Skip to main content
Mathematics Educators

Return to Answer

Fix typo ChatGBT -> ChatGPT
Source Link
Daniel R. Collins
Daniel R. Collins
  • 26.9k
  • 75
  • 129

You may be trying to band-aid a missing leg.

As another answer mentions, the internet provides a number of means to easily cheat fairly easily. You can take some measures to stop obvious cases of cheating (and you should take some such measures), but there's only so much you can do.

As fedja puts it, the job of teachers is "to provide an opportunity to learn to everybody who is willing to learn".

My conclusion is slightly different, however: you shouldn't necessarily based grades (almost) fully on proctored exams. A student can have a moderately decent understanding of a topic and still not perform all that well in an exam due to the time constraint, due to stress, due to not having memorisedmemorized the appropriate formulas, because they were having a bad day or week, etc.

If it weren't for the possibility of cheating, I'd say the best grading scheme is likely mostly one without exams (doctors, for example, probably still need surgical exams). There are also measures to focus on testing whether students actually understand the material rather than testing whether they were able to e.g. memorisememorize a bunch of things that they can recall within a short timeframe, and that they can write quickly enough, but I won't go into that here.

For different courses (even within maths), it may be more or less easy to cheat on homework (with ChatGBTChatGPT or through other means).

My suggestion is to possibly reduce the weight of homework so having a degree continues to actually mean something for what you know (even if it already often means very little). But also don't go to to extreme measures to avoid having someone pass mostly through cheating: if a student cheats, they're shooting themselves in the foot by squandering what's hopefully a good opportunity for them to learn things they'll use later in life. Don't then shoot other students in the foot to put them on equal footing with the cheater.

I can't, however, tell you how exactly you should weight homework, quizzes, bigger tests, projects and exams to come up with a final grade. That's something you'll need to figure out based on all the factors involved.

One option to consider is to have an overall grade that's calculated in some way (which has a minimal passing grade), but then to also have a minimal grade for the exam only, which is lower than the overall passing grade.

Or one could combine homework, etc. with exams in some variable way. You could, for example, say that the maximum distance between the two is 20%, and the exam mark will drag the homework mark down, or raise it up, if the distance exceeds 20%. The downside here would be that the grading would be less obvious to students (even if you're transparent about how it's weighted, which you should be), and this may make it feel unfair.

You may be trying to band-aid a missing leg.

As another answer mentions, the internet provides a number of means to easily cheat fairly easily. You can take some measures to stop obvious cases of cheating (and you should take some such measures), but there's only so much you can do.

As fedja puts it, the job of teachers is "to provide an opportunity to learn to everybody who is willing to learn".

My conclusion is slightly different, however: you shouldn't necessarily based grades (almost) fully on proctored exams. A student can have a moderately decent understanding of a topic and still not perform all that well in an exam due to the time constraint, due to stress, due to not having memorised the appropriate formulas, because they were having a bad day or week, etc.

If it weren't for the possibility of cheating, I'd say the best grading scheme is likely mostly one without exams (doctors, for example, probably still need surgical exams). There are also measures to focus on testing whether students actually understand the material rather than testing whether they were able to e.g. memorise a bunch of things that they can recall within a short timeframe, and that they can write quickly enough, but I won't go into that here.

For different courses (even within maths), it may be more or less easy to cheat on homework (with ChatGBT or through other means).

My suggestion is to possibly reduce the weight of homework so having a degree continues to actually mean something for what you know (even if it already often means very little). But also don't go to to extreme measures to avoid having someone pass mostly through cheating: if a student cheats, they're shooting themselves in the foot by squandering what's hopefully a good opportunity for them to learn things they'll use later in life. Don't then shoot other students in the foot to put them on equal footing with the cheater.

I can't, however, tell you how exactly you should weight homework, quizzes, bigger tests, projects and exams to come up with a final grade. That's something you'll need to figure out based on all the factors involved.

One option to consider is to have an overall grade that's calculated in some way (which has a minimal passing grade), but then to also have a minimal grade for the exam only, which is lower than the overall passing grade.

Or one could combine homework, etc. with exams in some variable way. You could, for example, say that the maximum distance between the two is 20%, and the exam mark will drag the homework mark down, or raise it up, if the distance exceeds 20%. The downside here would be that the grading would be less obvious to students (even if you're transparent about how it's weighted, which you should be), and this may make it feel unfair.

You may be trying to band-aid a missing leg.

As another answer mentions, the internet provides a number of means to easily cheat fairly easily. You can take some measures to stop obvious cases of cheating (and you should take some such measures), but there's only so much you can do.

As fedja puts it, the job of teachers is "to provide an opportunity to learn to everybody who is willing to learn".

My conclusion is slightly different, however: you shouldn't necessarily based grades (almost) fully on proctored exams. A student can have a moderately decent understanding of a topic and still not perform all that well in an exam due to the time constraint, due to stress, due to not having memorized the appropriate formulas, because they were having a bad day or week, etc.

