Epistemic status: This post is mostly a data point in the EA community building pursuit. All conclusions are based on intuition, aided by sparse data - both acquired during a short time period. Also, this post assumes that students are an appropriate target audience for community building efforts.
Intended audience: Anyone involved with supporting student groups, involved in national groups or city groups with multiple universities, or curious readers. Observe that the report on which this post is based makes recommendations for EA Sweden. Any generalizations to other groups should be made having their specific context (e.g. cultural, practical) in mind.
This post was written partly during my employment with EA Sweden. The views in this post, however, represent my own.
EA Sweden, the national EA group in Sweden, supports the country’s eight local groups, all but one of which are university groups[1](I will use “university group” and “student group” interchangeably). During March-December 2022, EA Sweden tried out having a University Specialist (USp) supporting university groups in the Stockholm region[2], to see what value, if any, there would be having a separate person on the team with this responsibility. This position was held by me and partly by Robert Praas.[3] As part of my exit, I wrote a report[4] reflecting on this question and made some general[5] recommendations for supporting university groups based on what activities led to the best ROI, and what I thought could be improved. This post is a shortened version of that report.[6]
I and Robert had together with two other community members been tasked to reboot the university group at KTH in the autumn of 2021 which still, at the time, was one of the more active of four groups in the region. When we transitioned to the role of University Specialists, we kept the focus on the KTH group to (1) develop an internal “proof-of-concept” before supporting other groups, and (2) make sure that the KTH group would be left in good shape until both I and Robert would leave in June. Together with the group, we set up weekly project sessions and organized a team retreat and other social events. Together with two of the other three groups, we also helped organize an Impact Hackathon.
We also spent time laying the groundwork for the fall, e.g. identifying and talking to potential new organizers, and designing an overall strategy which was implemented as below:
In general, these recommendations are weakly held due to the limited data on which they are based, and I encourage EA Sweden to continue experimenting with group support activities. I have numbered the recommendations in order of importance.
1) Make sure[7] intro lectures are organized and at the start of academic years;
8) Make sure university groups are present at university welcome fairs at the start of academic years;
A lunch lecture is a great opportunity to do high-fidelity EA communications - we reached 150 students with 5 lectures, with ~7% of the audience taking further action. Similarly, welcome fairs attract students when they are extra open-minded and information-seeking. For the two I was present on, each group received ~100 sign-ups to their mailing lists. The benefits of outreach during the start of academic years have been stated elsewhere on this forum.[8] However, planning for activities during this time is often done before or during the summer, which is difficult for new or rebooted groups, in which case EA Sweden can help.
2) Implement routines for inducting new organizers, including relevant community building resources and guidance for activity plans;
Routine induction meetings with new (lead) organizers are a low-cost way for EA Sweden to:
3) Be in regular touch with university groups and ensure their continued existence (if relevant and sustainable), e.g. by keeping an eye on annual meetings and ensuring good succession;
In Sweden, most university groups are student associations, meaning that they have a board of directors who are re-elected once a year at an annual meeting. I think ensuring the continuation of a group is important for the future growth of its presence on campus. As the continuity of EA Sweden is more robust than that of groups, they are well suited to support groups in electing a new board (e.g. by nudging potential organizers) and ensuring that a handover occurs. It may make more sense to reboot a group if there are not sufficiently many organizers who can or want to continue, rather than trying to convince people to stay just to keep the group “alive”. In general, regular communication easens support (e.g. contact via text/calls/lunch once a week/month/quarter) and keeps EA Sweden updated on the groups’ progress.
4) Make sure one bigger event for university students is organized each year;
In May, the USp supported a couple of the student groups to organize an Impact Hackathon, where thirty participants met up for a whole day, formed teams, and came up with ideas for solving some of the world’s most pressing problems. It seems to me that the most important outcome of this event was the post-event engagement of seven of the thirty participants, who went on to do intro fellowships, career courses, and/or engage in local groups. I dare to hypothesize that organizing bigger events like this (compared to e.g. a weekly book club) where participants are directly engaged in the pursuit of doing good in some way together with others can be great catalysts for engagement as well as creating positive associations with effective altruism (which I think it deserves!).
5) Continuously disseminate local and global engagement opportunities directly to student groups and other groups who have previously shown interest in EA-related material.
