Theoretically based behavioral interventions havedemonstrated effectiveness in stopping smoking,increasing physical activity and improving nutritionin order to prevent major chronic diseases such ascardiovascular disease, cancer and diabetes [1–7].Despite the development of effective behaviorchange programs, the magnitude of change in thesebehaviors has been relatively modest [8, 9]. Therising rates of diabetes and obesity, for example,indicate there is still much to learn about the mech-anisms of behavior change and how to maintainnewly acquired behavioral skills. One problemthat has slowed behavioral intervention research isthat the validity and reliability of our measures hassometimes lagged other innovations such as thedevelopment of effective tailored interventions [10,11] or analytical techniques for assessing moder-ators and mediators of behavior change, hencemaking it difficult to understand the mechanismsof behavior change and making it difficult toimprove our interventions [12–15].This special issue of Health Education Researchprovides an opportunity to consider an importantadvancement in our behavioral measurement meth-ods, specifically, how we can apply item responsemodels to improve our psychometric methods inhealth education and health behavior research andpractice.Althoughitemresponsemodeling(IRM)hasbeen used in educational testing over the last threedecades [16], it is an emerging method in healtheducation and health behavior research [17]. As, forexample, in health research, IRM is being widelyadopted to improve and revise quality of life ques-tionnaires [18–20]. There are many other innovativeapplicationsofIRMandastheninepapersinthisissueshowus,theseapplicationscanaidourunderstandingof the psychometric properties of scales beyondmaking questionnaires shorter and beyond whatmostofuslearnedinoursurveydevelopmentcoursesbased on classical test theory (CTT). In this briefafterword, we highlight current and future applica-tions of IRM and discuss how these methods mighthelp us improve the efficiency of our research. Webelieve these methods could lead to considerableimprovements in intervention methods, understand-ing the mechanisms of behavior change and de-veloping and refining theoretical models to makethemmore parsimonious aswell asprovide a founda-tion for considering the feasibility of computerizedadaptivetestingformeasuresofbehavioralconstructs.For those who are not familiar with IRM methods,the two papers by Wilson colleagues [21, 22] pro-vide an excellent tutorial by presenting an exampleusing the Rasch one-parameter model to examinea measure of self-efficacy using both dichotomousand polytomous models and then comparing IRMand CTT methods. In the first paper, readers learnthe basics of IRM, including the relationship ofindividual items as well as the role of item difficultyresponse patterns. In other words, item responsemodels estimate a result of the underlying construct