If it weren't for the possibility of cheating, I'd say the best grading scheme is likely mostly one without exams (doctors, for example, probably still need surgical exams). There are also measures to focus on testing whether students actually understand the material rather than testing whether they were able to e.g. memorize a bunch of things that they can recall within a short timeframe, and that they can write quickly enough, but I won't go into that here.

For different courses (even within maths), it may be more or less easy to cheat on homework (with ChatGPT or through other means).

My suggestion is to possibly reduce the weight of homework so having a degree continues to actually mean something for what you know (even if it already often means very little). But also don't go to to extreme measures to avoid having someone pass mostly through cheating: if a student cheats, they're shooting themselves in the foot by squandering what's hopefully a good opportunity for them to learn things they'll use later in life. Don't then shoot other students in the foot to put them on equal footing with the cheater.

I can't, however, tell you how exactly you should weight homework, quizzes, bigger tests, projects and exams to come up with a final grade. That's something you'll need to figure out based on all the factors involved.

One option to consider is to have an overall grade that's calculated in some way (which has a minimal passing grade), but then to also have a minimal grade for the exam only, which is lower than the overall passing grade.

Or one could combine homework, etc. with exams in some variable way. You could, for example, say that the maximum distance between the two is 20%, and the exam mark will drag the homework mark down, or raise it up, if the distance exceeds 20%. The downside here would be that the grading would be less obvious to students (even if you're transparent about how it's weighted, which you should be), and this may make it feel unfair.

added 4 characters in body
Source Link
NotThatGuy
NotThatGuy
  • 111
  • 3

You may be trying to band-aid a missing leg.

As another answer mentions, the internet provides a number of means to easily cheat fairly easily. You can take some measures to stop obvious cases of cheating (and you should take some such measures), but there's only so much you can do.

As fedja puts it, the job of teachers is "to provide an opportunity to learn to everybody who is willing to learn".

My conclusion is slightly different, however: you shouldn't necessarily based grades (almost) fully on proctored exams. A student can have a moderately decent understanding of a topic and still not perform all that well in an exam due to the time constraint, due to stress, due to not having memorised the appropriate formulas, because they were having a bad day or week, etc.

If it weren't for the possibility of cheating, I'd say the best grading scheme is likely mostly one without exams (doctors, for example, probably still need surgical exams). There are also measures to focus on testing whether students actually understand the material rather than testing whether they were able to e.g. memorise a bunch of things that they can recall within a short timeframe, and that they can write quickly enough, but I won't go into that here.

For different courses (even within maths), it may be more or less easy to cheat on homework (with ChatGBT or through other means).

My suggestionMy suggestion is to possibly reduce the weight of homework so having a degree continues to actually mean something for what you know (even if it already often means very little). But also don't go to to extreme measures to avoid having someone pass mostly through cheating: if a student cheats, they're shooting themselves in the foot by squandering what's hopefully a good opportunity for them to learn things they'll use later in life. Don't then shoot other students in the foot to put them on equal footing with the cheater.

I can't, however, tell you how exactly you should weight homework, quizzes, bigger tests, projects and exams to come up with a final grade. That's something you'll need to figure out based on all the factors involved.

One option to consider is to have an overall grade that's calculated in some way (which has a minimal passing grade), but then to also have a minimal grade for the exam only, which is lower than the overall passing grade.

Or one could combine homework, etc. with exams in some variable way. You could, for example, say that the maximum distance between the two is 20%, and the exam mark will drag the homework mark down, or raise it up, if the distance exceeds 20%. The downside here would be that the grading would be less obvious to students (even if you're transparent about how it's weighted, which you should be), and this may make it feel unfair.

You may be trying to band-aid a missing leg.

As another answer mentions, the internet provides a number of means to easily cheat fairly easily. You can take some measures to stop obvious cases of cheating (and you should take some such measures), but there's only so much you can do.

As fedja puts it, the job of teachers is "to provide an opportunity to learn to everybody who is willing to learn".

My conclusion is slightly different, however: you shouldn't necessarily based grades (almost) fully on proctored exams. A student can have a moderately decent understanding of a topic and still not perform all that well in an exam due to the time constraint, due to stress, due to not having memorised the appropriate formulas, because they were having a bad day or week, etc.

If it weren't for the possibility of cheating, I'd say the best grading scheme is likely mostly one without exams (doctors, for example, probably still need surgical exams). There are also measures to focus on testing whether students actually understand the material rather than testing whether they were able to e.g. memorise a bunch of things that they can recall within a short timeframe, and that they can write quickly enough, but I won't go into that here.

For different courses (even within maths), it may be more or less easy to cheat on homework (with ChatGBT or through other means).

My suggestion is to possibly reduce the weight of homework so having a degree continues to actually mean something for what you know (even if it already often means very little). But also don't go to to extreme measures to avoid having someone pass mostly through cheating: if a student cheats, they're shooting themselves in the foot by squandering what's hopefully a good opportunity for them to learn things they'll use later in life. Don't then shoot other students in the foot to put them on equal footing with the cheater.