When marketing our intro fellowship, career course, and EAGxRotterdam, I found the following to be the most effective outreach method (compared to e.g. general social media marketing):
Of the 150 people reached this way, around ~15% ended up applying for either the courses or the conference, or both. On another note, hosting digital preparation workshops before EAG(x) conferences seems to be a valuable low-cost community building tool in itself.
6) Organize in-person intro fellowships at least once a year (if groups are unable to);
7) Organize in-person career courses at least once a year;
11) Develop clearer learning criteria for evaluation for both courses and iterate on data collection using the reports from 2022 as a basis.
I think courses are great community building tools, but they can potentially be too heavy to organize for university groups, in which case a USp could assist them with e.g. facilitation, logistics, and applications. EA Sweden independently organized both an intro fellowship and a career course primarily targeted to students, both of which were well received. However, the exact learnings of participants were unclear.[9] I think we want to make sure participants actually update their worldview, in addition to having enjoyable intellectual discussions with like-minded people (which seemed to constitute a big part of their satisfaction, although this is also a good thing).
9) Define what success looks like for local groups and in the context of EA Sweden’s strategy, and develop relevant ways of evaluating group support accordingly;
When are groups “successful”? It varies, but at least having a definition I think can help track and motivate progress. Firstly, helping groups establish their own measures of success[10] can be a good way of supporting them. Secondly, this will create a clearer measure against which group support can be evaluated (i.e. "How have we helped the group become successful?"). Thirdly, I think EA Sweden creating their own definition of what a successful group looks like[11] will help them reach their wider community building goals better.
10) Design a more intentional and systematic strategy for university marketing/outreach that can be iterated on and improved from data collection and analysis;
Outreach for events was done on quite an ad-hoc basis. An overarching strategy could clarify target audiences and specific outreach methods and evaluate what works via participant forms or statistics. I think this would make it easier to streamline effective outreach and prevent re-inventing the wheel.
I expect there to be better and more cost-effective ways of supporting university groups than what I have recommended. But these serve as a good start that can be iterated upon. You could categorize the recommendations as follows:
Ideally, direct community building is left to university groups, but depending on the groups’ capacities, EA Sweden can make a counterfactual boost by helping out. Recommendations 9-11 are more “one-off” jobs and wouldn’t need to take more than 1-2 hours each. However, they are also the least important.
There were a few things that did not stand out as particularly effective or important and so did not make the list, although they may have had some benefit:
Here, I am more speculative and uncertain than with the general recommendations.
Looking back, I am 65% certain that the work done by EA Sweden’s USp counterfactually (compared with EA Sweden having no USp but similar workload/priorities) contributed to:
However, given that EA Sweden already has concrete plans in 2023 for most of the recommendations above with their current team, I do not think there is much value in adding a separate USp at the moment.
Bear in mind that any generalizations to other countries/cities/groups should be made having their specific context (e.g. cultural, practical) in mind.
Having that said, I think having a separate person handling university group support (be it a University Specialist or 30-40% of the responsibilities of a co-director/community manager) in a multi-university city like Stockholm makes it easier to get to know the group. For example, by attending the group’s events or collaborating in organizing them, not only Zoom-ing them every now and then from your office. I think this has helped:
In line with recommendations 2 and 9, this also helps oversee the general quality of the student group’s work.
We can also take a scenario in which a city or national group wants to seed, reboot or stabilize several groups in a multi-university city. Having a separate person may then make it easier to not compromise with the city or national group’s comparative advantage, which is their capability to organize bigger events and oversee strategy and development. Compare this to student groups that are better suited for “direct” community building work where members are located (e.g. discussion evenings, socials, book clubs).
UGAP seems like a good resource for community building training to recommend to all new/rebooted university groups, and national groups (e.g. via a USp) are in a good spot to encourage groups to apply. In the case of university groups in Sweden I do not think, however, that UGAP should be seen as a way of outsourcing induction by the national group entirely, as this helps with providing better group support (see also recommendation 2 above). Some university groups for which UGAP may be relevant may also not be suitable[12] to take the program for whatever reason, in which case a somewhat more thorough induction from the national group may be relevant.[13]
If hard-pressed, I would answer intro fellowships and/or career courses, with a confidence level of 80%.[14] Firstly, organizing courses seems highly counterfactual at the moment due to the current capacity of the groups. Secondly, I think the practical, time-boxed, thorough, and collaborative nature of the courses makes it an appealing activity to participate in, making long-term engagement with EA ideas more likely than other interventions do. Although courses may increase participants’ resources and dedication towards having a greater impact - drawing people in - they are not so good tools for developing and maintaining a community structure that makes people want to stay engaged and thrive although I think the former certainly helps with the latter.
