Epistemic status: Based on my own and our group’s experience. Please note that the context I’m in (I was the first community builder in the capital of a second-world country) is different than that of many current community builders, so what I’m writing will be somewhat less relevant in some contexts, but still potentially useful for people who live in EA hubs or cities with several already existing university groups as well.

This is the first post in a sequence called Experiments in Local Community Building.

How my university group turned into a city-wide university group

I was very eager to start my own university group, EA ELTE. After a stressful 2 weeks of marketing, I got ~20 signups from students at ELTE. At the same time, I met a student at a LessWrong meetup who was from a different university. She was interested in learning about EA, so I invited her to participate in the intro fellowship. She invited 3 friends from the same university, so I ended up facilitating 4 groups of students from my university, and one group from this other university. I also ended up organising the socials and the end-of-fellowship meal together, as I didn’t want to, nor have I had the time to organise a separate event for just 4 participants. In the next semester, again I spent most of my time marketing at my own university, but we again had a couple of people trickling in from other places, and of course, we took them in. After that, I just started making a city-wide, fellowship poster in addition to the university-specific one, figuring that we “have never really been a “one university-only group”, and more like a city-wide university group.

Here are the pros and cons:

Below I will outline how I see the pros and cons of this approach and what I think happened in our case. Note that whether this is preferable for you will to some extent depend on your own theory of how CB should work, so I recommend evaluating each item for yourself. To help with this, I also added what I see the cruxes are. I welcome you to write a comment if your cruxes are different!

The PRO(s) of doing this:

1. Bigger pool for program marketing

You should be able to get more applicants, as your fellowship is not limited to people from just one university. Budapest is a city of ~2 million people with 5-6 universities that I think could eventually have promising EA groups. With some caveats (see below) I would say the more universities your city has, the better.

The cruxes of this for me are the following questions:

Crux 1.1: How soon do your marketing efforts hit diminishing returns for a given university? 

If spending 20 hours on marketing at a specific university won’t yield a lot more applicants than spending only 10, then you are probably better off shifting your focus to another one, if you can.

If there is low-hanging (marketing) fruit to pick up at different universities, then you should probably prioritise that. 

For example, I felt really jealous (in a good way!:) ) when I found out that student societies at UW-Madison can email every student about their program. If universities in your city have something similar that you can easily get into, then prioritizing this is probably more useful than emailing individual professors to let you pitch the course before class.

One thing I would highlight though is that whenever it is appropriate, you should still market yourself as a university-specific group.^[2] This is because students will probably be more excited to join stuff that is specific to their university. I think you will also be more likely to get permission to eg. put out a poster in a dorm as a university-specific student group, than as a city-wide student group. 

On the other hand, you want to make sure you are not misrepresenting what you are offering. I think saying that you are university group A, which is part of a wider network of a [city’s name] university group should be fine. You should try to put applicants from University A in the same discussion groups, but you also want to clearly communicate to people who sign up that this is a city-wide student project, and they might end up in a group with students from other universities.

Crux 1.2 Can you use social media ads to effectively recruit participants for your program?

I think your answer to this question should to a large extent determine whether you would want a city-wide university group or not.

I highly recommend testing social media ads for your context, as this has been one of the best ways our group got diverse, highly motivated applicants for our program. (I plan to post about this in the near future.)

Crux 1.3: How many places can you cold email?

If you are going to do outreach at universities you are not connected with, that means you will have to write more cold emails (eg. to ask them to let you put out posters) compared to relying on personal connections (eg. a professor you are on good terms with will let you pitch the fellowship before class). The number of replies you get per hour of work put in should determine whether this is worth it for you to pursue. 

One thing I recommend doing is to try to put some status markers in your cold emails. After all, they don’t really know anything about you, so any information you can show that points you towards being legitimate/prestigious should be useful. Of course, you shouldn’t misrepresent who you are, but luckily you don’t need to, as EA/AIS has plenty of status markers you can use! In my emails, I tend to mention that the course we are running has started at elite universities (and we are using the same curriculum). In our group, the person facilitating most of the AGISF technical courses studied at an elite university and got into SERI MATS, so I mentioned her involvement there as well.^[3]

Crux 1.4: Which universities and programs will you get high-quality applicants from?

