TLDR: 

Three universities in the greater Stockholm area in Sweden organized - in collaboration with EA Sweden - a student hackathon for outreach. The event took much more time to organize than the usual student outreach events, yet provided high information value and seems like an effective way to engage people with high agency related to entrepreneurial ventures. Read this article for a new experience in student outreach.

Introduction

Halfway May, our EA university groups (Uppsala, SSE, KTH) in cooperation with EA Sweden organized a hackathon for students that are new to EA. The hackathon challenged groups of 3-4 people to come up with project ideas for solving problems within one of four topic areas; global health, animal welfare, nuclear safety, and future pandemics. It was a full-day event consisting of an introductory lecture, ideation, hacking and finally presentations of project ideas for impact before all participants and a panel of judges. 10 Teams participated, which corresponds to a total of 32 people. The winners tackled the problems of vaccine spillage, local vaccine development, and communication around antibiotics. They won a prize of 25 000 SEK (~2400 euro) to invest in their project idea or in some other effective intervention, while the two runners-up received funding to attend an EAGx conference this fall, conditional on engaging further with EA.

The main goal of the hackathon was to let students engage with EA ideas and principles in an active manner, and invite them to join the EA university groups afterward. With 50% credence, we believe 3-6 people who previously had not heard of EAGx want to go and will get in.

Theory of impact

The main value that we hoped to generate, was getting students excited about EA by facilitating the active application of related ideas. This can be a proactive and fun way to learn about EA and it also promotes independent thought, which are all things that we want more of in EA. Moreover, we expected that organizing a hackathon would have high information value: It could teach us whether the amount and level of engagement of participants scale with the size of the event. How does a hackathon compare to other ways of doing community building? These are some pros and cons of running a hackathon that we identified:

Pros

  • An intense and focused experience relating to the principles of effective altruism.
  • Selects for high agency individuals as they’re more likely to join and succeed in a hackathon.
  • Participants have the opportunity to get a high-quality lecture on EA, get interaction with mentors related to EA, and get to know peers who are interested in social impact.
  • The setting incentivizes brainstorms to look for cause/project-X.
  • If done well, it can give the impression of EA as a serious and professional movement with many opportunities for ambitious people, something participants may want to further engage with.
  • For university groups, it offers opportunities to interact with startup labs and similar organizations/unions.
  • It promotes an entrepreneurial culture and running projects in your local group.
  • It is a good way for local groups to test and improve many aspects of their operations.

Cons

  • It asks for a relatively large time commitment from participants. This can filter for people who are prepared to commit or have an entrepreneurial mindset, but it can be a big obstacle to getting a large number of participants on board, especially those who are interested but skeptical.
  • As it is a larger event it takes more time to organize compared to other activities. If you don’t have a concrete plan or experienced organizers, it can require a considerable amount of effort.
  • It can be hard to find good judges and mentors
  • When considering the monetary costs (venue, food, marketing), time investment from organizers as well as mentors and judges, the event can become quite expensive per participant.

Our lessons learned

  • The amount and quality of the marketing to be done is important. Many of our participants were persuaded to join by a friend who was already familiar with EA, indicating that marketing can be difficult. Perhaps there are ways to incentivize the “bring a friend” approach? In our experience, about 15% of our time was spent on marketing, whereas we will aim for 25% next time (Note this is a very rough guess).
  • Hackathon participants can be highly positive about the experience, but not directly inclined to join local groups.
  • Many people don’t know what a hackathon is, or they have a preconception that it involves coding. This can be an obstacle when marketing. An alternative name might help alleviate this. (“Impact Combinator”?)
  • Students that signed up closer to the event were much more likely to show up. Increasing marketing efforts in the last days towards the event seems promising.
  • Have the event before the end of the school year. By holding the event halfway May we were competing with other events that weekend, students were busy with exams, and there were not as many later activities in the semester that we could point participants towards joining. Hence, having a hackathon earlier in the year would make it easier for participants to engage with local groups afterward.
  • We planned the hackathon too casually. Having stricter deadlines for marketing targets and content creation would be less stressful and improve quality. However, the bulk of the work was done by student volunteers who were busy with courses.
    • To combat this, there should be a clear project plan that includes a timeline, roles, and responsibilities, and where all the important things are included.
  • Marketing something as a “hackathon” can filter for ambitious people
  • Organizing something big like a hackathon created development spillovers for the university groups. A big event is a great excuse to try out all facets of the marketing arsenal (flyers, tabling, social media, etc.), and through talking to collaborators EA KTH even secured access to office- and event space at the startup lab of their university.
  • Having one or a few mentors go from group to group and ask them to briefly explain their ideas, then provide some short feedback (“can you make a quantitative estimate of this?”, “what might the counterfactual be?”), worked quite well for getting groups to prioritize the right information and avoid some pitfalls.

