Expansion and evaluation of a step-based tutorial program for linear circuit analysis

The current status of a computer-based tutorial system that uses a step-based tutoring approach to teach elementary linear circuit analysis is described. The system features automatic problem generation (varying both element values and circuit topology) and solution of those circuits by a variety of methods, including node and mesh analysis and current and voltage division. A wide variety of student inputs, such as algebraic and matrix equations, numerical answers, and redrawn circuits are accepted. A web-based interface is being developed with the capability to pose questions of various types in a sequence that can be specified by a tutorial writer in an authoring interface. Various pedagogical features such as color coding are employed to aid in student learning, and typical student misconceptions are being addressed. Recent work includes the addition of a web-based instructor interface where instructors can create class sections and continually monitor student progress, including access to detailed logs of student activity. Graphics are now displayed directly within the program to eliminate the need for PowerPoint and facilitate wider usage. A web-based waveform sketching tool is being developed that has the potential for wide use in other courses such as calculus as well. More complete problem solutions are now available including explicit equations for the desired voltages, currents, and powers. The software has been used on a mandatory or strongly encouraged basis in 10 sections of a linear course at Arizona State University (totaling over 560 students) and by 42 students at the University of Notre Dame, and a few students at the University of Virginia and two community colleges in 2013. Student satisfaction has been very high at all sites. A controlled, randomized laboratory-based study showed that learning gains are approximately 10X higher using the software tutorials than when working conventional textbook problems for the same period of time, with a statistically significant effect size (Cohen d-value) of 1.21 standard deviations.