There is a lot that can be said about predictors of engagement with EA,^[4] but I think it is agreed that talent and altruism play a major role. As a rough guide, I would recommend prioritising the top universities in your city (other things being equal), while keeping in mind potential diminishing returns (as discussed above).

One caveat I will add is that at least in Hungary, there aren’t any selective universities. Instead, it is specific programs that are hard to get into within each university. If you have a similar system, I think that is a positive (but in itself not sufficient) argument for opting more towards a city-wide uni group.

2. Inclusivity

Even if you focus most of your marketing efforts on a specific university, it’s nice if people who randomly find EA from universities without an EA group can join something. You can also buy social media ads, which should attract people from more than one university (and as far as I know you can’t limit them to just one university). 

3. You can speed up seeding other university groups

The way I see this model work is at the beginning, you could get university students from all over the city, and those who become very excited can eventually start a uni-specific group.

You will be well equipped to support these groups, as you have already “felt the waters” for a given university and have some sense of what outreach works or doesn't. Just note that currently, CEA’s UGAP program is the main way uni groups are supported, so you want to make sure to coordinate with them.

4. You don’t necessarily have to be a university student to start such a group

If you already graduated and/or just moved to a new city, you can still do outreach to university students without being affiliated with a specific university. I would recommend recent graduates consider this as an option, as (other things being equal) you are in the best position to support students at university.

The CON(s) of doing this:

A) Group cohesion

Students might be able to connect more easily with others from the same university, as opposed to students from other universities. Students might also feel safer knowing that the only other people at the event will be others from the same university.

Crux A.1: How much does it matter which university a student is from?

I sense that it matters a bit, but this does not outweigh the PROs of having a city-wide university group. I think it is likely that this is more important for students in elite universities.^[5] It is also possible that it is more important for students at the top universities of a given country, but I think likely to a lesser extent)

Having said that, I would be extremely grateful for you (yes, you!) to share your experience with this, even if it's anecdotal.

Crux A.2: How scary is it to attend a university-specific social compared to a city-wide student social?

This is somewhat covered in Crux 1, but I thought it would be worth touching on it separately. Again, I think this matters to some extent. We haven’t been doing a good job of getting people to socials, but I’m not sure how much of that is due to having a city-wide uni group, as I can think of a lot of confounding factors.

 I think for people who seem shy and/or whom you are very excited to have at the social, you should give them a friendly personal invite either way.

B) Blurring the line between university and city groups

Keeping this in mind is something that I feel I didn’t do an excellent job of. Given that we ended up getting students from several different universities, at first, and (partly because I was the only organiser), I just opted for having city-wide socials, where both professionals and students were invited. I feel like this didn’t work really well, for a couple of reasons. 

Given that in our city we weren’t doing outreach towards non-students, there were only a handful of professionals in our group, who mostly found out of EA on their own several years ago, often through reading LessWrong. This meant that those members were already really familiar with EA/LW ideas and often spent years “in the trenches” debating moral philosophy etc. As a result, the inferential distance between the two groups was quite big, and newcomers ended up feeling like they didn’t know enough, while the professionals couldn’t enjoy complicated/weird/fun discussions. 

On the other hand, you kind of want your new members to at least eventually meet with older and/or professional members, as they can benefit a lot from connecting with these groups.

Crux B.1: How accessible is the city group to university students new to EA?

Perhaps you could have a separate meetup for both students and professionals. Of course, life is not that easy, as you have limited time, and might not have a critical mass for one or the other group.

Generally, I would be less worried about blurring the lines if the older members of your group are also new to EA, or if experienced members are especially mindful that they are talking to newcomers. If you end up organising joint socials, I recommend reminding people of this at the beginning of the event. I think the best of both worlds can be having separate meetups for students and professionals, but inviting a couple of people from the other groups.
 

