My (working) Model of EA Attrition: A University CB Perspective
My (working) Model of EA Attrition: A University CB Perspective
Background/Why I’m writing this post:
I’ve been co-organizing an EA student group at Queen’s University in Canada for about a year now. When I first joined Queen’s Effective Altruism (QEA) on campus as a general member in January of 2023, the club was small (only ~5 active members and the only HEA was the sole organizer). The following semester, when my fellow co-organizer and I inherited the club, we were pleasantly surprised with the number of members joining QEA—we attracted 40 members over the semester, with an active member base of ~17. Everyone seemed pretty interested and engaged, and we managed to send a total of 8 members to various GCPs and EAGs. We noticed some attrition, but not enough to pass it off as anything more than what was expected (in fact, our club collected more members than we lost over the semester.
Then winter semester rolled around. All of the sudden, our active member base fell to ~10, albeit we absorbed some of the previous EA members into exec roles. Our attrition rates were also notably higher: By the end of the semester, some of our introductory fellowship meetings, which initially attracted around 17 attendees, dwindled to just one or two participants.
This difference seemed significant. Of course, there are many expected variables (winter semesters tend to involve higher attrition rates, the recruiting period is less bustling than the fall recruiting period, etc.). But, with the knowledge of the success we had in the fall, the contrast was stark and rather confusing, especially since we put in the same—if not more—effort into the club during the winter semester. Some of the members who we thought would stick, didn’t, leading to even more confusion.
I’ve grown increasingly interested in attrition—namely, questions related to what qualities the people who reject/tend towards EA ideas possess, what factors contribute to attrition, and how to discern whether it reflects issues within our university club or is just noise.
As I’ve thought about this more, I’ve come to discern some patterns in attrition, which, I think, hint at the potential for a comprehensive model. Developing mental models has been incredibly helpful for me. When I notice confusion about something with no clear way out—no model to grip onto and make sense of the thing—I succumb to inertia. The same can be said about this model of attrition: The more I cultivated a model, a mental heuristic to employ when considering membership for QEA, the more clarity I’ve had about thinking about membership, and thus the more confident I felt about taking action according to it. This post aims to go over some of these trends/little things I’ve noticed that I think are helpful to be aware of as a university community builder.
Note: I’ve been uncertain about the usefulness of this discussion. Much of what I’ve noticed might seem self-evident, but maybe this is due to the illusion of transparency or the curse of knowledge fallacy. Regardless, I’ve concluded that I think there is at least no harm in posting this. At its best, I think this post could open up further discussion about attrition in EA broadly—not as a means of reducing it, as attrition is healthy and normal, but as a means of having a better prediction model.
It’s also important to clarify that this model isn’t meant for preemptively judging potential members. Instead, it aims to provide community builders with strategic insights—identifying where to intensify or ease recruitment efforts based on understanding likely member engagement. This approach isn’t about convincing everyone of EA; rather, it’s about efficiently selecting and connecting with individuals who are naturally inclined toward the principles of EA. Gaining even a basic understanding of who these individuals might be can better our focus and effectiveness as club leaders, ensuring our energies are well-positioned for optimizing our CB efforts.
My (very rough) model of what repels/attracts university students to/from EA:
*This is mostly based on conversations with QEA members of varying commitment levels—so all anecdotal, qualitative observations, with a small sample size.
Of course, there are overarching factors that largely do not lay in the control of community builders:
Wider economic incentives, self-interest: Duh...the opposite of altruism...makes sense. People who are more money-motivated or driven by other more superficial factors like status and so on. One moderately engaged QEA member hit me with the classic “play the game or the game plays you” to explain his focus on accruing personal wealth more than donating significant chunks of his income. This is largely the product of the wider structural incentives underpinning our (Canadians) socialization.
The factors I’m more interested in are those that come down to individual factors, like personality, interests, dispositions, etc. Here are some other factors and/or general considerations:
Lack of intrinsic intellectual interest: Some people just aren’t as intellectually motivated as others! It is evident that the sorts of people who tend towards EA are people who also enjoy the intellectual “game” of it—pondering EA ideas, continually updating, etc. If a student doesn’t seem to possess this kind of intrinsic curiosity, they’re probably less likely to dive further into EA in their spare time.
