It seems to be pretty well established that the manner in which information is digested has an impact on both short term and long term retention of the material.
For example, people who study in short bursts with frequent breaks in between retain more information than people who sit down for a 2 hour uninterrupted chunk of studying. The same seems to hold true of people who do a little review each day throughout the semester (as opposed to people who return to their notes the night before the exam).
With this in mind imagine two students who have tests and papers that average out to a 95 A. However, also imagine that one student was always in attendance. He never missed a day, but the other student missed half of the class.
If you had to bet on which one would be more likely to retain the information long-term, knowing the positive impact that slowly digesting information has on long-term retention, the smart money is on the student who attended class each day and slowly digested the material over the course of the semester.
Now to connect this to the final grade. One goal you might have in a class is that students master the material to the degree by the end of the course. You would test for that level of mastery with exams and papers. However, a lot of professors have as a goal that students retain some level of mastery of the material beyond the course. They have long-term retention as a goal.
One primary purpose of a final grade is to be a measure of the degree to which students have, in the professors well-reasoned opinion, achieved the goals that the professor had for the students. If long-term retention is a reasonable goal, and regular attendance is a reliable guide to long-term retention, then it is perfectly reasonable to have absences be reflected to some degree in the final grade.
The interesting thing about this defense is directly ties attendance to the thing that grades should reflect. Students sometimes seem to think that the only justification for enforcing an attendence policy is some kind of overly paternalistic attitude toward the students. But notice that this defense is not paternalistic at all because it is justified by the fact that attendance is directly correlated to something should (on the assumption that long term retention is a reasonable goal) be taken into account come evaluation time.
I’m afraid this reasoning is patently invalid.
If the student goes to class, then he is likely to have long-term retention
The student did not go to class
Therefore, the student did not have long-term retention. (ergo, deserves a lower grade)
You’re still reasoning that you know better than the student what will best lead to long-term retention, and if he’s not following your method, then he’s not achieving the goals of the class. That’s clearly paternalistic. (Though the question of whether a professor should have this attitude towards students is perhaps still an open question)
You’ve extracted my argument incorrectly.
I argue that failure to attend regularly is likely to result in worse long-term retention, and as a result it is reasonable to believe that a student who doesn’t attend as a lower probability of retaining material long-term than had he attended. If you keep the “likely” in the the conclusion, then the argument is valid.
Why is it permissible to have exams and papers included in the final grade? Because they are indicators that students have likely achieved some part of the course goals. Good grades on exams and papers don’t entail that students have achieved course goals. At best, good grades are reliable, but of course fallible, indicators that students have achieved course goals. However, it is perfectly permissible to use these as measures.
That’s all I’m saying about attendance. It is a reliable, but fallible, indicator of achievement of a reasonable course goal. And so, it reasonable to take it into account.
>It is a reliable, but fallible, indicator of achievement of a reasonable course goal. And so, it reasonable to take it into account.
What if it turned out that blonde-haired students are more likely to have long-term retention, significantly enough that it can be used as an indicator of achievement? Then, would it be reasonable to grade blonde-haired students higher than brunettes? (include a ‘hair color’ factor in the final grade)
Regarding the argument form, I’m afraid keeping ‘likely’ in the conclusion doesn’t help it.
If I’m eating a taco, I’m likely hungry
I’m not eating a taco
Therefore, I’m likely not hungry
I’m sorry. I missed the negation in the conclusion, but now that I see the negation the argument seems even more uncharitable. Just so we’re clear.
1. The student did not attend class regularly.
2. If the student did not attend class regularly, then it he is less likely to have long-term retention than he would have if he had attended class regularly.
3. Therefore, the student is less likely to have long term retention than he would have if he had attended class regularly.
The blonde point is interesting. I’ll get back to you about that.
Your premise 2 is not obvious to me. It seems like you made a good case that going to class can aid in long-term retention, but did not show that the student was not doing something much more valuable with his time that could do an even better job of encouraging long-term retention. I see no reason to assume time in your class is worthwhile (better than any alternatives available to the student), and assuming you know better than the student is definitely paternalism.