C) Potentially Crowding out other organisers

If you end up having a city-wide group, it’s really important not to make the impression that you/your team have a monopoly on community building in the city. In terms of motivating people to get involved in community building, whether on a voluntary or paid level, I think it is likely important for people to feel like they own a project or idea. “Starting your own university group” just sounds way cooler than becoming the “Xth volunteer for a city-wide uni group”, especially for highly ambitious people. 

To mitigate this, I think you should make sure to reliably display that people can and are encouraged to start university-specific groups and that you are also happy to support them. Specific ways you can do this are 

1. Encourage people to apply for UGAP.
2. If you have a newsletter, in addition to a call for volunteers for your city-wide uni group, you should have a call for university group founders. 
3. If your country/city has a website that lists eg. Uni group A, Uni group B, and your city-wide uni group, then as an additional slot you could put “start your own group”, with a call to action.
    

D) Location of your events

At some universities, you can get your own office as a student society for free, or at least use get to use some rooms for your fellowships/socials. I think this is cool, and it can save you a lot of money! Having students from other universities attend there could make things a bit more complicated, as university staff might be less excited / more worried about people from outside of the university using their infrastructure. I would recommend checking in about this with the people who are allowing you to use these spaces.

At my university (ELTE) we weren't able to get an office and the room that I was able to use was not ideal, so we ended up having to pay for room booking outside of the university. On the other hand, we recently figured out that we can use libraries for coworking (some libraries have spaces/floors that allow for semi-quite chatting), and some even have rooms that you can book for free if you are a student - which you can use to hold your fellowships in.

What did I miss?

If you can think of other PROs and CONs, please mention them in the comments! I would be open to editing the post to add your comments!

Should you do this?

My view is that all things considered, having a [City's name] Universities Group is preferable to one university-specific group if the context is right and you are mindful of failure modes.
Even if you are convinced and want to do something similar, I highly recommend you coordinate with CEA and talk your plans through with your mentor. As far as I know, the only groups that do this are us in Budapest and the London EA Hub has a similar system. This means that we have much less data on how to make this model of CB work, including data about potential pitfalls that we have not encountered so far.

The best of both worlds

Now that we have a city-wide uni group that is running programs every semester, I would like to start helping seed university-specific groups in the city. This way hopefully we will be able to mitigate the cons while having a strong team in the background that can funnel members to the university groups, help prevent them from going dormant, and help out with operations so they have more time for meaningful activities. Wish us luck!

Thanks to Annake and Milán for rewieving the draft of this post.

1. ^{^}

   Eötvös Loránd University (ELTE) has ~28k students. It is considered as one of the top universities in Hungary, however, what "top" here means is different than in the UK/US. Here top universities can be still very easy to get into (even with a full scholarship), depending on which specific program you are applying to within the university. Some are very competitive, some are less so. This means that for some programs of lower ranking universities, it is harder to get into than many programs of higher ranking universities.

2. ^{^}

   For example, if you go to University A, I think you should ask for dormitories at University A as a university-specific student group, and write the posters accordingly too. You should consider having a separate sign-up form for students from University A, either through your website or your social media page.
   If you don’t go to University B, but you have someone in your group who is excited about EA from University B then you could either cowrite an email with them as as a student society from University B, mentioning that you are part of a network called “[City’s name] universities”.
   If you don’t go to university C and don’t have anyone you know who is involved in EA from that university, you can email dormitories as “[City’s name] universities” and ask for permission to put out your posters.

3. ^{^}

   One thing I'm also considering is asking permission from EA groups at elite universities to allow us to put out some indication on our website that, well, they exist and we are doing similar stuff to them. This one I feel the iffiest about, as we are not too much in collaboration with them, but I think showing that there are similar groups to us in elite universities would be pretty valuable in cold emails, especially in second and third-worldCrux 1.4: Which universities and programs will you get high-quality applicants from? countries.

4. ^{^}

   Including questioning whether that's something we should try to optimise for, but this is outside of the scope of this post. 

5. ^{^}

   I wouldn't recommend someone to change an established university group at an elite university into a city-wide university group. My sense is that at those locations there would be enough excited people to run a city-wide uni group in addition to the uni-specific groups that already exist. Of course, if people join events from the uni-specific group, that can be a plus!