This chapter reports creation of a prototype assessment using virtual simulation for a test intended to assist in decision-making for non-native English-speaking military air traffic controllers’ placement into appropriate work assignments and to provide them with diagnostic feedback. The research and development focused on the domain definition inference, whose warrant was that observations of performance on the virtual interactive tasks for aviation English assessment are representative of the knowledge, skills and abilities required of air traffic controllers in an Asian country. Evidence Centered Design (ECD) guided the domain analysis and methods from Task-Based Language Assessment (TBLA) were also used, including eliciting opinions of two domain experts, document analysis, and a survey of 71 participants in the domain. Results indicated good backing for the first two assumptions: (1) critical aviation English skills, knowledge, and processes needed for aviation English communication in the military ATC context can be identified and (2) assessment tasks that are representative of the English for Specific Purposes (ESP) domain of ATC can be modeled as assessment tasks. The third assumption, (3) assessment tasks that require important knowledge and skills for communication in aviation English can be simulated in Second Life, was only partially supported.

The goal of ESL courses offered in American universities is to prepare international students for communication in other courses. This chapter investigates the degree of correspondence in the writing tasks used for assessment in an ESL course and those in other courses. The argument-based validity study focused on extrapolation, the degree of correspondence between performance on the assessment tasks and performance in the context of interest to scores users. Accordingly, this study compared course syllabi, assignment sheets, and students’ written assignments as well as analyzing instructor interview data and student survey responses. The results suggest that the ESL writing and disciplinary course assignments differ with regard to genre, topic, information sources, purpose, rhetorical functions, and length despite their similarity of target audience. In addition, interview data show that students tended to notice more differences than similarities in assignments in the ESL and other courses. The findings do not support the extrapolation inference in the interpretation/use argument for the assessed writing in the ESL courses. Instead, findings point to suggestions for revision to these writing assignments in the ESL academic writing course.