Abstract

The modern era in psychometrics began with the publication of Lord and Novick’s ‘The theory of mental test scores’ which introduced item response theory (IRT) to the world. IRT , together with its simplified version, the Rasch model, form the core of developments in assessment research today. They enable efficient use of item banking procedures, computer adaptive testing, and more transparent and effective test construction and item choice procedures. More recently, the common ground and theoretical underpinning between Structural Equation Modeling and IRT have been successfully bridged, giving birth to new and powerful psychometric programmes such as Mplus, R and Stata, that enable pipeline accuracy in test production procedures. We can’t teach these here, but at the end of this session you should be familiar with the item characteristic curve and be able to recognize a good pattern of curves from a poor one in terms of the three parameters.