Reflections – a very personal view

Reflecting on teaching in higher education can be summarized as do I teach for students to pass the exam or do I teach them skills and knowledge beyond that? So far, I have aimed for the latter. Not only do I see lectures as supplement to the course material but also as an arena for asking questions and discuss the material.

[MOOC are supplemental but since we are social animals we learn best by discussing the material with another human being]

Lectures are not recitations of the course material, slavishly following the book. As I am now also designing exams, I take great care to ask ”thinking and understanding” questions and not fact questions which can be googled. In my opinion facts can be googled but it needs a brain and teaching to make sense out of the facts – and in the recent events also the skill to distinguish facts from fiction and fake news from facts.

The pedagogic course provided me with the terminology for it: assessment for learning not assessment of learning. That captures very well what my mom was teaching me. She did not ask her students who was N and what theory XY person N had but rather inquired whether the student understood the theory XY and could derive it and see the implications of it. It is this critical thinking skill, not just “learning by heart” (“pugging”) or rote learning. In the behavioural literature we distinguish between rote vs rule learning. Pigeons are rote learners, corvidae are rule learners.

We humans may learn a poem by rote but most other knowledge comes from rule learning, i.e. identifying the common elements, incorporating it into one’s own schemas, extracting the rule and generalizing it.  This is how we seem to learn the grammar of a language. This comes with biases and pitfalls, as our memory is not like a PC RAM nor are our brains computers. Rule learning does not kill creativity. Quite the opposite, applying knowledge across disciplines, or reasoning by analogy has let to great discoveries and insights. transfering a set of skill across disciplines is often a recipe for success. Accordingly, students do not just learn the terminology of their discipline, but being a student at a university teaches far more versatile skills, such as acquiring knowledge on your own, verifying the trustworthiness of facts and information, dissemination of research, and dealing with uncertainty or that the answer to ones question is more questions.

This is what drives humankind and makes teaching fun. Just think of a student suddenly making a connection between P and F where P stands for the topic you are right now teaching, like nerve cells’ action potential and its propagation, and F for football. I clearly see the link, football fans can behave like nerve cells, excited and singing or not excited and no chorus wave propagates through the stadium. Connecting the material to previous knowledge, and providing applications are an integral part of my teaching. After all, not all students become researchers. But all shall be critical thinkers with graduation.

Thus, I will continue my way of teaching for life rather than exam, improving on presenting the material but not much changing my attitude towards teaching and assessment.

The course on teaching in higher education made me aware of three main aspects I “kind of know” but have not given much attention to it. Firstly, it is about the choice of method. Why have I chosen to give a lecture instead of a seminar? Secondly, it is about how I evaluate students including giving feedback, but also designing exams and grading. Admittedly, I underestimated how much time it takes to make good exam questions, and I have not yet gotten faster at that. Thirdly, it is about the ethical guidelines of teaching, ranging from gender issues, minority aspects to considering disabilities like dyslexia. Accordingly I will take up each of those three aspects in depths.

To give or not to give a lecture

Lecturing is great, you can show off your knowledge, impress students and avoid any kind of interruptions if this is your style of teaching. It requires some preparation of the material you are lecturing but everything else is voluntary. The use of blackboards, hand-outs, powerpoints or other tools are up to you and the type of material you are teaching. There is only one snag: lecturing this way is for a student like watching a movie: passive and keeping them engaged requires a damn good movie. Either action or comedy. The danger is to become an entertainer instead of lecturing the material. A lecture’s goal is to teach new knowledge, it requires effort. It shall not be just entertaining.

The opposite method is “guiding” a seminar. Students present whatever they have prepared and expect feedback, morally as well as scientifically. Sounds easy during the seminar hour, as the students shape the lecture, but here the snag is in preparing good exercises for them that they find not too easy but also not too hard, and giving them the right timeframe and support along the way. It requires more from the students, but they receive direct feedback on how well they have understood the material. Karpicke and Roediger found clear support that not the number of “sampling” but the number of testing yourself matters. Seminars are a better arena for assessing testing, whereas lectures can repeat the material (sample) but unless one uses Kahoots or alike do not allow testing.

