ABSTRACT
Automatic Item Generation (AIG) is the process of using item models to generate test items with the aid of computer technology. This chapter describes and illustrates the technology-enhanced three-step process for AIG that could prove useful in producing new multilingual items in an efficient and cost-effective manner. It explains how n-layer item modeling can be used to generate tasks in multiple languages by adding language as an additional layer in the model. The chapter also describes the three-step AIG approach in which test-development specialists first, and identifies the content that will be used for item generation. It then creates item models to specify the content in the assessment task that must be manipulated, and manipulates the elements in item models using computer-based algorithms. Item quality is evaluated using judgments form the basis of scrutinizing the items. There is a view that the science of educational assessments prevail to guide the design, development, administration, scoring, and reporting practices in educational testing.
