Tuesday, June 1, 2010

Want to improve student learning? Then fail more!

Six months ago I mentioned an excellent post by the Female Science Professor about the struggle of teaching graduate students to write. There are a lot of perceptive and helpful comments on that post. There are two comments which have stayed with me and I wish to highlight because they are relevant to much broader issues in undergraduate education.
Linda said... I don't believe that this is a skills issue.

The issue is that there are no consequences for writing badly. That student, one way or another (including buying papers), will continue on in his/her education.

When we take this seriously, students will be told "no, you cannot continue on in your education until you improve this skill".

Oh, the horror! Holding students (at ANY level) to a standard. Isn't that unconstitutional or something?

Anonymous said...

My English 101 writing teacher threatened to fail me unless I improved. My paper was,

- too long ... 5x requested length
- contained too many parenthesis
- contained too many footnotes
- thesis unclear
- etc.

I made those improvements. Compared to professional writers I know I could improve much more. Still, the professor made her point.

What's difficult to understand here? Fail the student if the writing under performs.

I attended university in my mid-30s. I applied the basic principles operating in most work environments. Those in charge identify areas of weakness and require improvement in their subordinates. And either the worker improves or risks being replaced. Ideally this improvement is done with the cooperation and support of the boss.
Currently I am teaching a second year undergraduate course on thermodynamics and condensed matter. To enrol students should have passed a course on multi-variable calculus and a first year calculus based physics course. Yet there are many students in the class who struggle to do basic things like:
  • Sketch a graph of a one variable function.
  • Find the value of x at which a specific function is a minimum.
  • Evaluate a partial derivative of a simple function of more than one variable.
  • Keep track of physical units in a calculation.
I believe the reason students struggle to do such basic things is because they have never HAD to learn to do these things "in their sleep." They can pass courses without really being able to do them in a consistent and disciplined manner.

We all need to improve the quality of our teaching. But sometimes I think the best thing we could do for our students is fail more of them. Furthermore, as one of the comments above notes dealing with failure (or at least the threat of it) is important preparation for life in the "real world."


  1. Oh Ross, I think this is the biggest problem in Australian Universities today. I was talking to a guy in Leipzig on Sunday night who was thinking of dropping out of 3rd year undergrad physics because he didn't realise it would be so difficult. I guarantee you he'd pass quite happily in the university I went to. My current boss is running a grad course on QFT in condensed matter which would have made my head explode after undergrad.

    Soft marking is rife, it's no secret. But I feel like it's doing the universities no favours at all. It infuriates me when I hear that 70% failed statistical mechanics which the university feels is unacceptable, so half of them are bumped up to a pass. Sure there may be circumstances when that's ok - fine.

    Incidentally on the grammar side of things, I was the only person in undergraduate labs who (right or wrong) actually marked students down for poor grammar.

    Your link to the Female Science Professor was very interesting. It had never occurred to me that perhaps people are trying, but still can't string a cogent argument together. I'm not sure I believe it actually.

    Phd Comics had a nice grammar and marking reference a week or two back:

    Sorry if this was a little rant-y. Enjoying the blog immensely as always.

  2. I didnt learn anything about units until graduate quantum mechanics. I also take no personal responsability for that one. To his great credit Brad Marston made me learn units, and later my thesis advisor reinforced it.

    The reason was simple: it was never required previously. My electricity and magnetism courses used different units each term --to this day i loathe cgs and prefer SI even with mu's running around.

    as for the writing, it was not required of me either.
    I really like ideas like this:

    and maybe, just maybe, we should require physics majors to take a writing class --something i never did either!

    now the math bit you speak of is new to me, although i have this issue with my own student: im teaching introductory classes and sometimes even basic algebra or trigonometry is a problem, so i cover it and tell them they need it. most students can learn and are willing to work, i just wonder how i got them without the other parts in the machine of education doing their job. one thing i did learn in physics was plenty of math.

  3. I agree that the students are struggling with simple things because they don't have to do them. However I don't think that failing more students is how to get them to learn these simple things. If they don't have to do these things to do the material of the course it would be unfair to fail them on it. "I'm going to fail you on this question because even though you got it completely right your grammar is atrocious" would not go down well to any student and I believe that all it would accomplish would be to lose you respect. Especially if you aren't going to teach the student how to do it right.

    Instead, make it an additional criteria and go through it in every example. Want students to learn to carry through units? then do examples every lecture where you carry the units through. Set out your problems like how you want them to answer. I gurantee that will produce better results than just failing more students.

    Students wouldn't have to undertake an additional writing course if this was done in all physics courses. I know that in all my education courses grammar, spelling and general structure is a criteria and FEEDBACK is given on these things. feedback is the most important thing here. The student won't get better if you fail them and they don't know what to improve.

    We're probably saying the same thing in different words. it's not really the student's fault if they never HAD to learn something but more a fault of the course. Also I want to emphasise how the solution isn't to fail more people but to make it a part of the course, a criteria. It looks like the same thing, but there's a fine line seperating the two.