What do I mean by a community structure? For example:
In general, most of my recommendations (except perhaps 3 and 5) are about helping groups to draw people into the community, rather than developing a good community structure. I think that national and city groups can support student groups with this without removing the groups’ independence, and I would love to see more reflections on the topic.
If you have read all this way and have any thoughts you’d like to share, please consider leaving a comment! I also welcome feedback on how I could make posts like this easier and perhaps more entertaining to read.
Thanks to Vilhelm Skoglund, Cecilia Tilli, and Robert Praas for guidance, mentoring, suggestions, and fruitful discussion during the year. Thanks to Amarins Veringa for recommendations regarding this forum post, and to Vilhelm Skoglund and Robert Praas for extensive feedback on the post. Thanks also to Emil Wasteson, Kiryl Shantyka, Anna Ek, Mimmi Thor, and Krummi Kristjansson at EA Sweden for input and collaboration during the year. Last but definitely not least, thanks to all the amazing members and organizers I have gotten to know at KTH, SU, SSE, Uppsala, and in the wider Stockholm area - you know who you are!
The groups in Gothenburg, Lund and Uppsala are focussed on the universities but also include the wider city, compared to Stockholm, where the city group and the university groups are separate.
Specifically, KTH Royal Institute of Technology, Stockholm University, Stockholm School of Economics and Uppsala University. Attempts were also made to seed a new group at Karolinska Institutet - some progress was made in terms of acquiring useful contacts and information, but no group was started. Initial contact was also made with an engaged organizer in Linköping.
Between March and June I and Robert shared the position for 0.2 FTE each, and between July and December I worked 0.45 FTE. Robert initiated the idea of a University Specialist to EA Sweden in early January 2022.
If you want access to any of the other documents that are linked within the report, please DM me.
I also made quite a few specific ones for organising intro fellowships, career courses, and outreach - you can read more about those in the report and the attached documents.
Since I got to spend quite some time writing this post, a few of my views changed along the way. Thus, in case there are any contradictions, the content of this post should take precedence over what is said in the report.
This could mean that EA Sweden does most of the initiative, or encourage groups to pull most of the weight, and a judgment should be done for each group. Also, to clarify, I do not think that national groups should manage university groups - the keyword really is 'support'.
But I can’t remember what post! If anyone knows what I am talking about, please leave a comment!
We used a few vague measures of learning - for example, asking “Did you change your mind about something during the course?”, as well as comparing cause prioritisations/career plans before and after the courses to see if something had changed.
This is roughly similar to goal-setting, which I take to be more active than merely establishing a success metric. It may be more relevant for some groups to set goals, but this may also yield more pressure.
For example, I think any of the following would be independently good proxies for success, in order of importance:
*Most active members either taking 1-1 Career Coaching, and/or attending an EAG(x) conference, and/or taking an introduction fellowship (virtual or IRL);
*University group organizers spend more than 5h/week on the group and enjoy it; relevant smaller events are organized by the group at least once a month;
*Steady increase in membership numbers over a year;
*Organic development of a sustainable community structure (see later in the post);
*They establish collaborations with other groups at their university;
*They organize bigger events.
However, as far as I'm aware, Effective Ventures recently made it mandatory to complete the first stage of UGAP in order to apply for grants, which I think is a sensible thing to require.
The two groups in Stockholm for which UGAP was relevant went through its first stage but not further (which I have gathered to be mostly an introductory session and further resources), saying that “time” and “uncertainty whether the program was the best way to spend their time” were their main reasons to not continue. To get a better picture of the counterfactual, it will be interesting to watch the progress of the group in Gothenburg, who has gone through both stages of UGAP during the fall, without much contact with the USp.
Note that this is different from "If asked to keep only one thing, what would it be?", in which case I would take number 1 on my list which is intro lectures.
These all don’t have to be written down in a document for them to exist.
