Using Rasch Analysis for Determining the Cut Score of a Computer Science Placement Exam

• In Event: Rasch Modeling Methodologies

Abstract

A key strategy for broadening computer science (CS) participation in a large midwestern urban district has been the enactment of a high school CS graduation requirement. The Exploring Computer Science (ECS) curriculum and professional development program serves as a core foundation for supporting policy enactment. However, students with prior CS background might find the course repetitive. This paper reports on district efforts to develop a placement exam for students to take an advanced CS course in lieu of ECS. The placement exam tasks were modeled after the ECS exam tasks but with higher difficulty. Research partners used Rasch modeling to equate the placement exam tasks to the ECS exams and to establish a cut score for passing the placement exam.

Authors

• Steven McGee, The Learning Partnership

• Everett V. Smith, EVS Psychometric Services, LLC

• Andrew Rasmussen, Chicago Public Schools

• Jeremy Gubman, Chicago Public Schools