Online Exams Make the Grade
Most faculty would wince at the mere thought of putting their exams online. Security risks and the potential for academic dishonesty have kept consideration of this topic to a minimum—until now. Dr. Mark Stanton, associate professor of psychology, has not been deterred by the pitfalls of electronic testing.
In the summer of 2003, Stanton and graduate research assistant Barney Pagani formulated a daring vision for web-based assessments for Psychology 312: Learning and Motivaton. Through a Technology-Enhanced Course Redesign grant proposal, the Stanton-Pagani team crafted a plan for a fully-equipped secure testing lab. Patterned after testing center guidelines set forth by Educational Testing Services, the proposal called for room design, carrels, computers, network equipment, video surveillance, and .
Why go to such lengths to give exams? Stanton uses an instructional approach called the Keller Method with favorable results. In keeping with the method, students take objective tests to a mastery criterion before attending classes. According to Stanton, “the hallmark of Keller is mastery learning.”
Prior to class meetings, students must pass exams as evidence of basic concept mastery. It’s a carrot and stick system where passing an exam is the stick and subsequent admission to intensive class meetings is the carrot. As a result, class time is more effective because students can apply knowledge to in-depth class discussions.
The Keller Method is also known for its feedback mechanisms. Incorrect exam responses are followed by feedback that directs the student to the correct source material. “Students need to know not just when they’re wrong but where to go to clarify their understanding,” Pagani mentioned.
The Keller Method stipulates that exams are to be taken at the student’s discretion. In previous semesters, Stanton’s exams were in a pencil and paper format. Since students were able to set their own test-taking schedule, administering and grading exams and providing feedback was taxing. Enter the need for an automated testing system.
The summer-long grant project produced a testing environment that is as secure as it is supportive of the instructional method. Aided by PRESENT staff members Justin Schakelman and Jeff Whisler, the team set out to launch the lab by fall of 2003.
Two major pieces make up the secure testing site: software and hardware. Exams are managed by . The learning management system “was a natural fit,” Stanton said. “It organizes and delivers exams to our exact specifications,” Pagani added, “and it provides the necessary feedback to students.”
automates the time-consuming task of grading and guards against cheating. “Features like student authentication and randomly generated questions from sets of items are important to our use of the Keller Method,” said Stanton. has several lines of defense for those foolish enough to attempt cheating. “We configured to track students’ responses, scores, and attempts right down to the second,” Pagani noted. “There’s simply no easy way for students to take an exam fraudulently.”
Tracking students’ movements is central to the security of this instructional innovation. Stanton made it clear early on that “protecting against academic dishonesty is critical to the success of the Keller Method.” The team assembled a hardware infrastructure to combat the cunning. They installed an Internet-based video surveillance system and computer monitoring program. The ceiling-mounted camera records and archives all lab activity. Students cannot begin until the proctor releases the exam using software called NetOp. At any time, the proctor may use NetOp to take over the student’s computer remotely if there’s an appearance of impropriety. The room has its own local area network which gives added control in keeping intrepid hackers out. The network has a thick shield that “makes it virtually impossible for outsiders to capture exam content,” said Jeff Whisler, designer of the lab’s security system. All six computers function as terminals with no access to disk drives, ports, or any other medium to which content could be saved.
Amid all the passwords and precautions, the lab does “a very good job of streamlining the administrative side of the Keller Method,” said Stanton. More importantly, it works to improve student achievement with less administrative overhead. In a recent course method comparison study, Stanton and Pagani found that test scores were significantly higher in the Keller Method sections as compared with traditional lecture sections.
Some of Stanton’s students were especially impressed with how easy it was to stop by the lab, take a test, get feedback, and get on with life. Students also reported that they learn more in the Keller group, “and that’s the name of the game,” Stanton added with a smile.
Return to Faculty Profiles Page
Students reported that they learn more “and that’s the name of the game.”
|©2006 University of Delaware|