How to plan a hackathon

You can be the judge of whether it seems an effective use of organizer time, but we deem that it is since it promotes agency and initiative in the local group whilst also being an outreach activity where participants spend 10 hours on effective altruism in one day. Here’s how to run one:

  1. Book a venue at least 3 months in advance and approximate how many people you can get there. (For us, around 45% of the people that signed up attended, so take hours spent marketing * participants per hour * ½ to get an approximate number). Ofcourse, participants per hour is not a fixed number.
  2. Use the EA Groups Resource Centre checklist on in-person events to plan out the event.
  3. Use The EA Hackathons Participant Guide document for a curriculum or create one yourself.
    1. We are hoping to create a document for different curricula you could use in running the hackathon. So far we only have our own added which is a more general EA one, but if you have a curriculum you want to add feel free to do so.
  4. Market the hackathon, but be prepared for people to be a bit more resistant than usual as it is a larger commitment than other things.
    1. What we will try to do next time is to market and run some type of workshop or lecture around 3 days before the hackathon, and compare that to directly marketing the hackathon. This is because the time commitment that the participant has to put in is large and it might thus be easier for them to make a decision if we run an event introducing people to Effective Altruism beforehand.
  5. Feel free to reach out! We are happy to help where we can.

Conclusion

A hackathon seems like an effective way to engage people with high agency related to entrepreneurial ventures. It is a larger event and requires more time put into it, but if one compares it to e.g. a larger lecture, the time people spend on putting the ideas of EA into practice is probably a lot higher. It is, however, difficult to do really well and there are many ways we want to improve until we do this next time (Which will probably be next  fall/spring).

One thing to consider is if it might be possible to scale this up by having multiple hackathons in different cities and/or neighboring countries, and then let the winners of each go on to a finalist round. This could make the events more attractive to participants, produce a greater variety of ideas, and filter for a winning project that is more effective and likely to be pursued by its creators.

Our main lessons are:

  1. Marketing for the hackathon is difficult and efficiency seems to be highly dependent on the approach and timing.
  2. People got to implement the ideas of effective altruism and seemed very excited after the hackathon (the average enjoyment in the feedback after was 8-9/10)
  3. We will plan the hackathon earlier next time so that we can scale it more (to maybe 60-70 participants), the biggest cost for this would be increasing the marketing efforts, but it could more than double the impact.

If you are interested in replicating/building onto this, send us a direct message and we can link you to our documents!


 

17

New Comment
5 comments, sorted by Click to highlight new comments since: Today at 7:09 PM

Thanks for this write-up! For me, a 10 hour hackathon sounds rather short, since with the lectures and evaluations it only leaves a few hours for the actual hacking, but I have only participated in hackathons where people actually programmed something, so maybe that makes the difference? Did the time feel short to you? Did you get any feedback on the event length from the participants or did somebody say they won't participate because time commitment seems too big (since you mention it is was a big time commitment from them)?

Great questions! From the 10 hours only 5-6 were spent hacking, which felt as short. Some participants mentioned in the feedback form they would have loved to see a 2-day event, whereas others mentioned that they thought the length to be great. When we marketed the event there were students mentioning they preferred to spend the weekend not doing much or that they thought of themselves as not being ambitious enough. I think this amount of time balanced well the quality of the event, the entry barrier for participants (time commitment), the costs of venue and food & the ask from volunteers. Next time we could do a networking & briefing session the evening before the hackathon (perhaps online if necessary?), in that way we can add some hacking time during the day itself.

thanks for the info! I didn't really get the part on ambitiousness, how is that connected to the amount of time participants want to spend on the event? (I can interpret this in either "they wouldn't do anything else anyway so they could as well be here the whole weekend" or "they don't want to commit to anything for longer than 1 day since they are not used to committing to things".)

I encountered both types of participants (the ones that showed up because they had 'not much better to do' and the ones that are not used to committing). My impression was that most participants were ambitious and that they liked a challenge. The effect of the length of the event on potential participants with varying levels of ambition can be a bit ambiguous here. With a longer event it is also more likely that potential participants have other things planned during part of the event. My gut feeling says that making the commitment bigger than a day for an introduction hackathon (without coding) makes it less likely for people show up. 

Wow, nothing helpful to add, just wanted to say think you for writing!!  These tips will be so helpful in prepping for this event:  https://forum.effectivealtruism.org/posts/BL8Rny4hBgDvLsjci/hackathon-on-mon-12-5-to-follow-eagxberkeley

 

☝️ if you guys have any thoughts for potential software engineering projects to do good better, please feel free to add them as a comment to the post.