SMABS 2004 Jena University
SMABS 2004 Home Organization About Jena Sponsors Links Imprint / Contact SMABS Home

European Association of Methodology

Department of methodology and evaluation research

Jena University

Contributions: Abstract

In this presentation (to be given by the first author), we will discuss how to extend explanatory measurement in two directions: External factors and internal factors (De Boeck & Wilson, in press). In the first direction, external factors, there are two sorts of extensio - one is associated with the nature of the external factors (i.e., person predictors, item predictors, and person-by-item predictors); and the other is associated with the nature of the effects (i.e., fixed or random effects of the external factors). The three types of predictors are:
(a) person predictors, including multilevel models for person predictors of the person group type,
(b) item predictors, including multilevel models for item predictors of the item group type (e.g., the hierarchical item model - Janssen, Tuerlinckx, Meulders & De Boeck, 2000), and
(c) person-by-item predictors (e.g., local item-dependency models such as extensions to DIF models).
In the second direction, internal factors, we will stretch the meaning of the term "predictor" to include predictors with latent values (parameters), and predictors that are themselves random variables, such as the responses on other items. We will call such predictors internal factors, because they stem from the data to be analyzed. This extension leads to the two-parameter (logistic and normal-ogive) model, multidimensional models (including models with item weights), mixture models, and local item dependence models.

De Boeck, P., & Wilson, M. (in press). Explanatory Measurement: Generalized linear and non-linear mixed models for item response data. New York, Springer-Verlag.
Janssen, R., Tuerlinckx, F., Meulders, M., & De Boeck, P. (2000). A hierarchical IRT model for criterion-referenced measurement. Journal of Educational and Behavioral Statistics, 25(3), 285-306.