deterring cheating in online environments
play

Deterring Cheating in Online Environments Henry Corrigan-Gibbs - PowerPoint PPT Presentation

Deterring Cheating in Online Environments Henry Corrigan-Gibbs Nakull Gupta Curtis Northcutt Stanford University Microsoft Research India MIT Edward Cutrell William Thies Microsoft Research India Microsoft


  1. Results: Cheating vs. Score Cheating doesn’t pay! 40 (p < 0.0001) 30 Frequency 20 10 0 0% 50% 100% 0% 50% 100% Scores (Honest) Scores (Cheating)

  2. Results: HC and Warning 50% 40% 31.7% Percentage cheating 25.9% 30% 20% 14.6% 10% 0% Control Honor Code Warn Online Exam

  3. Results: HC and Warning 50% Honor code group saw a drop in cheating rate 
 40% (not stat. signif.) 31.7% Percentage cheating 25.9% 30% 20% 14.6% 10% 0% Control Honor Code Warn Online Exam

  4. Results: HC and Warning 50% 40% 31.7% Percentage cheating 25.9% 30% 20% 14.6% 10% 0% Control Honor Code Warn Online Exam

  5. Results: HC and Warning 50% Warning has an effect 
 ( N =398, p < 0.001) 40% 31.7% Percentage cheating 25.9% 30% 20% 14.6% 10% 0% Control Honor Code Warn Online Exam

  6. Results: HC and Warning 50% 40% 31.7% Percentage cheating 25.9% 30% 20% 14.6% 10% 0% Control Honor Code Warn Online Exam

  7. Results: HC and Warning 50% 41.6% 35.8% 40% 31.7% Percentage cheating 25.9% 30% 18.3% 20% 14.6% 10% 0% Control Honor Code Warn Control Honor Code Warn Online Exam MTurk (India)

  8. Where we go from here…

  9. Where we go from here… • Does the effect of a warning diminish over time?

  10. Where we go from here… • Does the effect of a warning diminish over time? • Do warnings scare off honest and dishonest users at different rates?

  11. Where we go from here… • Does the effect of a warning diminish over time? • Do warnings scare off honest and dishonest users at different rates? • Can we use honey pots to automate the detection of unethical behavior online? Abuse, etc.

  12. Where we go from here… • Does the effect of a warning diminish over time? • Do warnings scare off honest and dishonest users at different rates? • Can we use honey pots to automate the detection of unethical behavior online? Abuse, etc. • Can warnings help deter other bad behavior online? Nasty comments, etc.

  13. Summary 50% 40% 31.7% Percentage cheating 25.9% 30% 20% 14.6% 10% 0% Control Honor Code Warn Online Exam

  14. Summary • We studied the effect of an honor 50% code and a warning at deterring cheating in an online course and 40% 31.7% Mechanical Turk . Percentage cheating 25.9% 30% 20% 14.6% 10% 0% Control Honor Code Warn Online Exam

  15. Summary • We studied the effect of an honor 50% code and a warning at deterring cheating in an online course and 40% 31.7% Mechanical Turk . Percentage cheating 25.9% 30% • We find evidence that the warning 20% 14.6% was effective in online settings… 10% less so the honor code. 0% Control Honor Code Warn Online Exam

Recommend


More recommend