Keep students from using test banks to cheat

Many textbook publishers now make exam questions and answers available to faculty who use their books. These “publisher test banks” (PTBs) can be a boon to faculty members who are pressed for time. “Making up test questions is really time consuming,” said Christine Cheng, Ph.D., who will start this fall as an assistant professor of accounting at the University of Mississippi in Oxford, Miss. “Time is really precious. I need to spend every waking minute on my research to make tenure and to make a change in the world.”

But test banks come with one major drawback: Students have found ways to access them. Now, with a quick and easy online search, test takers can anonymously read numerous test questions and, in some cases, can even see their actual exam ahead of time. While some students use test banks as study aids, others look up test questions intending to cheat.

Cheng and D. Larry Crumbley, CPA/CFF, Ph.D., professor of accounting at Louisiana State University in Baton Rouge, La., studied test bank use by students in accounting classes and wrote about their findings in an article in the Journal of Accounting Education, “Student and Professor Use of Publisher Test Banks and Implications for Fair Play” (Vol. 42, March 2018). They determined that 48% of students used PTBs, and that those who did gained a performance advantage of around 30%.

Cheng and Crumbley’s findings suggest that PTBs could be a serious threat to fair play starting in the accounting classroom. We spoke with them about their study, their results, and their advice on how faculty can make it more difficult for students to use test banks to cheat.

The genesis of the article

Crumbley conceived of the idea to study test banks after being shocked to learn that the exam questions he helped write for the Forensic and Investigative Accounting textbook were readily available on the web. He looked into the matter and discovered an academic gray area that not everyone will discuss.

Some students, he said, don’t view using test banks as problematic. “You can’t necessarily fault the students,” he said. “Many of these students don’t think it is cheating.”

He voiced his concerns to Cheng, then an assistant professor at LSU. “This is a very complicated, but extremely important issue,” he said.

Initially, Cheng “didn’t think it was a big deal,” she recalled. But Crumbley convinced her otherwise.

The study and the findings

Students who use PTBs to cheat often memorize answers to PTB questions along with specific textual cues. For instance, say a student sees a multiple-choice question about the amount of income tax withheld from the bonuses Cubs baseball players received. The answer to the question is $6,250. The student could memorize the cues “Cubs” and “income tax withheld” along with $6,250 and thus be able to answer the question on an exam without understanding the accounting principles behind it.

Cheng and Crumbley developed a forensic method of determining whether students really knew how to perform the calculations necessary to answer exam questions, or whether they were using prior knowledge of PBT questions to answer them. They gave students a multiple-choice exam containing both PTB questions and those from other sources. They also altered some PTB questions, retaining textual cues like “Cubs” and “income tax withheld” and making the original “correct” answer from the PTB one of the choices, but changing the figures involved. Students who consistently chose the original answers from the PTB, they assumed, would likely have been memorizing the questions.

By performing a statistical analysis of students’ answers, Cheng and Crumbley determined that 48% of students used PTBs and that those who did performed about 30% better on exams than those who did not.

Ethical and pedagogical implications

The use of PTBs can make it more difficult to evaluate student learning, the authors argued. Faculty who use PTBs, they noted, might not be able to distinguish between students who did well on a test because they knew the material, and those who did well because they memorized answers.

PTB use also has some implications for fair play. Though some students do use test banks to study, Crumbley said he doesn’t think that’s fair to the students who don’t have access to them or who won’t use them for ethical reasons. Students who don’t use PTBs can “suffer real consequences,” he and Cheng wrote. “Since in-class exams often comprise a significant portion of students’ grades, students who do not use PTBs likely receive lower test scores.” That, in turn, could affect students’ course grades and overall GPA, and, perhaps, ultimately their job prospects.

Test-bank usage may also have significant consequences for the profession and for accounting education. Employers expect certain fundamental accounting knowledge from new hires. Students who only memorized test answers may receive poor evaluations or struggle to pass certification exams. This could hurt their chances at promotion and, ultimately, reflect poorly on the schools they graduated from.

Cheng and Crumbley label their study a wake-up call for academia. Their goal, they wrote, is for the research to “serve as a catalyst which sparks discussions among all shareholders (e.g., students, faculty, administrators) on how to ensure a level playing field in education.”

What faculty can do

Faculty, who are often strapped for time, are unlikely to give up using PTBs altogether, Cheng and Crumbley said. However, there are ways faculty can make it more difficult for students to use PTBs to cheat. Cheng and Crumbley recommend the following:

• Discuss proper use of test banks with students at the beginning of the semester. Remember that what you allow may be different from what another professor does.
• Write your own questions using PTB questions as a starting point.
• Alter PTB questions so that the answers aren’t the same as the ones readily available online.
• Use PTB questions from multiple publishers (note that some publishers will not allow faculty to use their PTBs unless they’ve purchased their textbooks).
• Rely more upon essays and open-ended questions that require students to demonstrate learned knowledge.

Cheng and Crumbley also note that the Teaching, Learning, and Curriculum section of the AAA has a new committee devoted to creating a multiple choice test bank not tied to a textbook publisher. They invite faculty to contribute questions and answers.

Dawn Wotapka is a freelance writer based in Georgia. To comment on this article or to suggest an idea for another article, contact senior editor Courtney Vien at Courtney.Vien@aicpa-cima.com.