Well, I am not an actor and moviestar, respectively. My lecturing is tailored to the topic and the audience. But crucially, my lecturing tries to engage students by having short tasks incorporated in the teaching material. For example in my statistics course, I use Kahoots, this gives me and the students an immediate feedback on how well they understood the (often) tricky concepts in quantitative methods.

As an undergraduate I started with supervising lab exercises and seminars, i.e. I gave feedback on how well students have understood the material. This is a very effective method for small groups but a seminar-like style brakes down in large groups, as too many can hide and do nothing. Small groups are ideal and benefit a lot from such group assignments. This was also my reasoning behind developing the new PSY-2553 with 6 modules, where each module is a lab exercise run by a small group of students, testing their peers.  Briefly, every module has three 2h seminars; two before the lab exercise, providing some background for the experimental task, assistance in writing a consent form, and practical instructions; and one seminar focusing on data analysis after the experiment has been performed. Students prepare a poster and receive once feedback on it. They then present the posters in a poster session where peers provide feedback. This way, we combine cognitive concepts, classical experiments and how to do them, with dissemination of science – concisely but not too concisely summarizing the experiment to peers.

When it comes to transfer knowledge to a large group, lecturing is by far superior. The quietness in the lecture hall and the attention paid to the lecturer guarantees a certain knowledge-transfer from teacher to student. This can be enhanced by various methods, but one should keep in mind the main point of the lecture: presenting the material in a way that is accessible to the students, exploiting the dynamic inherent in a lecture (graphics, mathematical derivations etc). It should not be misused as a recitation of the written material anyway available to the students. Another important point is that both lectures and seminars are the public meeting points between the curious student and the also curious but slightly more knowledgeable lecturer in the topic at stake. Knowing about this unwritten deal is important. I find it ridiculous to forbid or limit questioning during or after lectures or seminars. Lectures are the meeting arena, and preferred over meeting unannounced. Therefore, I clearly encourage and signal that questions during my lectures and seminars are welcome.

To praise or not to praise

Lecturing and seminars also differ in the chance to give individual feedback. During a lecture I cannot give tailored feedback but I can provide the solution to a question I have ask, and hence a student can compare his/her result with the solution. From looking at their faces, I can gauge how well they have understood it and reassure myself to carry on or rather repeat the material. If a student asks a good question I can mention this explicitly and hence provide a kind of formative evaluation (Allen 2012). However, this is still rather unspecific if it is based on asking a question, as my reply how good the question was can be understood as a polite remark. Here, we can see the advantages of seminars. By letting students present something I can evaluate them how far they have come to understand the topic and are prepared for taking the exam. That is a major reason why I have always one seminar session in my lecture series. It pushes the students to start early reading the material, not just 1-2 weeks before the exam, as well as it is a tool for me to see how well I can explain the topics. Of course, this is not a 1:1 relationship. A batch of students may have had the material before, are highly motivated and use other resources, and therefore excel. Or the reverse, they might have heavy working load with another practical assignment and deliver a mediocre seminar work. Judging from the performance to my lecturing is not straightforward but it can hint at challenges students have, may it be outside of the course (study load) or with the topics themselves. Chats with students in the breaks often clarify this matter.

At a different level is the feedback one gives during supervision, either by commenting on written parts of a thesis or during oral meetings. I find this direct and 1:1 feedback very helpful. The student gets an evaluation on his/her understanding of the material but also working style. This is rarely just an evaluation, it often includes advices on how to improve. However, such feedback blurs the distinction to management. Supervision is not only about scientific feedback but also teaching personal skills and academic “rules”.

With respect to summative evaluation I have made recently a very strange experience. I have been asked to read four exams where students appealed their grade. An external sensor and I had pretty much the same conclusion which grade the students deserve. To our surprise was this in a somewhat stark contrast to the grade they received in the first round. We gave them better grades. This is strange insofar as I graded a Masterthesis very tough and a home essay assignment as well, with just one A, some D’s and many B and C’s. However, it all became much more clear when we looked at the so called sensurveiledning. It was good but not sufficient. It gave us two who were unbeknownst to the exact “pensum” too much freedom how to interpret what a student should write to get a good grade. This illustrates that large discrepancies cannot be avoided if instructions leave ambiguities. Having had the honour to grade exams from Høgskolen i Innlandet provided me with  an opposite example. Clear and detailed instructions, and every semester there are few complaints by the students.

