Flat course fees for online teaching: is the cure worth the disease?

Work in progress

This opinion piece is a work in progress and will change.

Goal

This paper started at the NBIS Training Steering Group meeting of 2026-09-21. It was announced that a flat fee of 1000 SEK will be introduced for online courses. Here I try to find out who benefits and who is harmed by this idea.

Flat course fees are announced

The flat course fee was announced in the meeting notes as such:

Online Courses
- Flat fee of 1 000 SEK (administrative costs + technical operation cost)
- Scrap no-show fee

In that meeting, questions were asked by the course leaders.

The course fee did not arise because of the money: NBIS is not in bad shape financially. Instead, the idea is that it will help registered participants to actually show up.

Course attendance is a problem

Course attendance is, indeed, a known problem. Ideally, we want it to be 100%, i.e. that all learners that register for a course show up. In practice, for this is what it may look like in practice:

Figure 1: course attendance. See Appendix 1

By eyeballing (and ignoring the 100% outlier, where 3 out of 3 showed up) we see that in those course we expect around 1 in 3 registered learners to actually show up. Although this figure shows the problem of a less-than-100-percent-attendance rate, it also shows a solution to this problem that is used at NAISS and UPPMAX: they estimate how many will show up.

A flat course fee increases attendance and other positive effects

The idea behind a flat course fee is to increase attendance (i.e. the percentage of people that have registered that actually attends the course): if a learner has paid 1000 SEK before attending (and it cannot be refunded after 3 weeks before the course starts), this should motivate him/her to actually attend.

At first glance, a flat course fee like a good idea to increase course attendance. Here I assume a flat course fee increases course attendance to 100, although similar-but-different setups hint that this assumption is false [Bhayani et al., 2021][Ketel et al., 2016]. Having (close to) 100% course attendance solves multiple problems:

• A fee is more fair to the rejected participants, i.e. those that could not attend when a course has reached capacity, as there will be less empty (virtual) chairs that they themselves could have occupied
• A fee helps schedule the right amount of teaching assistants: as it is predicted to have a higher course attendance. This may help course coordinators

A flat course fee impacts others negatively

However, there are also some drawbacks to having a flat course fee to be able to more reliably estimate attendance:

• A fee means the course is no longer open for all: This effects all teachers that care that their courses are open to all, even to those registrants with little financial resources. It would mean that NBIS online training will benefit mostly those that can afford them.
• A fee solves a non-existing problem: the are courses that deal with this uncertainty by estimating attendance. Those courses will have a lower number of participants due to a problem that is non-existing for them.
• A fee misaligns with our mission: instead of helping our participants to do 'research otherwise not possible', we put an administrative burden on them to secure a course fee
• A fee punishes registrants that cancel for the right reason: Participants can have good reasons to cancel a course (even last-minute!) and now have to unjustly pay a fee for a course they (rightfully) could not attend. Examples are funerals and accidents involving hospitalization.
• A fee reduces the number of registrations: Not every learner has access to get a course fee paid for them, nor not every learner wants to deal with the administrative burden of securing money for a course.

Intentions are good

It is with good intentions that management tries to help its training staff to have more certain estimates on the amount of participants that will actually show up. We can expect staff to trust management on making good decisions or -the more negative one- we can speculate on staff to 'just' obey management without questions. Whatever relation our colleagues in training have, a flat course fee will benefit those that crave knowing how many participants show up.

Other solutions

Taking a look at similar course providers, we can see a simpler solution at NAISS and UPPMAX, where this uncertainty is simply embraced and the amount of participants that will actually show up is estimated. From the estimated amount of participants, the fitting amount of teaching assistants are asked to be present at the start of a course. When there are less participants than estimated, those course assistants are sent away, back to their regular jobs.

Taking a look at the literature, we know that a having course fee (versus a course being free), will have an impact, as 75% of MOOC participants state that having no fees is 'influential' for their enrollment [Clow, 2013].

However, there seems to be no research (to the best of my knowledge) that helps us predict the effect of a flat course fee on registration numbers. The closest relevant paper is by Han and colleagues that investigate enrollment rate for variable pricing of online courses. Based on emperical evidence and assuming a market where online and traditional courses compete, they found that per 10% price increase, there are 15-18% less enrollments [Han et al., 2019].

