Parsegon’s New Academic Honesty Tool.

Screenshot from future banner on the website, deploying on June 1st.

We’re announcing a new feature today addressing the dark word: Cheating. Available on June 1st, Parsegon’s optional extension will allow educators to detect cases of cheating automatically and take action if appropriate.

Academic honesty is a growing concern with the digitization of educational resources. While the age-old concern of Student A copying Student B’s homework has existed in eternity, known digital providers like Cengage and Pearson’s have led to solutions manuals being available online for both textbook and digital assignments.

Naturally, educators are growing concerned about the academic honesty of their classroom. At the end of the day, academic honesty hurts everyone. The student doesn’t learn the material fairly, it alters the averages of the class leading to skewed classroom statistics, and can lead to false expectations of future cheating-proof examinations.

Parsegon naturally addresses cheating in part. (1) Our platform allows teachers to create their own content minimizing the “Googla-bility” of solutions. (2) Moreover, uniquely on Parsegon, students must demonstrate their work to a solution which requires more upfrontness. If Student A copied Student B’s work, the cheating would be much more apparent if the work was provided, not just the final answers.

However, we are happy to announce an optional Academic Honesty extension to the Parsegon platform. This extension has two intentions: (a) an organized reporting to educators to see unethical copying between two students that checks against randomized, arbitrary, or natural trends and (b) deep analysis to check if that trend has persisted over time.

Screencapture of current build

The platform has two statistics at the top: the anomaly detection rate and average coagulation rate. A friend recently corrected my spelling of anomaly so please forgive the typo in the screenshot!

Anomaly detection rate is the number of students in the classroom that have anomalies in their submissions. A high rate is indicative of a likely persistence of cheating but a very high rate could also indicate an assignment is very discrete in nature making the data less confident. In the ending, it’s up to the educator to survey the cases of alleged collaboration to determine if it is worth investigating.

Average coagulation rate is of the surveyed cases of anomalies, how severe the copying is. Unlike doing a blind comparison, Parsegon’s Academic Honesty module focuses on outlier data that correlates between users. Since outlier data is naturally “unlikely” by discrete nature, correlation between points is extremely unlikely if not for collaboration.

The platform shows side by side comparisons, statistics on how different they were from the rest of the class, and likelihoods that they are by randomness. It also does deeper analysis into other assignments to determine if such students had a possible link in the past alone.

By its nature, we hope this feature isn’t used often. But with the growing concern of Academic Honesty, we also hope educators can utilize it’s strength to ensure student’s stay truthful in their submissions.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.