“If you don’t have rich statistical models behind rich items, you’ve got nothing,” according to Measured Progress psychometrician Michael Nering. Nering recently co-authored a book about a new type of mathematical model, used to describe sophisticated items. “A lot of the success of Measured Progress has to do with our belief in items that measure more than multiple-choice items can, items that go into depth of knowledge and creative problem solving. These are constructed-response items,” Nering said. “Polytomous Item Response Theory models are used to help us understand how students are interacting with constructed-response items, and without these models, I don’t think that we’d be where we are today.” 

Nering collaborated with Remo Ostini of the University of Queensland, Australia, to publish Polytomous Item Response Theory Models. This 120-page book, part of the SAGE Publications series, “Quantitative Applications in the Social Sciences,” provides a comprehensive introduction to polytomous models and Item Response Theory (IRT). According to Nering and Ostini, the mathematical foundation of IRT is a function that “describes, in probabilistic terms, how a person with a higher standing on a trait (i.e., more of that trait) is likely to provide a response in a different response category to a person with low standing on that trait.” The authors’ research focuses on the advantage of using polytomous models over dichotomous models, which have only two response alternatives (correct and incorrect). Although they are “categorical items in the same way as dichotomous items,” polytomous models differ in that they have more than two possible response categories. 

Polytomous models—which include rating scale items, test items that provide partial credit for partially correct answers, portfolio assessment test formats, and multiple-choice items in which each response is scored separately—are becoming more prevalent. Nering and Ostini examine the range of available polytomous IRT models, focusing on the two types of response probability that are unique to polytomous models and their associated response functions. The book describes the major polytomous models, important variations on these models, and the relationships among different models. 

Polytomous Item Response Theory Models is a resource for professionals in the areas of testing and measurement, psychology, and other social sciences. It includes practical examples of major models with real data, and figures to illustrate important elements. Mathematical models are increasingly prominent in the social sciences, and so new and more complex approaches to measuring human traits are increasingly necessary. This book provides a comprehensive introduction to and case for one such new approach to measuring learning. 

Nering said that working with Ostini, was “a lot of fun.” Ostini and Nering met in graduate school. Ostini is a post-doctoral research fellow whose research interests include psychology of morality, bioethics, applied social psychology, and psychological measurement. Nering is the director of the Psychometrics department at Measured Progress, responsible for overseeing the daily operations of all psychometric functions. They decided to collaborate on the book because they saw a need for an introductory text in polytomous Item Response Theory models. The book is intended for students taking advanced, graduate-level courses in measurement, and is essentially the first to address the subject of polytomous IRT models. “A book like this would have made life much easier for Remo and me in grad school,” Nering noted. 

Currently, Nering and Ostini are working together on a new project—a book entitled Handbook of Polytomous Item Response Theory Models: Development and Applications. They are co-editing the text, which will include chapters written by leaders in the measurement field. The book will discuss the development of the most commonly used polytomous IRT models and their location within two general theoretical frameworks. A second section will demonstrate the applied settings where these models have been used. The authors’ intent is, “to bring together in one book the primary actors in the development of the most important polytomous IRT models to describe their work in their own words.” Handbook of Polytomous Item Response Theory Models is scheduled for publication in March 2008, prior to the annual American Educational Research Association meeting.