4 ms·
Beating cheating
- diego 16y agoOr you can just have one-on-one oral examinations. Nah, that's a lot of work.
- popinyay 16y agoCall it a chat, then. Chatting is perhaps the fastest way to get to know someone and discover what they're interested in. And working with them is the best way to see if they're up to the job. Testing is unpleasant and it doesn't work anyhow, IMO.
- yummyfajitas 16y agoChatting is also a great way to introduce subjectivity into the examination and provide an advantage to the students you like the best.
- Xk 16y agoAnd short of "fill in the bubble", any written exam also provides an advantage to the students you like the best.
- drdaeman 16y agoTesting does work, it's just that the results are of lesser quality. The simpler the testing procedure the less it's error-prone. Edit: However, there are edge cases where testing could provide more accurate results - for example, a socially-inept person, may be too nervous to talk in person, but could pass the test with a good results.
- popinyay 16y agoI don't think testing can measure depth of knowledge, character, creativity or enthusiasm. The reason being that these things simply aren't measurable. They don't come in units. Depth of knowledge, for example, depends on the connections one forms with other areas of knowledge.
- greenyoda 16y agoI'm a bit disturbed that students could be accused of cheating, and their academic careers potentially ruined, because of events that could happen purely by chance. For example, let's say I take an exam and get the same two wrong answers as the person sitting next to me. While the probability of this happening is small, it's not zero. With millions of students taking standardized tests, it would be strange if it didn't sometimes happen. Just like some percentage of the time, you're going to toss a coin and get 20 heads in a row.
- pjscott 16y agoI wonder how it compares to false positive rates in other, more conventional methods of detecting cheating. There might be a status quo bias at work here.