Tales from the latest exam

# There seems to be a negative correlation between how smart students think they are (and how vocal they are about their mad skillz) and how well they do on my exams.

Welcome to the Dunning-Kruger effect.

# Another odd data point: The students who are at the top of the class and blowing away the rest of the class on these tests....are not CS majors.

Maybe they're the ones too smart to go into CS and were going to become investment bankers?

NB: I wasn't one of the smart ones. I have the CS degree, but I spend my time designing at a somewhat lower abstraction level. Funny how things work out.

Jeez, thats sad Jane, I only took 2 programming classes when I was convinced I was going to do Biology for a bit, dropped that habit real fast when I found out how much I didn't like it, but I must say id never make those mistakes.

Especially the programming portion, in my tests that was 60% of the grade consisting of 2 questions usually, if I didn't know how to do a part, id usually at least enter everything I did know, declare everything relevent, and insert something like 'insert proper recursive steps to derive x' and fully finish what I was doing. Sure id lose a boatload of points, but its better to get 5 points than 0.

Another odd data point: The students who are at the top of the class and blowing away the rest of the class on these tests....are not CS majors.

At a guess, those students either (a)know they'll need it for something they want to do, aor (b) are actually interested in and enthusiastic about the topic.

I've actually noticed the same thing in some of my organic chem/molec bio/etc classes - the ones going in for science were more likely to be interested/capable then those planning to go into medicine (which made up a significant set of the students). However, I'm only a student so I don't have their grades to prove that.

As a CS major I was surprised when my intro teacher informed me that CS majors were almost all procrastinators. In my case this is completely true. On the day of our first test I walked in not knowing we had one. I wonder if this bit of knowledge could be true.

Hey, are the kids doing the best... female?

I'm teaching math and I've got a lot more young women in my class than my co-teacher; my class always outscores his on exams and a lot of it has to do with reading the questions, turning in the pages, etc. (The women are smart, too -- but the key thing with my class as opposed to his is that my students allow us to reward them for that smartness by turning in the work! what an idea!)

At certain ages the maturity gap seems to line up with the gender gap. I find it strange that I still have a hard time convincing my (female, biology-loving) top exam scorer that she's good at math, though. She thinks she gets good math grades because she has nice handwriting.

kt, except for the one that's failing, the women are doing pretty darn well (and one of the top ones is a woman). In my intro classes, where the gender mix is a bit closer to 50-50, the women do tend to cluster at the top of the class. Attention to detail really does help when you are learning to program.

Lewis, I haven't seen the procrastination bug as much with this group. Of course, I have somewhat of a reputation for being a hard teacher, and my assignments are not trivial, so I'm thinking the word is out: start early or face the consequences. :)

dreikin, it could be a bit of both (a) and (b). I'm still hoping to convince one or two of the top ones who still have time to change their majors to reconsider CS.

NI, now see, if they had done something like that, I would have given them somewhere up to 1/2 credit, depending on how close their algorithm was. At least what you suggest shows a good-faith effort to solve the problem.

DC, I have now learned something new. Thanks for the link!

One of the questions on the exam asked the students to explain the efficiency of a particular solution to a common problem. We've been doing variations of this analysis since Day 1 in this class. I believe we even did an almost identical problem in class one day. Two-thirds of the class got the problem wrong on the exam.

This brought back memories of a physics test freshman or sophomore year...I got the problem wrong and in the post-mortem in the next class session I was just stunned to realize the question that completely boggled me was just a slight variation of a problem I had easily solved in the homework. A lightbulb went off and I realized, they aren't all individual physics problems, you can recognize categories of problems if you know how to pay attention to the cues. What is so obvious to us can seem so opaque to young minds. Even as I write this it seems hard to believe I was ever that...I wanted to write stupid, but I will be more charitable with myself and say naive in my approach to physics. It was the same with calculus, and after I had the click experience in calculus, everything afterward was oh so much easier. You would think the click experience in calculus would have translated into physics but it didn't seem to. For me, I had to struggle for that click in many different subjects, until I finally got to a point where I was able intellectually to just search for it almost without thinking. It became a habit. But it wasn't inborn in me. So I pity your poor students.

I also pity you having to teach them! It isn't easy!

Zuska, as always you are wise beyond belief....I hadn't even thought of that as a plausible explanation as to why so many students got that problem wrong. And this is something I can easily address in class, and sometimes do: pointing out how this problem is just like that problem we did last week, except for X, Y, and Z. I guess I tend to be better about tying similar concepts together than tying similar problems together. Thanks for giving me some food for thought!

What is so obvious to us can seem so opaque to young minds.

There seems to be a negative correlation between how smart students think they are (and how vocal they are about their mad skillz) and how well they do on my exams.

Honestly, I think these types also drive off some of the good students from the major. Speaking as one of the quieter students in classes like that (or at least I used to be, now I like to make jokes during Analysis, but that's another story), these people are really annoying. I recall my freshman intro CS course, and these guys were at least half the reason I didn't continue as a CS major (I was/am interested in Computational Neuroscience, and had planned a double major in CS and Neuro). We actually have two separate intro CS sequences at my school, one that's a full year for people without programming experience and one that's a single semester for people with experience who want to get into the major faster. I was in the slower section having no previous experience and I still had to deal with these jerks who would ask advanced questions that had nothing to do with lecture solely to look smarter than the rest of us.

My lecturer was terrible and these students became so annoying to me that I stopped attending class and dropped the major (I still made an A though). This is despite the fact that I actually enjoyed working on the projects, enough that I would spend extra time to insert jokes and stylistic elements to my code. Fortunately for me, I eventually realized I still liked the thought process and ended up studying Mathematics (and focusing on Combinatorics!)

My point however is that, for the sake of those like me, if you can find some way to shut those people up, we might be more amenable to the major. Not that I would have necessarily have been the best CS student ever, but I'm sure a lot of those better students you want to convince might be in a similar situation, either annoyed or intimidated by these people who basically try to show off their "mad skillz" in class.