But there is reason. There is the data that the manner in which information is digested can aid in long term retention. The data gives us good prima facie reason to accept the consequent. Absent defeaters it’s reasonable to accept the conclusion.
Unless the student demonstrates that they are taking steps (in their absences) to generate long-term retention. It still seems perfectly reasonable for the professor to take this into account. Which would, perhaps, be one way around this. If a student could demonstrate that they are doing something (in lieu) of attending that makes long term retention just as likely, then the prima facie reasons we have here would be defeated.
But I leave it to you (or the student) to tell me what other methods would be just as likely to increase long-term retention, and how that could be effectively demonstrated. The bar for proof would have to be pretty high. Afterall, the fact that the student attends class is something that I can be highly certain of. The fact that the student does whatever it is that does just as good a job of preserving long-term retention, is something that I think it’s going to be difficult to have as high a degree of certainty has actually obtained. Again, I’m open to suggestions on this front. Both on what this alternative method would be and how a professor could have compelling reasons to believe that the student is actually engaging in that alternative method.
None of this gives any reason to think you’re not being paternalistic. You think you know better than the student what is the best use of his time, and will penalize him for deviating from your instructions, unless he gives you compelling proof that what he’s doing works better than what you want him to do. While it’s perhaps an open question whether professors should be paternalistic towards students, that’s definitely what you’re doing here.
Also notice the disanalogy between this method of grading and tests.
1. Learning the material can cause you to do well on tests.
2. Going to class can cause you to retain the material better.
If you think 1. is a good reason to grade based on tests, then that doesn’t necessarily tell you anything about whether 2. is a good reason to grade based on attendance.
It does if (as I stated earlier) long term retention is a reasonable course goal. Assume (1) is a good reason to factor in exams into a grade. Why? Because our awareness of the causal relationship in (1) makes good performance on an exam evidence that a reasonable course goal has been achieved – namely learning the material.
If long-term retention is a reasonable course goal to have for students, then (2) is a good reason to factor in attendence for precisely the same reasons that (1) is.
Which is why I don’t understand the paternalism charge. I am factoring in something that is evidence of having acheived a reasonable goal to have for a course (long-term retention). Part of my job, as a professor, is to assign a grade that reflects the degree to which a student has satisfied certain course goals. So, I don’t see it being paternalistic if attendance is as a matter of fact reliable evidence that a certain course goal hasn’t been acheived to the degree that it could have.
Perhaps we mean different things by paternalism. When students decry paternalism with respect to attendance, they usually are objecting to the fact that the professor is penalizing them for “life decisions” that are irrelevant to whether or not they have satisfied certain course goals. If this argument works. I’m not being paternalistic in that sense.
>When students decry paternalism with respect to attendance, they usually are objecting to the fact that the professor is penalizing them for “life decisions” that are irrelevant to whether or not they have satisfied certain course goals.
Ah – I had never encountered that objection. I was using a meaning of paternalism like “Thinking you know better than me what is good for me, and so 1. taking actions to cause me to do what you think is best rather than what I think is best, or 2. penalizing me for doing what I think is best rather than what you think is best.” That’s why I said it’s an open question whether professors should be paternalistic towards students – some would claim that professors do know better than students what will better achieve course goals (I am skeptical of this).
>If long-term retention is a reasonable course goal to have for students, then (2) is a good reason to factor in attendence for precisely the same reasons that (1) is.
You seem to be missing (or just not acknowledging) the disanalogy I pointed out:
In (1), A causing B is considered a good reason to use B as a metric for A.
In (2), A causing B is considered a good reason to use A as a metric for B.
One should not assume that (1) being based on a valid form implies that (2) is based on a valid form. Frankly, I’m more intuitively skeptical of (1) than (2), but that’s beside the point.
Actually, I was aware of the difference in order – it just didn’t seem relevant here in terms of what was evidence for what.
Here are some examples that illustrate the point.
Smoking can cause lung cancer…so the fact that someone smokes increases the probability that they will contract lung cancer. That someone smokes is some evidence that they will contract Lung Cancer. That someone contracted Lung Cancer is also some evidence that they smoked.
