Using Paraphrased Test Bank Questions Reduces Academic Dishonesty in Remote Classes
What is this Research About?
In remote classes, invigilated exams are not always possible. Academic dishonesty in remote classes has become a widespread problem, where students have easy access to online databases and test banks. Instructors may use questions from published test banks in their exams to assess student learning, however, this practice can lead to an increase of academic dishonesty. Students can use search engines to look up questions and answers, purchase test banks, and release questions and answers on websites for other students to access. In this study, the researchers investigated students’ performances in online exams using verbatim and paraphrased test bank questions, along with honour codes and proctoring technology to determine whether paraphrased test bank questions can reduce cheating in remote exams.
What did the Researchers Do?
The researchers delivered three timed multiple-choice exams in online accounting classes at two universities. Question pairs were used for each topic, consisting of a test bank question taken verbatim from a textbook and a paraphrased version of the same test bank question. One question from each question pair was randomly chosen for each topic and thus, exams contained approximately 50:50 test bank and paraphrased questions. The questions were pre-tested to ensure that they measured the same concepts at the same level of difficulty. The first exam used this method alone. On the next two exams, the researchers implemented honour codes and proctoring technology, respectively. The researchers compared student performance (i.e., grades) on the verbatim and paraphrased test bank questions on the three exams.
What did the Researchers Find?
The researchers found that students scored higher in verbatim test bank questions than paraphrased questions (80.4% of questions correct vs. 69.1% of questions correct) across all three types of exams. These results indicate that paraphrased questions are a more accurate representation of students’ knowledge than verbatim questions, which students can look up online. The use of honour codes was found to be ineffective to reduce cheating. The researchers also found using proctoring technology overall decreased the incidence of academic misconduct for paraphrased and verbatim questions.