I don't give out 100%s

Jack21222

So this semester as a teaching assistant, I've decided that I simply do not give out 100%s on lab reports. On quizzes and homeworks I'm happy to give out perfect scores, but on something like a lab report, I have determined that nobody is perfect and there is always something that could have been done better.

As a concrete example, I could find nothing obvious wrong with a lab report last week, but his conclusion was a bit too long-winded and some of the stuff presented there should have been in the analysis, so I took off a point. He got a 99%.

I feel that if somebody received a 100%, they stop trying to improve and perhaps get complacent. If somebody gets a 99%, they have a better chance at trying to figure out ways to improve. Meanwhile, the 1 percentage point difference on their lab grade is exceedingly unlikely to affect their letter grade, particularly if they're good enough to be getting 99%s on a regular basis.

Does anybody have any thoughts about this? I can see the argument that perhaps it's unfair to take a point off from one person's paper that I don't take off from a paper that got, say, an 80%. But in my mind, the person who got an 80% has enough major stuff to work on without getting nitpicky.

Are there any other arguments against this? Will this actual reduce the morale of higher-achieving students? Is anybody aware of any physics education research about this tactic? Has anybody used this to good effect?

micromass

Many teachers of mine never gave out a 100% for exactly the same reason, but I never really liked the system.

I feel that beforehand, you should set some criteria about what you want to see on an assignment and then you should grade according to that criteria. If somebody meets all criteria, then I don't see a reason not to give him 100%.

If I grade papers that students need to write, then I always keep in mind the eventual goal: the students are learning this stuff in order to write good research papers. If I read the paper of a student and if I think that it is the quality of a good research paper, then I feel that the student has met the goal and he should get a 100%.

Besides, getting a 100% as a student can be an extremely good motivator. I don't think students are likely to start slacking off or stop improving. On the contrary, I think that they want to keep repeating their success and they will put in even more time in writing a good assignment (at least, that is what I would do).

I'm not saying a 100% should be easy to get, on the contrary: a student would have to work very hard to get it. But I disagree that it should be impossible.

Number Nine

You should have clear guidelines about what makes for a good lab report, and the grades should correspond to how well the students adhered to those guidelines. Docking someone marks because they could conceivably have done better maybe somehow doesn't seem like a good practice.

Jack21222

Thing is, so many of the criteria are subjective. There is a grading breakdown in the beginning of the lab manual which I have to follow, but it has things like this:

Things like "neatness and organization" of the lab report is extremely subjective. When I'm grading, I have a list of key points I look for in every lab report, each worth between 3-6 points. If somebody loses points on a few of them, I don't subject the paper to any more scrutiny, I just give them their 88% or whatever they earned and move on. Maybe I should have taken off a couple more points because in addition to being wrong or incomplete, their paper wasn't quite organized properly. But, it's more important for them to get the major points right and it's less time-consuming for me.

On the other hand, if a paper survives my "first pass" where I am specifically looking for my list of key points, I'll re-read it to make sure I haven't missed anything. If it survives the second pass, I give it a 3rd read to find something they could do better next time, and that's when they end up with a 99%. I'm sure at some point, if everything really was perfect, I'd give up and give a 100%.

So I guess the way I grade, I make it really difficult to get a very poor grade. As long as they cover most the main points adequately, they get a high grade. The average grade I gave this semester was 86%. Out of the 160 papers graded, I've only given a 99% 3 times, and on one of them the point was taken off on a prelab question that was wrong. In reality, I should probably be harder on those people getting in the low 80s, because they're doing a lot of things I could nitpick for points (weak introductions and conclusions, mainly), but I want them to focus on actually doing the analysis properly.

So, it isn't a situation that happens often, but I am still interested in hearing more peoples' opinions.

For me, I didn't mind getting nitpicked for points because I saw it as a motivator. If I got a 100% on something, I'd pat myself on the back and move on. If I got anything less, I'd try to learn something. I guess everybody's different.

pongo38

What students need (but may not want) is proper constructive feedback other than a grade.

Jorriss

I have graded 6 labs (so 108 lab reports) this quarter so far and given out two 100%'s. They were not perfect but they were exceptional by the standards of a freshman course so I feel comfortable with giving a perfect score.

EricVT

Consistency of grading was very important to me as a student. If I'm docked a point for something, then I expect every other student to be held to the same standard. Of course I would never ask that other student's grades be lowered for any reason, but if I found out that I was being penalized for something that other students weren't being penalized for (even if my overall grade was higher), then I would take issue with that.

I also agree that perfect scores (for students who deserve them) are motivators for continued hard work, not motivation to suddenly become complacent.

mfb

If you never give 100%, students cannot improve compared to a 99%-grade, at least in your grading system. Where is the difference to 100%-scores then?
In addition, every grade has some (very fine) internal structure - there might be a "good 99%" and a "bad 99%" (which is very similar to the former one). So where is the problem with a "bad 100%", which is better than a "good 99%", but not perfect?

