If you’ve tried to use “Forced Completion” in the test options in Blackboard recently you may have noticed a new warning. Of all the features in Blackboard, the forced completion setting for test/quizzes causes the most headaches for students, instructors, and system administrators. This seemingly innocent setting can cause problems should anything interrupt the session. In the end it can result in more cheating than it prevents because it provides a plausible reason for needing a test/quiz reset – my computer froze or my Internet connection dropped out.

Forced completion requires students complete a test/quiz in a single session. On the surface this setting seems reasonable. If an assessment is low stakes you may want to prevent students from focusing too much energy on a single item. Or you may want to limit how long the student takes on a longer tests. Although these intentions are good the way the tool works will cause problems if a connection is lost or reset. Leaving the test for any reason ends the student’s session and they can’t get back in to complete the test.

What can you do?

Set a time limit but do not select forced completion. The Blackboard timer does not stop if the student exits and returns to the test. If a student does have a problem they can get right back in. This students time should not exceed the time limit or if it does it might be by a few minutes – the time it took to reconnect. But if a student enters an exam, looks at the questions, exits to find answers, and reenters, the timer still runs and you can see how long the student took from initially clicking the link to the test until they finally submit the test. So if a student clicks on the link for a test with a one hour time limit at 2 PM on Wednesday afternoon and looks at the questions then exits and looks up all the answers and reenters the test at 8 PM to answer the questions and finishes at 8:30 PM, their total elapsed time is 6.5 hours – well over the 1 hour limit. You can now take appropriate action.

Require students to report any problems taking the test to you immediately. Then if a student requests a test reset, you can see from the test access log when the student started the test, how long they spent in the test before the “problem” occurred, what questions they accessed, and which questions were saved. Based on what you see you can  decide whether you will honor the request or if a penalty should be applied. If a student reports a problem and goes over the time limit by only a few minutes, you’ll see evidence to back that claim up.

When setting time limits, there is an option to have the test auto-submit when the time limit is reached. A caution here is that if a student has had a problem, their test will submit when the time from first access is reached and they will not have any additional time to make up for having to reenter the test. This setting can also be a problem for students receiving additional time on tests as an accommodation for a disability. For those students you will need to set up testing exceptions to give them additional time.

Create pools of questions and let students have multiple attempts. You can choose to average the results, or have the highest grade count. Because the test is pulled from a pool or pools of questions, they will see different questions each time they take the test. (See related post on pool size and question repeats.) So even if they look up the answers from the first attempt, because the questions on the second attempt will be different, students don’t gain much by taking the test again.

There are ways to reduce cheating without using forced completion. Take advantage of the tools that can help you manage the normal and reasonable problems students might encounter without being overly restrictive.