Convenience/cherry-picking: Some individuals engage with EA selectively, embracing aspects they find enjoyable or beneficial while disregarding elements they perceive as inconvenient or broadly unhelpful. For example, members might be enthusiastic about pondering the arguments for global health or effective interventions, but stray away when it comes to the practical commitment of personal donations.
I think some people also find the obligatory undertone of some of the donation discourse unsettling. This feeling of compulsion may detract from the otherwise voluntary spirit of EA, making it seem less appealing.
Being more impressionable/less independent: University-aged individuals are still trying to find their footing in the world, and because they tend not to have yet formed a clear identity, they latch onto their inner circles as a way of finding meaning/identity. Students are then very influenced by their inner circles/friend groups. Those who place a high priority on fitting in—whether consciously or subconsciously—may find it challenging to deviate from the norm and maintain their commitment to EA (value drift is real!).
This also relates to the idea of bounded ethics/consciousness: The idea that our capacity to make ethical decisions is frequently limited by both internal and external pressures. Environment is perhaps the largest factor contributing to the degree one is “bounded.”
Different thresholds of obligation: Members have expressed vastly different beliefs about where the line of obligation (that they have to others) is drawn. Some are really sympathetic to the drowning child thought experiment, others are not. For example, a popular opinion raised in meetings is that our obligations are prioritized in a sort-of “concentric circle” format, starting with our immediate family, and any responsibility beyond that is considered supererogatory. Some reject the concept of obligations altogether. Those who believe we do in fact owe things to each other tend to be more passionate about EA.
I also haven’t noticed university-aged EAs being as driven by shame/guilt as older EAs I’ve met. I have nothing to substantiate this idea with, it’s just a quick observation I’ve made.
The degree to which someone is naturally “rational”: Many people don’t think in a quintessentially “rational” way, and so have a hard time relating to EA discourse. Any idea that includes “optimization” or rational thinking in the classical EA sense seems to be alien to more ordinary university students. I’ve found it challenging to bridge this gap and make the principles of rational, optimized decision-making accessible and relatable to those who do not instinctively gravitate toward such approaches.
I’m having a difficult time articulating this point…hope it makes some sense.
People enter with different ideas of altruism, and attrit once they perceive EA as pushing back on their initial altruistic ideas: Some members often join with various understandings of what constitutes altruism; Many altruistic students are motivated by traditional forms of altruism such as climate change initiatives, food banks, or cancer charity runs. It goes without saying that while each is a noble and necessary cause, core EA principles like impartiality and cost-effectiveness can sometimes seem threatening to their existing altruistic views.
More specifically, this manifests in a kind of disillusionment, where members passionate about endeavours not at the top of EA’s cause prios regard EA as being broadly unsympathetic (and perhaps too utilitarian).
As we know, emotional narratives and personal stories often drive altruistic behaviour. EA’s sometimes analytical approach might not resonate with (and may even be seen as abhorrent) those who are motivated more by emotional and personal connections.
I have many more ideas but these are the most thought out/helpful ones.
Other things worth saying:
I am not at all trying to paint non-EAs as bad people (and doing so would actually be pretty detrimental to the ethos of EA). We are not intrinsically morally superior; understanding and humility are key! I even think back to my younger self first joining QEA: While I probably tend more toward Peter Singer-type obligation ethics ideas more than the average person, I also think I was just as susceptible to attriting from EA back when I initially joined. All it really would’ve taken was a dislike of the other members to stop showing up. The act of continuing to participate in EA ideas is quite complex, and I generally don’t think blame helps push the needle in expanding the EA community.
Self-selection effects are real: Particular school cultures also obviously come into play. Queens’s University, while maintaining a good academic reputation in Canada, is also well-known for its vigorous party culture, and so the students pouring in every year may on the net be less classically “EA.” Schools with more academically rigorous reputations might have a disproportionally higher sample of people who would be better suited to EA
Please reach out if you have any thoughts on this :)
Solid piece. I like lists of things and I appreciate you taking the time to write one.
