Abstract

In this chapter an item response modeling approach is discussed and illustrated that can be used for measurement and explanation at the same time. The main purpose of using item response models is the measurement of latent traits, to be considered as unobserved individual differences underlying human behavior and performances and also their test responses. In standard item response modeling, the items and the persons are represented by one or more parameters. The person parameter estimates provide us with a measurement of the latent trait(s). In an explanatory approach properties of the items and the persons are used to explain the responses. We will call this the “explanatory” approach (De Boeck & Wilson, 2004). The approach does not preclude measurement. Explanation and measurement can go hand in hand. The domain of application we will focus on is anger and aggression, but as will be explained in the Discussion section, the potential is much broader and includes other health-related variables.