Evidence Based on Test Content

Abstract

The previous chapter focused on the need for latent constructs in the study of affective characteristics and the array of scaling and measurement issues that need to be addressed in order to justify their treatment as mathematical structures in affective instruments. In this chapter, we introduce the most fundamental concept in instrument development: validity. While validity has no single agreed definition in the larger scientific community, in the area of instrument development it does refer to a very specific evidence gathering process for inferences made about latent quantities of latent constructs. As a result, Chapter 3 can be seen as an extension of the latent construct operationism started in Chapter 2. This evidence gathering process is a recurring one with elements discussed also in Chapters 4, 5, and 6. The test content validity discussed in this chapter is part of a non-statistical judgmental process done by content experts before it is present to respondents. As a result, the chapter is devoted to the methods involved in helping instrument developers translate their conceptual understanding of the affective characteristic into a well-functioning operational construct. This chapter introduces the researchers to the most commonly employed approaches to the process of establishing validity through the test content