A simple and efficient markup tool to generate drawing-based online assessments

N. Nytko, M. West, and M. Silva

in Proceedings of the 2020 American Society for Engineering Education Virtual Annual Conference (ASEE 2020), 2020.

Due to an increase in student enrollment in engineering programs, many instructors are now adopting automated computer-based systems to deliver homework and exams to students. Commercial and free online learning systems such as Gradescope, Pearson MasteringEngineering and PraireLearn give instructors the ability to import or create auto-graded questions involving a mix of multiple-choice, multiple-select, numerical, and symbolic input. However, the ability to auto-grade questions that involve graphing or sketching is still very limited. This constraint has great impact in introductory engineering classes where the ability to hand-draw diagrams and graphs is an important learning objective. For example, drawing free-body diagrams in Statics, bending-moment diagrams in Strength of Materials, and circuit diagrams in Electronics. In this study, we present an online tool that uses a simple HTML markup language to create automated drawing-based questions, allowing students to draw diagrams, graphs and design solutions on the computer that are instantly auto-graded by the computer. While solving homework questions, students receive immediate feedback about their drawing, and can practice this skill until they achieve mastery. Instructors can also use the drawing tool to generate randomized drawing questions for computer-based exams and homework. A key advantage of this new tool over previous work is that the question author does not need to write any explicit programming code. The drawing tool has been used to generate homework and exam questions in four large courses (Statics, Solid Mechanics, Dynamics, and Numerical Methods) at a Midwestern University, and it can be easily extended to other courses. This paper will focus on three aspects of the drawing tool: (1) description of the markup tags that create mechanics objects, such as pins, rods, pulleys, forces and moments, and how they can be adapted to create drawings in different contexts; (2) how these objects can be created using random parameters, so that students receive different versions of the same question; and (3) the algorithm used to grade student drawings and the type of feedback provided. We present results from student interaction data with the system, student surveys, and feedback from instructors and question authors.

DOI: 10.18260/1-2--34049

Full text: NyWeSi2020.pdf