There is, however, another positive effect of having a course fee: it increases engagement with the course. This has been found in the massive open online course (MOOC) provider Coursera. They also struggle with more people registering than being active [Goli et al., 2022]. They tested out a flat course fee, as well as a certificate. This research, however, does not provide data on how these interventions changed the number of registrations. It does show that a course fee increased engagement (i.e. those that attend do spend more time on the platform) of 17-20% percent at the start and 4% after 10-20 days , where a certicate increased engagement 8-9% during the entire course.

Support for/against a flat course fee within NBIS

It is unknown to which extent there is support for a flat course fee. Hence, I have asked NBIS and non-NBIS teachers in online courses for their response to having an online course fee to increase the certainty of a high attendance. I did so without disclosing my own views.

So far, the reactions are around 50-50.

Suggestions given by those in favor of the course fee:

• Consider using a no-show fee instead: this may be a good option too
• Charging a fee increases the investment with a course
• The course fee amount not a problem, as this will be paid by supervisors, not the students themselves
• The course fee is too high

Suggestions given by those that are against a course fee:

• A course fee introduces more hurdles for entry, such as convincing a PI to pay for it, followed by the administrative processes that follow.
• The course fee is too high and it should not be more than 500 kroner.
• A fee will reduce the amount of participants, which will hurt courses that have a low amount of participants. Instead, learners will increasingly turn to AI (as long as this is perceived as free)
• A fee solves a problem that does not exist, as there are enough teaching assistants within NBIS

Other suggestions:

• It depends on the course setup: If courses need to book more teaching assistants for more learners (e.g. when there are a lot of discussions), then, yes, having a course fee is a good idea. However, some activities (e.g. automated tests) do not require more teaching assistants when there are more learners. In such a context, a course fee in not needed.

Although this small poll shows some numbers, the winner is not decided by mere voting (in that case, the course fee would be removed). Instead, a more useful question is to what kind of training should be catered too. For example, should we cater to teachers that struggle with an uncertain amount of participants, or to those that do have the pedagogic tools to deal with such uncertainty. Additionally, should we focus on teachers that see teaching in an economic businesslike way, or to those that intimitely care about the profession.

Support for/against a flat course fee outside of NBIS

I have asked non-NBIS teachers in online courses for their response to having an online course fee to increase the certainty of a high attendance. I did so without disclosing my own views.

So far, the reactions are clear: all are against a course fee.

Below is an aggregate of the reasons they provided:

• The fee costs time and money: the administrative cost is higher than the fee brings in. This is especially true when attendants from different countries (e.g. with a different VAT rate) apply.
• A fee penalizes the wrong people: a flat course fee hinders those people that we want to attend, such as early-career researchers, students, or participants from less well-funded institutions and countries. It is exactly those that we should provide this service to.
• A fee turns students into customers: this is a big difference in mindset with lots of risks involved
• A fee is inferior to better communication: It was found by experience that better communication is more effective at increasing attendance, such as sending personal confirmation reminders, requiring explicit re-confirmation a week prior, or using waitlists with clear cancellation deadlines
• A fee is an alternative to free methods: There are other ways to increasing attendance that do not cost money e.g. a short supervisor approval note or asking people to write 100 words explaining why the course might be useful for them
• A fee reduces the amount of participants, where we want as big as a reach as possible
• We are not allowed to have a course fee at all: this happens to be the case for other (i.e. not NBIS) public-funded research infrastructure
• A fee misaligns with higher goals, such as the principles of Open Science and FAIR training principles for public-funded research infrastructure.

Conclusion

Not all learners that register for our online courses actually show up. This is a known problem for NBIS and other online course providers, such as NAISS and UPPMAX.

