Sponsor Videos

Canvas
Pearson
Software Secure

Keypath Education

Conference News


Twitter
LinkedIn FaceBook YouTube GooglePlus www.instagram.com/onlinelearningconsortium

 

Download the Mobile App
IOS  |  Android
OLC Mobile App


Make your travel arrangements


Yoga with Jan

Add to my registration

 

American Higher Education in Crises book cover

Join keynote speaker Goldie Blumenstyk for a book signing.

Books are available for pre-purchase for $16.95 (+tax). 
Read more


Conference Program now posted! This year's line-up includes:

 

OLC Excellence and Effective Practice Award Recipients Announced

 

Add/remove sessions from the Program Listing on the website or in the mobile app to create a list of sessions you want to attend!

My Schedule



Join Keynoters Goldie Blumenstyck (Chronicle of Higher Education) and Phil Hill and Michael Feldstein (MindWires Consulting)


BYOD to learn, explore, and share knowledge within this lab environment

Technology
Test Kitchen

Save the Dates

22st Annual OLC International Conference
November 16-18, 2016 | Orlando, Florida | Walt Disney World Swan/Dolphin Resort

OLC Innovate 2016 - Innovations in Blended and Online Learning
April 20-22, 2016 | New Orleans, LA | Sheraton New Orleans Hotel

Examining the Likelihood of Cheating in Online Classes and the Effect of Proctoring on Cheating

#Twitter: 
#olc45453
Presenter(s)
Helaine Alessio (Miami University, USA)
Nancy Malay (Miami University, USA)
Beth Rubin (Miami University, USA)
Session Information
October 16, 2015 - 11:45am
Track: 
Technology and Emerging Learning Environments
Major Emphasis of Presentation: 
Research Study
Institutional Level: 
Multiple Levels
Audience Level: 
All
Session Type: 
Information Session
Location: 
Oceanic 4
Session Duration: 
45 Minutes
Session: 
Concurrent Session 11
Abstract

Evidence from within and across course sections is provided indicating that Proctoring software influences test scores in online courses, and promotes student accountability and integrity.

Extended Abstract

Online courses have grown in popularity with estimates of 20-50% of college students enrolling in at least one online class, depending on the type and location of college. This development creates opportunities for servicing a wider audience of students by alleviating obstacles such as geographic location and time zones.

Although self reported data has found no difference in cheating between online and traditional format tests (Watson and Sotile, 2010), questions about online course integrity and student accountability have recently surfaced, given the relative ease of cheating by substituting another person for tests or using resources during a test that are explicitly prohibited. The purpose of this study was to examine emerging learning environments including the likelihood of cheating in online courses and the effect of proctoring on cheating.

Methods

Four instructors agreed to use proctoring software for sections of the same course using common exams that apply concepts from WCET's best practice for online education, including timed tests, random items from a common question pool, and random responses. Three instructors selected a few of the tests to be proctored using Software Secure, a remote proctoring software that uses videotaping of both the student in their surroundings, as well as videotaping their desktop during the test. One instructor had all of the tests proctored using Respondus Lockdown Browser and Monitor, which utilizes both locking down the browser and videotaping the student taking the test in their surroundings.

Following the completion of the tests, videos from Software Secure are reviewed by the company to detect rules violations or suspicious activity, indicating a breach of academic integrity. The instructor for the course is given feedback of the review and can watch the videos at each point of potential breach. Lockdown Browser/Monitor generates brief, randomly timed thumbnails of the full video recording that can be reviewed and flagged by the instructor for potential violations. Five instructors did not use the software and therefore all of those tests were not Proctored.

Not all tests were proctored, so the comparison set up was as follows:
Test 1 - 1 test (n=14) was Proctored; 8 tests (n=148) were Unproctored
Test 2 - 3 tests (n=48) were Proctored; 6 tests (n=109) were Unproctored
Test 3 - 2 tests (n=31) were Proctored; 7 tests (n=129) were Unproctored
Test 4 - 2 tests (32) were Proctored; 7 tests (n=130) were Unproctored

Students in all nine sections were informed that tests were to be taken by themselves with no notes or other resources allowed during the test. All tests were timed with a range of 15 min - 60 min. Tests were not identical, but covered similar material, and questions were randomly drawn from a shared question bank. Student enrollments were tracked in all sections.

Results

Scores on Proctored tests were significantly lower by nearly 15 points compared with scores on the Unproctored tests. The total mean score on all Proctored and Unproctored tests averaged 74.6% and 89.1%, respectively. This represents the difference between a grade of approximately C for the Proctored tests and A- for the Unproctored tests on virtually every test comparison.

Within sections, when all tests were Unproctored, mean scores were higher and ranged from 77.6 - 96.9 compared with Proctored tests, with mean scores 15 points lower, ranging from 68.7 - 91.3. Within the same class, many individual test scores had grades of A or B when tests were Unproctored and C or D when tests were Proctored.

Only 7 of the 100 (7%) students initially enrolled in sections with Unproctored tests dropped the class compared with 15 of 80 (19%) students initially enrolled in sections with Proctored tests who dropped. Despite several suspicious cases, the Proctoring software accurately identified only one student who committed academic dishonesty during a test. Conclusions

These results indicate grade disparity when comparing students within the same class that took Proctored and Unproctored tests as well as between sections of the same course where some sections used test Proctoring and others did not. Mean test scores that were Proctored were always lower (by 15 points) than mean test scores that were Unproctored. Furthermore, attrition was more than double in students enrolled in Proctored vs. Unproctored tests. Use of Proctoring software appeared to prevent academic dishonesty during test taking, as only one student was identified as using resources that were prohibited during the test. Evidence is provided that Proctoring software influences test scores and instructors should consider using Proctoring software in online courses to promote student accountability and integrity.

References:
Watson, G. & Sottile, J. (2010). Cheating in the digital age: Do students cheat more in online courses? Online Journal of Distance Learning Administration, 13(1),.

Lead Presenter

Helaine Alessio is a Professor and Chair in the Department of Kinesiology and Health. She is a Fellow of the American College of Sports Medicine and Scripps Gerontology. Her research has been funded by the National Institutes of Health, National Institute on Aging, and she is active in several research projects that use human and animal models. Her teaching includes online, hybrid, and traditional teaching methods, and she is actively involved in several STEM projects in the Center for Excellence in Teaching at Miami University.