Tove Dahl, a colleague of mine, criticized ambiguous instructions in her article (Dahl, 2006). The solution is not more bureaucracy around one big exams, but rather to have repeated assessments. This is far more common in the natural sciences as it is in the medical sciences or humaniora. For example, the course I gave at NTNU (2010) had two parts, an oral exam and a written exam. The oral exam was half-way in the semester, providing me and the students a feedback. They had at least to pass to get permission to take the written exam. What mattered for most was the more informative feedback on how well they did compared to their peers. Unfortunately, I do not know how well their oral performance correlated with their written performance at the end of the semester. It will not be a perfect correlation, but those who started early to learn and took it seriously (no procrastination or minimal effort for the oral part) most likely did better than those who gambled on passing the oral exam. This is in agreement with literature on students’ metacognition and confidence judgment, i.e. knowing how much they know and hence have to put effort into filling the gaps (Dunning et al. 2003).

To encourage earlier learning was also a driving force for the development of the new PSY-2553 (see above).  The modules require deeper learning throughout the semester, and students get feedback on how well they understood it in the poster sessions. It is not yet perfect, as the group work still allowed some students to be more passive and postpone intensive learning to the very last, but with a few exceptions, the exam grades were better than in a former course without the practical part (Psy-1003). This was only partly  because  40% of the final grade came from the poster (written, oral was just “arbeidskrav”). We may have achieved better grades if students would have received their grades on their poster before the 2h written exam, but we decided to grade all posters at the same time to minimize biases (recency effect, liking of the topic by external examiner).

 Ethical aspects in academic teaching and supervision

In the recent years, I became main or co-supervisor of 7 PhD students (by 2018).  Having a PhD student is a huge responsibility. The goal is not to have someone to do the research – I’d rather like to do research myself. It is to train a new researcher. Any PhD student is an apprentice. We supervisors transfer some research skills, but a lot of academic skills and assist in developing personal skills – said more directly: academia selects for persons with a decent frustration tolerance. Not everything works as it should the first time, things take longer than expected, and reviewers – well – can be mean. The PhD journey is often bumpy, and requires far more manager skills than being an excellent scientist (as recent cases have shown, e.g.

One grows through taking a PhD. This responsibility also means that I have to educate in a very strict sense what ethically good research is; i.e. ranging from teaching about plagiarism to avoiding fraud, stealing ideas, incomplete lab reports, fluffy-minded thinking or writing and so on. I had some really good discussions with my PhD students on all these topics and we re-assure us that we interpret the way new knowledge is gained similarly.  Since 3 years I actively encourage open science practices at all levels, with preregistration of studies, pre-prints, and providing open data.

The responsibilities of the supervisor and supervisee are explicitly stated at UiTs website ( I use them and can highly recommend going through them even with a Masterstudent and Projectstudent. The 1 hour invested in discussing this topic is well spent. If one figures out a major disagreement, one should not supervise this student. Similarly, a student should know her/his rights and duties.

But ethical aspects also emerge in the teaching of students. Here I can think of gender issues, minority issues and disabilities that can influence how I see a student and hence bias my teaching and supervision. As I mentioned in previous paragraphs, there is a deal between students and lecturers. Students pay for their education (indirectly through paying back studielån) and we get paid for providing this service (to society in general but the student in the case situation). I am not a big fan of consumer market approaches to education, but some truth lies in this view. Good education is transparent, open-minded, flexible and importantly has no fear for being evaluated from the outside. Being aware of being a “somewhat public” figure as teacher helps in providing a good teaching environment. Easiest is to apply Kant’s axiom: behave to others as you want others behave towards you (free translation of the core meaning of: “Handle so, daß die Maxime deines Willens jederzeit zugleich als Prinzip einer allgemeinen Gesetzgebung gelten könne.”). I still remember being a student, and this is my guideline on how to behave towards students (see also my teaching philosophy).


Allern , M (2012). Vurdering med vekt på summative vurdering – eksamen. UiT,

Dahl, T I (2006). When precedence sets a bad example for reform: conceptions nad reliability of a questionable high stakes assessment practice in Norwegian Universities. Assessment in education 13(1), 5-27

Dunning, D., Johnson, K., Ehrlinger, J., & Kruger, J. (2003). Why people fail to recognize their own incompetence. Current Directions in Psychological Science, 12(3), 83.