A programmed instruction tutoring system for Java™: consideration of learning performance and software self-efficacy

摘要

An undergraduate (n=23) and a graduate (n=23) class of information systems majors used a Web-based tutoring system during the first 3-h session of a 14-week course in interface design and implementation. The tutoring system taught a simple Java™ applet as the first technical training exercise, and the instructional design was based upon programmed instruction, which is a competency-based tutoring system. Software self-efficacy was assessed prior to using the tutor and at the end of the 3-h period. Students’ interactive performances (errors and help selections) were recorded for all interfaces in the tutor. The results showed that the undergraduate students made more input and test errors than did the graduate students, but the number of students in each class who completed all eight tutor stages (18 undergraduates and 17 graduates) was almost equivalent. Forty-four of the 46 students completed the fourth tutor stage, which presented frames of information explaining the items in the program. Students who did not complete all eight stages showed more errors on the initial four stages, in comparison to students who did complete all stages. Software self-efficacy increased from pre-tutor to post-tutor occasions for both classes and for both completers and non-completers. No significant relationship was found between software self-efficacy changes and tutor learning performance. Neither was gender related to software self-efficacy changes or learning performance. Evaluations of the tutor were favorable by almost all learners. A competency-based tutoring system may produce both skill and earned self-efficacy at the level of the individual learner, without regard to variations in the learning process leading to mastery.