Having a flat course fee is a well-intended way to increase the certainty that all registered participants will actually show up, although the academic literature provides no evidence that a course fee will increase attendance. Knowing more certain that all registered participants will actually show up is is meant to help assure that the right amount of teaching assistants can be scheduled. Also, it is more fair to those that were not accepted by the course due to the course being full: a person that is accepted cancels, that empty spot could have been used by the rejected participant. Drawbacks are, among others, a lower attendance, an administrative burden, a penalization for the wrong people, injustice to those that rightfully need to cancel, change of mindset in whcih students are treated as customers and misalignment with higher goals such as Open Science and FAIR principles.

The most elegant alternative of having a flat course fee is to estimate the amount of participants actually showing up, as is done by, for example, NAISS and UPPMAX. Advantages are, among others, a higher attendance, no administrative burden, the mindset to treat students as learns and alignment with higher goals such as and FAIR principles. A drawback may be that not all teachers or teaching assistants have the mindset or pedagogic tools to deal with such uncertainty.

I conclude that it would be better to have no course fee, similar to other government-funded course providers and this idea would have support from within NBIS staff. I hope that NBIS Training will re-evaluate using a flat course to enable research otherwise not possible in a more welcoming, just and efficient way.

Epilogue

These are some things that do not add to this opinion piece, but it informative about the background from which it was written.

My assumptions about NBIS Training

I assume that NBIS Training is open to opposing ideas, such as this opinion piece. My high expections of those in charge does contrast some sentiments I received during my polling, where it was voiced that such decisions are a one-way street and was given the advice to just obey. This is a sentiment that I do think is relevant, but I did not investigated this any further. Again, I do believe my colleagues are actually fine with opposing ideas.

Addressing my biases

On a more personal note: I am biased, as I care deeply about teaching and the effect my teaching has on my learners. Throughout this opinion piece, however, I have been open to opposing views too, to the best of my abilities.

I think learning is important. I think learning is so important that I sacrifice most Saturdays to coordinate and teach multiple courses. I do so as a volunteer, in the the so-called Saturday courses, which take place in the Uppsala Makerspace. The courses are set up around the motto 'Everyone is always welcome', embracing the fact that participants may just have migrated to Sweden (e.g. Ukranian refugees) and may have unexpected and irregular (e.g. weddings and weddings) events in life that are more important. The courses have grown to around 40 participants and have fluctuating attendance.

Those courses do have a course fee, to cover the costs of the courses, of 900 kroner per year.

However, the payment instructions open up with, I quote: 'money should never be a problem at the Saturday courses'. I know that around 1 in 20 parents come to me and ask the course to be free for their children, which has always been granted. In that way, we have been able to keep the participants of lower socio-economic status, which may be the ones that need our courses the most.

I am proud that, at those courses, we live up to our mottos that 'everyone is always welcome' and 'money should never be a problem'. I hope that I can do the same for NBIS too.

References

• [Admiraal et al., 2023] Admiraal, Wilfried et al., "Mind the gap: Early-career teachers' level of preparedness, professional development, working conditions, and feelings of distress". Social Psychology of Education 26.6 (2023): 1759-1787.
• [Bhayani et al., 2021] Bhayani, Ali. "Let students pay for their higher education: Debate concerning free and subsidized education based on sunk cost theory." Journal of Philanthropy and Marketing 26.4 (2021): e1700.
• [Clow, 2013] Clow, Doug. "MOOCs and the funnel of participation." Proceedings of the third international conference on learning analytics and knowledge. 2013.
• [Goli et al., 2022] Goli, Ali, Pradeep K. Chintagunta, and S. Sriram. "Effects of payment on user engagement in online courses." Journal of Marketing Research 59.1 (2022): 11-34.
• [Han et al., 2019] Han, Yongseung, Michael P. Ryan, and Kelly Manley. "Online course enrolment and tuition: empirical evidence from public colleges in Georgia, USA." International Journal of Education Economics and Development 10.1 (2019): 1-21.
• [Howarth et al., 2016] Howarth, Jason Paul, et al. "Learner motivation for MOOC registration and the role of MOOCs as a university ‘taster’." International Journal of Lifelong Education 35.1 (2016): 74-85.
• [Ketel et al., 2016] Ketel, Nadine, et al. "Tuition fees and sunk‐cost effects." The Economic Journal 126.598 (2016): 2342-2362.