There are a lot of cases where the causal relata are tightly connected enough so that what plays the evidential role is the correlation. The correlation can be some evidence (however, of course defeasible) that works both ways.
Actually here’s a clearer example than the smoking example…
I. Bats can cause caves to be covered in guano.
(A causes B)
So,
II. The presence of bats in a cave is evidence that the cave will be covered in guano. (A is evidence of B)
But it also seems plausible to suppose that
III. The fact that a cave is covered in guano is evidence of the presence of bats.
(B is evidence of A)
Granted, if it could be established that an effect had denumerably many causes. Then we might start to doubt the evidential force of a known possible cause. But this would come in as a defeater (in the Pollock, Alston, Bergmann sense – I think)…
So it seems we’re back to grading based on things that are correlated in any way with successful completion of course goals. For example, one whose parents went to college is more likely to succeed in college classes; therefore, whether one’s parents went to college should be a factor in students’ final grades.
More directly related to the example of attendance, we have good reason to believe that philosophy majors are more likely to have long-term retention of the things they learn in a particular philosophy class, since they’ll think about and refer to it later in their academic career. Therefore, students should be penalized for not being philosophy majors.
I generally do have an attendance policy, and for (I believe) a non-paternalistic reason. But my reason is very different: A significant portion of my class time is devoted to student-raised ideas. Each student, by coming to class and participating periodically, creates part of the course content. If a student doesn’t come to class or never participates, then that student is depriving the other students of the benefit of learning from his or her ideas. The trouble is not just that students who don’t come to class predictably learn less of the course material (although that’s certainly true), or that they fail to retain the material after the course ends (though this does sound plausible), but rather that they aren’t contributing to discussions in class, which is a required activity.
In addition to the benefit that the other students derive from a student’s good ideas, doing philosophy is not strictly a matter of writing a good exam or a good paper. I expect students to improve their ability to verbally articulate their arguments, too, and this requires attendance (for them to get practice, and for me to be able to evaluate their skills.
These reasons, of course, would not apply if I were teaching a large, lecture-style class.
Thom,
Thanks for taking the time to weigh in like this. It’s been helpful. This is a good point.
Notice, however, that this strikes me as a difficulty for any kind of grading policy or justifications for most metrics. Most justified metrics are justified because they are evidence of having likely completed course goals, and we can always imagine possible scenarios where some other metric (that seems completely unjustified) would also be evidence of how likely it is that the student has completed course goals.
There are a few things that can be said. Suppose we found that hair color really caused lower IQs and that brunettes in general had lower IQs. One might wonder what a college professor should make of that. One would be that professors are charged with the responsibility of assigning grades based on metrics that (a) are reliable indicators of having acheived course goals, and (b) are directly within the student’s control. This would explain in fact why we tend to not count it against student’s if they could not complete an assignment when factor’s outside their control prevent them from completing an assignment. This would give us some principled reasons to discriminate between metrics. So, retention affecting factors that are within the student’s control could (in a principled way) be held against them.
I have some more thoughts about this. This certain cases, but it doesn’t handle the philosophy major case (nor does it handle other cases e.g., if I know the student engages in other activities outside of class that effect long-term retention like heavy drug use – it would be weird to take off points in the class for heavy drug use…)
The general problem that I think you’ve touched on is that there must be some way to demarcate between permissible and impermissible metrics. I’m reasonably confident that class attendance will fall on the permissible side of that demarcation – but it will be no easy task to do that (of course no demarcation project in philosophy is an easy task)
I’ll get back to this. Right now I’ve gotta run. Thanks again.
Thanks for posting the discussion and responding to my comments – sorry if I come off as a little ‘argumentative’ (to use it colloquially).
I would hazard a guess that the reasons why attendance policies fall under ‘permissible’ would be:
-they’ve been around a long time
-students actually agree to be in class and so you can hold them to it
I doubt there’s any more to it.
I also use attendance in order to grade my students. We are working on group projects in class and only in class. If a student is not there, than they are not there to work or contribute to the project. If they are there and working, than they deserve to receive credit for the time and work that they contribute.