Next semester, I will be leading exercise sessions in a first year course of mathematics. More precisely, the course is Linear Algebra and the audience consists of young, first year undergraduate mathematics-students. The course is taught at a university in continental Europe. Judging from my personal experience, I know that most first year students of mathematics have plenty of troubles with the new, abstract material that is taught at a university. Hence, it would be good to make the student's experience in my exercise sessions as "easy" and enjoyable as possible.

However, I am not a fan of just doing simple, easy and short exercises. My opinion is that you do not make things easier for the students, if you omitt doing the hard exercises and just do the easy ones. Of course, this does not mean that everything I want to do is hard, but every now and then, something difficult will have to be discussed in the exercise sessions. Often, the hardest exercises are the most enlightening ones.

Moreover, my experience tells me that most students have wrong expectations of an exercise class: they expect to understand everything without (only a tiny bit) personal contribution and blame the instructor (which would be me in that case), if they do not understand everything. This has happened more than once to me and if I tell them that they have to work, I am considered to be the 'strict teacher', whose subject is the most important one and who does not understand that the new material is hard. I would be totally fine with that image, but this demotivates students and let them feel safe because they have someone to blame (Once, I heard the sentence "I would have understood the new material, but the exercise class was too bad and on a too high level" - in my opinion, I discussed everything in detail and pointed out which things are the most important ones and focussing on these).

In short, I find myself in the following vicious circle (it seems like you can achieve two of the three things below at a time, but not all three together):

  • Maximizing the students' enjoyment level (i.e., they should understand as much as possible and just have a good time during my exercise class)
  • Doing an exercise class which is on a reasonable (not too high, but also not too low) level
  • Keep the students motivated and tell them quite clearly, that they have to work to understand the new material

Or - in very short - the question is: Which aspects constitute a good exercise class (in mathematics) and which of these aspects are the most important ones?

  • 2
    "first year course of mathematics." That's very ambiguous: please clarify which degree program and include the broad geographic location (e.g. England, US, continental Europe, India...). In fact, even specifying the subject could be helpful, or at least clarifying whether there is a specific subject. I am a math professor, and I'm finding it hard to answer the question at the current level of generality. Commented Aug 23, 2015 at 20:24
  • 3
    Note that while this is not off topic here, you might choose to move it to Mathematics Educators if you want, if you think it might get better answers there. (But please don't post it on both sites, that is against SE network policy.)
    – ff524
    Commented Aug 23, 2015 at 20:43
  • 1
    Thank you ff524. I'll let the question here. And @Pete L. Clark, I have edited my question. Thank you too for your comment!
    – Slash_
    Commented Aug 23, 2015 at 20:46
  • 1
    Go for the second and the third points. When I took analysis, at the very first lesson, during the first proof on set theory, the professor paused to say: "You shouldn't expect to understand now the whole proof: try to grasp the concepts, and then work hard at home to develop a better understanding." But it was 26 years ago, maybe now the expectations are different. Commented Aug 23, 2015 at 21:00
  • 2
    You're presenting a false dichotomy. The best exercises are both challenging and enjoyable!
    – JeffE
    Commented Aug 24, 2015 at 2:22

2 Answers 2


As a grad student now who took undergrad linear algebra with a professor I really liked, I will attempt to generalize what I saw that worked.

The best way for students to learn something that is as potentially abstract as linear algebra is to relate it to something concrete. My professor always liked to point out that once you leave R2 and R3, everyone visualizes geometry differently. As a result, he often tried to keep his visual explanations in that space.

However, for higher dimensional vectors, you can still relate the material to multi-dimensional concepts without drawing a picture. A great example (and pretty interesting exercise) is Markov chains. When we went over eigenvalues and eigenvectors, we motivated all of the math by walking through Markov chains. He used Markov chain gambling schemes (e.g. holding a bet steady when you win, doubling when you lose) to illustrate eigen-decomposition in a practical way. In that way you can help students develop intuition by talking about something that makes sense outside of the math-sphere.

In terms of practical exercises, I've found that progressive problems tend to work best. That is, where each problem is working towards one ultimate problem solution. Generally, each step is a little more complex as well. For example, maybe the first step is finding the Eigenvalues and Eigenvectors, the second is diagonalization, the third is something tangential like inverting the matrix, and the last is applying what you've calculated to some practical problem like a Markov Chain.

In other words, ease students into the harder material by showing them the simpler ways to do it first. Once you can intuitively understand the general concept at basic level, it's easier to move on to more advanced views of the same and similar concepts. Plus, when students see complex material from the viewpoint of something simpler to understand (like maximizing gambling winnings), it provides a reference point for them to start from.

  • marcman.. .do you have notes, slides about what you just talked? :) I would have loved to take this one step at a time with practical examples because this is the only way I can learn something .. and I <3 Markov Chains
    – moldovean
    Commented Apr 12, 2016 at 9:48

In addition to the helpful comments and answer already provided:

Please let go of the idea of enjoyment and look for engagement.

I have fond memories, from my calculus classes, of the portion of the class when the teacher had us all go up to the blackboard to write up solutions to some of the homework problems. Each student had a section of blackboard to work in and a particular problem to solve. Then we all sat down again and compared what was in our notebooks with what we saw on the board. You could get up and go look at a problem closer if you wanted to. Discussion among students was permitted and you could ask the teacher a question.

When I was a language teacher, I was trained to include a variety of types of activities in each class. This is a good thing to do in math too. Some abstract reasoning, some mechanical matrix manipulations, some verbalization.

Make sure you do some spiral review, meaning, you should generally take 5 or 10 minutes to throw some problems at them from earlier units in the semester. If you notice that students are getting caught off guard by this, then tell them ahead of time what sections you'll be drawing review problems from during the next session.

You asked about the easier problems. A good place to put those is in the beginning, as a confidence-building, getting into gear, warm-up. The warm-up should last 3 - 5 minutes.

Keep your "lecturing" to an absolute minimum.

A discussion session (that's the term in the U.S.) is a great place for students to make connections with fellow students, that might result in small study groups being formed. Encourage these connections!

Watch the gender dynamics carefully. If you start to see some males behaving in a show-off-y way, and some females getting very quiet, that's a red flag, and it's time to try some new ideas.

I know you're not supposed to ask the identical question at the math educators site, but I would strongly encourage you to come up with a similar question to ask there -- you'll get more helpful ideas there.

Also, observe some other instructors' exercise sessions as a fly on the wall, and notice what works well and what doesn't work so well. And ask students what works well and not so well for them.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .