Skip to ContentSkip to Navigation
University of Groningenfounded in 1614  -  top 100 university
EDU Support Blackboard Instructor Assignments, Assessments & Exams Quality & Invigilation Cheating prevention tools and resources

Cheating prevention tools and resources

Besides online surveillance, a combination of didactic considerations and tools can be used in order to improve exam quality and minimize fraud risks.


Options

Didactic considerations / examination design

  • Conversion to individual exam forms as much as possible

    • Randomizing questions per student, so that each student receives a unique exam

    • Selecting questions from a large item bank or parameterized so that each student has different questions.

    • Calling in random students after the exam with image and sound

  • Open questions instead of closed questions

  • More interim formative examination

  • Not a summative examination every course unit separately: merging of course unit examination to reduce pressure. This examination could, for example, take the form of a cross-curricular oral exam, paper or assignment that assesses knowledge, understanding and skills

  • Time limitations (cannot always be technically adapted sufficiently and are sensitive to technical problems on the part of the University of Groningen and the student)

    • Strict start and end times to restrict access

    • Making backtracking impossible

    • Limiting the time per exam question

  • More didactic considerations and alternative assessment methods...

Awareness

  • Student pledge

  • Raising awareness of scholarly conduct and ethical behavior during lectures

Fraud Detection Tools

  • Essay plagiarism scanner Ouriginal

  • Analysis of logs and grades (Nestor logs download)

  • Analysing the data that was collected and generated about the student who took the exam (user logs and autosaves)

  • Analysing non-personally identifiable data on previous exams (e.g. results), to assess whether substantial deviations have occurred that indicate possible cheating in the online 

Extra measures (conditional)

  • Exam on campus (on request)

  • Exam with online surveillance (according to the guidelines)


Anti-cheating strategies per exam type

For different exam types, different strategies can be followed to prevent fraud. Furthermore, a few methods to get an indication of fraud using Nestor tools are mentioned (detection). Apart from prevention and detection, sanctions are the third component of combatting fraud. The Board of Examiners is authorized to pass judgement on students who are suspected to have committed fraud.

Disclaimer: Think carefully before adjusting your exam, as changing the set-up might introduce an extra testing variable. For instance, stricter time restrictions also test how quickly students are able to answer questions, even though this might not be a learning objective. Be sure to check the exam with the Board of Examiners if you do make any changes.


General recommendations

Independent of the type of exam or type of questions, there are a few general guidelines to follow to prevent or detect fraud.

Prevention

Detection

  • Have students sign the Student Pledge

  • Check the identity of students by asking them to show/upload their student card

  • Offer unique versions of exams/questions by offering a unique data set to every student or use their S-number as a variable (parameterizing)

  • Perform random oral checks afterwards.

  • View the user logs of students, to see whether students answered a question at the same time, whether their IP addresses match or whether a claim of a ‘technical problem’ is justified.

Note: matching IP addresses do not prove physically working together (the students could be housemates), nor do different IP addresses rule this out (one could be on a mobile hotspot).


Test - MC/closed questions

This type of exam has the highest risk of fraud, due to the lack of unique answers.

Prevention

Detection

  • Randomize questions and answers

  • Offer a different subset of questions to each student

  • Prohibit backtracking

  • Use severe time restrictions or consider a form of surveillance

  • Check the ‘Access log’ under ‘Test information’ for irregularities when viewing an attempt.

  • Analyse submissions by downloading all answers via the Grade Center: select column, click ‘Download results’, choose ‘by User’ - ‘All Attempts’ and click to download results.


Essay assignment

In (essay) assignments, students answer the same question but their answers are generally long enough to be unique and therefore the risk of fraud is medium.

Prevention

Detection

  • Use question flows to offer different versions of questions and offer questions in a random order (and check answers against other versions)

  • Prohibit backtracking

  • Backups of answers on essay assignments are stored every two minutes by AutoSave. If copy-pasting is suspected, chronology can be traced by combining user logs with the backups.

  • Use a plagiarism scanner such as Ouriginal to scan submissions for plagiarism.


Assignment

Because every submission is unique for this type of exam, the risk of fraud is small. The main concern is to make sure the students do the work themselves.

Prevention

Detection

  • Enable the Ouriginal plagiarism scanner

  • Provide different versions of the assignment

  • If Ouriginal is enabled, every submission is automatically scanned for plagiarism and given a score based on similarities to other submissions and a database of submissions.


Oral exam

Since the examiner can see the student at all times, the risk of fraud is small. 

Prevention

Detection

  • Prepare multiple questions per learning objective.

  • Ask the student for identification.


Last modified: 19 November 12:09 PM
In need of immediate support? 050 - 363 82 82
Follow Nestorsupportfacebook twitter youtube