Skip to main content icon/video/no-internet

The concept of validity is one of the most influential concepts in science because considerations about its nature and scope influence everything from the design to the implementation and application of scientific research. Validity is not an abstract property of any observable, unobservable, or conceptual phenomenon, such as a measurement instrument, a personality trait, or a study design. Rather, validity is a characteristic of the inferences that are drawn about phenomena by human agents and the actions that result from these inferences. Specifically, validity is always a matter of degree and not absolutes. This stems partly from the fact that validity is not an observable characteristic of inferences and actions but something that has to be inferred also.

The evaluation of the degree to which inferences are valid and resulting actions are justifiable is, therefore, necessarily embedded in a social discourse whose participants typically bring to the table diverse frameworks, assumptions, beliefs, and values about what constitutes credible evidence. Specifically, modern frameworks for validity typically list both rational and empirical pieces of evidence as necessary, but in each individual context, what these pieces should look like is open to debate. Put differently, a coherent statement about the validity of inferences and actions requires negotiation as well as consensus and places multiple responsibilities on the stakeholders who develop such a statement.

Negotiating Validity

A metaphor may illustrate complications that can arise in a discourse about validity. If an educational assessment is viewed as the construction of a house, inferences are markers of the utility of the house. In this sense, an evaluation of the validity of inferences can be viewed as an evaluation of the degree to which the house provides structural support for the purposes that are envisioned for it. Obviously, the parties who are envisioning a certain use of the house are not necessarily the same as the designers or builders of the house, and so discrepancies can arise easily. Of course, other reasons for a mismatch are possible and could stem from a miscommunication between the designers of the house and the users of the house or from a faulty implementation of the design plans for the house. In a sense, the search for inferences that can be supported can be viewed as the search for how a house can be transformed into a home.

In general, the stakeholders in an assessment can be coarsely viewed as belonging to four complementary groups. First, there are the test developers, who create a research program, a framework, or an instrument under multiple considerations, such as theoretical adequacy and feasibility for practical implementation. Second are the examinees, whose needs in the process are typically more practical and may differ quite substantially from those of the other stakeholders involved. Third are the test users, or the decision makers who utilize the scores and diagnostic information from the assessment to make decisions about the examinees; only rarely are the examinees the only decision makers involved. Fourth are the larger scientific and nonscientific communities to which the results of an assessment program are to be communicated and whose needs are a mélange of those of the test developers, the test users, and the examinees. Therefore, determining the degree to which inferences and actions are justifiable is situated in the communicative space among these different stakeholders.

...

  • 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

Sage Recommends

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

Loading