitarian model by using content, criterion-related, and construct validity simultaneously to gather evidence for validity. Messick (1995) proposes a similar approach, arguing that content, response processes, internal structure, and relations to other variables are all important aspects to consider when evaluating validity.
The trinitarian and unitary models provide different perspectives on the assessment of validity in psychological tests. The trinitarian model proposes content, criterion-related, and construct validity as distinct components, while the unitary model views validity as a global evaluation of inferences drawn from test scores. Despite their differences, both models highlight the importance of content validity and considering the relationships between test scores and external criteria or related constructs. It is suggested that the trinitarian and unitary models can be integrated to provide a more comprehensive evaluation of validity.