I can't, however, tell you how exactly you should weight homework, quizzes, bigger tests and exams to come up with a final grade. That's something you'll need to figure out based on all the factors involved.

One option to consider is to have an overall grade that's calculated in some way (which has a minimal passing grade), but then to also have a minimal grade for the exam only, which is lower than the overall passing grade.

Or one could combine homework, etc. with exams in some variable way. You could, for example, say that the maximum distance between the two is 20%, and the exam mark will drag the homework mark down, or raise it up, if the distance exceeds 20%. The downside here would be that the grading would be less obvious to students (even if you're transparent about how it's weighted, which you should be), and this may make it feel unfair.

You may be trying to band-aid a missing leg.

As another answer mentions, the internet provides a number of means to easily cheat fairly easily. You can take some measures to stop obvious cases of cheating (and you should take some such measures), but there's only so much you can do.

As fedja puts it, the job of teachers is "to provide an opportunity to learn to everybody who is willing to learn".

My conclusion is slightly different, however: you shouldn't necessarily based grades (almost) fully on proctored exams. A student can have a moderately decent understanding of a topic and still not perform all that well in an exam due to the time constraint, due to stress, due to not having memorised the appropriate formulas, because they were having a bad day or week, etc.

If it weren't for the possibility of cheating, I'd say the best grading scheme is likely mostly one without exams (doctors, for example, probably still need surgical exams). There are also measures to focus on testing whether students actually understand the material rather than testing whether they were able to e.g. memorise a bunch of things that they can recall within a short timeframe, and that they can write quickly enough, but I won't go into that here.

For different courses (even within maths), it may be more or less easy to cheat on homework (with ChatGBT or through other means).

My suggestion is to possibly reduce the weight of homework so having a degree continues to actually mean something for what you know (even if it already often means very little). But also don't go to to extreme measures to avoid having someone pass mostly through cheating: if a student cheats, they're shooting themselves in the foot by squandering what's hopefully a good opportunity for them to learn things they'll use later in life. Don't then shoot other students in the foot to put them on equal footing with the cheater.

I can't, however, tell you how exactly you should weight homework, quizzes, bigger tests, projects and exams to come up with a final grade. That's something you'll need to figure out based on all the factors involved.

One option to consider is to have an overall grade that's calculated in some way (which has a minimal passing grade), but then to also have a minimal grade for the exam only, which is lower than the overall passing grade.

Or one could combine homework, etc. with exams in some variable way. You could, for example, say that the maximum distance between the two is 20%, and the exam mark will drag the homework mark down, or raise it up, if the distance exceeds 20%. The downside here would be that the grading would be less obvious to students (even if you're transparent about how it's weighted, which you should be), and this may make it feel unfair.

Source Link
NotThatGuy
NotThatGuy
  • 111
  • 3

You may be trying to band-aid a missing leg.

As another answer mentions, the internet provides a number of means to easily cheat fairly easily. You can take some measures to stop obvious cases of cheating (and you should take some such measures), but there's only so much you can do.

As fedja puts it, the job of teachers is "to provide an opportunity to learn to everybody who is willing to learn".

My conclusion is slightly different, however: you shouldn't necessarily based grades (almost) fully on proctored exams. A student can have a moderately decent understanding of a topic and still not perform all that well in an exam due to the time constraint, due to stress, due to not having memorised the appropriate formulas, because they were having a bad day or week, etc.

If it weren't for the possibility of cheating, I'd say the best grading scheme is likely mostly one without exams (doctors, for example, probably still need surgical exams). There are also measures to focus on testing whether students actually understand the material rather than testing whether they were able to e.g. memorise a bunch of things that they can recall within a short timeframe, and that they can write quickly enough, but I won't go into that here.

For different courses (even within maths), it may be more or less easy to cheat on homework (with ChatGBT or through other means).

My suggestion is to possibly reduce the weight of homework so having a degree continues to actually mean something for what you know (even if it already often means very little). But also don't go to to extreme measures to avoid having someone pass mostly through cheating: if a student cheats, they're shooting themselves in the foot by squandering what's hopefully a good opportunity for them to learn things they'll use later in life. Don't then shoot other students in the foot to put them on equal footing with the cheater.

I can't, however, tell you how exactly you should weight homework, quizzes, bigger tests and exams to come up with a final grade. That's something you'll need to figure out based on all the factors involved.

One option to consider is to have an overall grade that's calculated in some way (which has a minimal passing grade), but then to also have a minimal grade for the exam only, which is lower than the overall passing grade.

Or one could combine homework, etc. with exams in some variable way. You could, for example, say that the maximum distance between the two is 20%, and the exam mark will drag the homework mark down, or raise it up, if the distance exceeds 20%. The downside here would be that the grading would be less obvious to students (even if you're transparent about how it's weighted, which you should be), and this may make it feel unfair.