Please consult the event program for date, time and location.
In the context of the US public school system, tests of English language proficiency (ELP) are used for multiple purposes, including whether to keep English learners (ELs) in language services, and if so, at what level within the program.
To measure such a broad construct, test designers are faced with two potentially conflicting constraints: the need to design a test that provides accurate measurements across the range of performance levels and adequate coverage of all of the critical areas of the domain, while keeping the test at a manageable length.
This paper discusses the approach taken in a linear paper-based test of academic English language for grades 1-12, and how the transition to a computer-based test allowed for a design that better fit the demands of the test. Ultimately a multi-stage computer adaptive (MST) approach was adapted.