DaleSpam

IMO, grades are a measuring device to determine how well a student has learned the material. If any student gets a 100 then your assignment is too easy and it was not a good measurement since the quantity measured exceeded the range of the measurement instrument. Similarly, an average score should be about 50.

This may be more appropriate for tests than for homework.

doppelganger

I find this attitude ridiculous and patronizing. You are basing your grade on how you 'feel' a student would react to that grade (without any real proof for your feeling) rather than on the quality of the student's work. I despise instructors who carry that type of attitude.

Pengwuino

When I was a student doing lab reports (and I feel MOST students think this way), anything greater than a 95% was pretty much the same thing. No one ever feels challenged to do better for that extra meaningless 1%. We would always just call it nit-picking and feel that the TA is just being a jerk instead of being motivated to get that extra 1%.

I had a professor who went a bit more extreme in a way that was far more effective. He wouldn't give 99%s or even 95%s. You had to go above and beyond what is normally expected out of an undergrad at the level of that class in order to even get an A on a lab. When your actual grade can be affected, THEN you see students get serious about perfecting a lab report. Of course, you need to be willing to give out the >95%s or whatever as well otherwise they will assume that you just don't want to give out the best scores and that it's meaningless to even try to get an A/A+.

Remember, there certainly are students who will strive for perfection, but all students are under time constraints and any bright enough student will realize that the extra work for the extra 1% is not going to be worth the time especially when they realize that they can't actually get that extra 1% anyways (and they figure this out very quickly).

Jack21222

The problem with this is that your two statements are contradictory. You wouldn't want me to lower everybody else's grade, but you also want them to be penalized for the same mistakes. You can't have both.

If I've killed somebody's grade over not answering the questions properly, and they're getting a 60%, it would be cruel to continue to beat them down by nickle and diming their grade with the small things, like not putting enough detail or background information into their introduction.

If you get a good grade on an assignment, it's petty to "take issue with" the teacher giving out "pity points" to somebody who is struggling.

I must say, though, this thread has made me reconsider giving out 100%s. Perhaps I can still comment on papers on things that need to be improved without taking points off for it.

And you're getting very angry over 1 percentage point on rare assignments for students who are getting an A no matter what. The feeling is mutual.

doppelganger

I am not angry about it, but I do have strong feelings on this topic. I have been subjected to this type of grading system many times in the past.

I find it ironic that you feel it is petty for others to take issue with pity points you might have given to poor students, but not petty for you to take away 'perfection' points from exceptional students.
This would be my preferred method. The internet works!

MarneMath

So, I used to be a TA for the most subjective class in the world, a writing class!

While I agree that, in theory, a person can always present information better, I think it's asinine to form a grading policy on this. In the real world, which I hope is something we are preparing students for, there is balance between perfection and time. No one wants to put in the extra work, if in the end, the person reading the material will just nit pick it, and to the same extent, you can't expect a person to not be able to do something better.

My approach when it came to grading was simple. On the first assignment, I laid out my grading criteria. If they met the criteria, but I found errors or had tips for them, I would make note of it in my massive journal, and leave them a note on their paper explaining what I expect them to improve upon. If they failed to improve, then I would mark off on the assignment, but if they did improve on it, I see no reason why I should still punish them.

EricVT

What I meant is that I wouldn't ask that another student's grade be lowered retroactively once I found out about a grading inconsistency, there is no contradiction. And I don't consider it petty in the least to expect consistent grading from an instructor.

Jack21222

Let me give you a concrete example, and let's see if you still have this sentiment.

A student has a two sentence introduction, where they give only the most barebones information about what is to be accomplished in this experiment. This introduction should really get about a 6/10. The rest of the paper is riddled with errors and is in many places incomplete. The student would get a 68%. Well, in comparison with the rest of the paper, the introduction isn't that bad, so I give an 8/10 on the intro, and the student gets a 70%. A second student turns in a paper with a very similar introduction, but the rest of the paper is nearly flawless. I give that student 6/10 on the intro for a 96% total, because that's what the intro deserves.

Your desire for "consistent grading" is the difference between the weaker student getting a C- and a D+.

What you fail to realize is that grading is inherently subjective. Lab reports aren't multiple choice; you can't just flatly says "this is right" and "this is wrong" and apply it evenly for every student. Specifically, in the case of these lab reports, "neatness and organization" are right there in the rubric. Well, organization is a very subjective thing. If you put information which should be in the introduction in the conclusion instead, to me, that's disorganized. However, if I spent all my time tracking down every bit of information that should be in another section and docking points for it, I'd triple my grading time and I'd only hurt the students who were on the borderline anyway. However, in a perfect world, such disorganization SHOULD have points taken off, so I see nothing wrong for docking points where it's warranted even if I haven't docked points from the weaker students who made the same mistakes. Yes, it's inconsistent, but it only raises the grades of the weaker students, it does not lower the grades of the stronger ones.

EricVT

You've just said that you would give 80% credit for a 60% quality answer to one student and 60% credit for an identical 60% quality answer to another student, arbitrarily choosing to give the better student less points because they did so well on the rest of the paper.

I just hope you understand how the higher scoring student might not appreciate being held to a different grading standard than others.