European universities are running the so called "Bologna plan" which includes the recommendation of keeping some percentage of the final grades to be filled with various activities during the lessons period. Like everything in life, it has pros and cons.
I will tell you the cons I have seen (in the particular case of mathematics related students), then I will appreciate a list of pros (because I don't see any of them).
- The world isn't perfect, so, although mathematics is really time consuming, there are people who work while they study, so they physically can't attend classes, which usually is one of the activities taken into account directly (signature lists) or indirectly (correcting problems in the blackboard). So they begin with a clear disadvantage compared to their classmates (they de facto lose a fraction of their grades).
- Again, the world is not perfect, there are teachers whose lessons simply aren't worth, and, for sure, a student who had spent the lesson time in the library reading a book about the subject will know more than his lesson-attending mates, and, probably, get a worse grade, despite having a way better done exam. In some (but not infrequent) cases passing the exam without attending lessons is formally impossible.
- Over again, the world isn't a wonderland, and every student has lots of subjects, every one of them with lots of continuous evaluation stuff (mandatory lesson attending, non-liberating partial exams, sets handing over,...), so, students tend (and sometimes are kind of forced) to do the work of each subject as fast as they can without really understanding what they are doing (really sad, even more in math-related stuff). I think university is about learning, not about getting constantly evaluated. With continuous evaluation student have $n$ weeks getting evaluated and $0$ weeks learning while the ideal will be $n$ weeks learning and maybe $3$ weeks getting evaluated.
- I have heard that continuous evaluation produces better overall results. I must disagree. I have seen really brilliant students getting a $7$ out of $10$ as a final grade with a perfect final exam, as well as people who really doesn't understand the subject getting a $6$ (pass) while failing the final exam with a $3$. Also I have seen people with a good exam $6$ or $7$ out of $10$ having failed the subject with a $3$ or a $4$. Is this fair?
- In general, this dedicated fraction of the grading (usually around a 30 %) is kept for the extraordinary calls. ¿Isn't it nonsense?
I know some students like continuous evaluation, some of them because it allows them to pass the subject without really understanding it deeply, some others because they say it helps them to study. So, why not let students decide? If some of them don't want this system, just evaluate them by the final exam while the system keeps running for the others. Maybe another solution is to make a good use of the $\max$ function?