I sometimes wonder how to combine many qualitative impressions like this into a more robust picture. Some thoughts:
Someone could survey groups on attrition rates
Someone could ask how many people group leaders recalled who were in each group type
Hi Nathan, I’m one of the co-organizers along with Juliana and I’ve thought a lot about quantitatively measuring attrition rates and the types of people more interested in EA.
We found it hard—at least on the level of one club—to measure things like attrition rate for a few different reasons:
First, there are so many factors that may cause someone to attrit from a club, ranging from them not being a good fit to them simply not having the time (as @DavidNash mentioned).
These factors also are so variable: as Juliana mentioned, if she didn’t like interacting with her group (me especially) as much, she probably wouldn’t have continued engaging in EA. One bad session or rude person in your group may make the difference between a person continuing to engage or leaving.
These factors all also have such small effect sizes and also intersect with each other, to the point that it becomes very hard to suss out any sort of causality or clear picture from quantitative data.
This is where (in my experience) these more qualitative approaches become more useful, as we are naturally attuned to gathering peoples’ impressions.
Second, there’s going to be a lot of selection bias: most people (at least I would suspect) in most universities aren’t interested in EA (again, for many factors) and so any confusion matrix of people ‘interested/not-interested in EA’ and ‘whether they join’ is going to be very heavily weighted towards ‘not-interested x not join’. And so the vast majority of people who come to an information night or express some sort of interest are not leaving for any particular reason, but rather because they just aren’t interested in EA [enough].
But as you mentioned, there may be a lot of value in sort-of qualitatively-quantitatively measuring attrition rates on the scale of CEA—instead of trying to find reasons as for why people are not staying on a group-level, the CEA Groups team could survey reasons for why group organizers think people leave, and perhaps use that to create helpful resources.
One category that you didn’t include are people that agree with the ideas and take action, but don’t want to or are too busy to attend lots of EA meetups.
Hi David, I think that is actually quite a big factor! I noticed in particular that there are people in our group we think are particularly well-suited for EA but don’t have the time and/or energy to engage. These sorts of people agree with the ideas, are motivated to make an impact, and also have sufficient work ethic to do so, but can’t make it to the meetings precisely because they’re too busy making an impact. I personally think these people are the ones we want to expose to EA ideas, but it is difficult to engage them.
We have a couple different ideas for engaging these sorts of people more:
First, having a strong community and friend-group associated with our group creates a social obligation (for lack of a better word) to attend, and also makes it more fun to attend. The people in this community will also have lots of EA-adjacent people in their network and/or other friend-groups, who would also be exposed to EA ideas.
Second, having low-commitment events at non-busy times of the year (e.g., a social event early in the semester) means that those sorts of people are more likely to attend. This also introduces them to the ‘community’ I mentioned earlier, which allows them to be more engaged.
Third, as you mentioned, we also find a lot of value in having a more ‘wide’ network of EA-affiliated people on our campus, rather than just having a tight-knit community and encouraging people to join. Not everyone is looking for an EA-style community, but may still want to engage with the ideas, or want career advice (we’ve noticed a few people like that). This also allows people to move in and out of the wider ‘network’ and tight-knit group with ease, (hopefully) reducing attrition.
Executive summary: The author presents a working model of factors that attract or repel university students from Effective Altruism (EA) based on their experience as a co-organizer of an EA student group, aiming to provide insights for community builders to optimize their efforts.
Key points:
The author noticed significant attrition in their EA student group membership, prompting them to develop a mental model to understand the factors contributing to this trend.
Overarching factors like economic incentives and self-interest play a role in attrition, but the author is more interested in individual factors like personality, interests, and dispositions.
Lack of intrinsic intellectual interest, convenience/cherry-picking of EA ideas, being more impressionable/less independent, and different thresholds of obligation are some key factors that may repel students from EA.
The degree to which someone is naturally “rational” and their preconceived notions of altruism can also impact their engagement with EA ideas.
The author emphasizes the importance of understanding, humility, and avoiding blame when considering EA attrition, acknowledging the complexity of factors involved in continued participation.
School culture and self-selection effects may also influence the likelihood of students being drawn to or repelled from EA.
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.