Hide table of contents

As a result of hosting the People of Colour meetup at EA Global London 2019, I (Vaidehi Agarwalla) conducted a community survey on racial and ethnic diversity in December 2019. The goal was to gather peoples’ experiences and thoughts on the topic to share with CEA and the community to inform future Diversity and Inclusion efforts.

A big thank you to Sky Mayhew from CEA and many others for providing valuable feedback and edits. If you have any questions about the data please email Vaidehi at vaidehiagarwalla@gmail.com.

You can view the full report here.

Key Takeaways

  • 94 respondents consented to sharing their data, 27 (29%) of whom identified as a non-white race or ethnicity. A breakdown of respondents and key question responses is in the table below.
  • The top two reasons for thinking ethnic diversity is important are:
    • missing potential members, and
    • missing important perspectives that would influence our actions
  • The graph belows shows a summary of reasons for thinking ethnic diversity is important:
  • The main observations from participants’ experiences with race/ethnicity were:
    • Many non-white respondents hadn’t personally had negative interactions*
      • *Of the 3 respondents who shared personal negative experiences relating to their identity as a nonwhite person, all either currently or previously worked at EA organisations.
  • Respondents shared many recommendations; they clustered around the following categories: creating space for discussions about diversity, targeted outreach, changing communication of EA ideas, improving the global community, soft or hard quotas, and improving awareness and training. The summary table below shows the full breakdown:

  • The survey offered to connect anyone with a negative experience to CEA staff for support. No respondents requested this support.

About The Data

Survey responses were collected from November to December of 2019. The survey invitation was advertised on the EA Forum and the following Facebook groups: Ethnic Diversity in EA, EA Hangout, and the Effective Altruism Facebook group. The table below shows the racial and ethnic identification of the respondents by question:

Clarification on Identification with non-white race or ethnicity: Yes = The respondent identifies as non-white race or ethnicity, No = The respondent does not identify as non-white race or ethnicity, Complicated = The respondent gave has a complex relationship to identification and/or is mixed race.

94 people took the survey and wanted their information shared with the broader EA community and CEA. 31 (33%) respondents identified as non-white or complicated (mixed race). The goal of the survey was to solicit the experiences of people from ethnic minorities in EA. The survey has a slightly higher proportion of respondents to this survey were non-white. It has a much smaller sample size, but due to the niche topic and limited advertising it was unlikely to be high. I believe the survey can still provide insights because the survey questions were not asked in the EA survey, and no previous community surveys have covered this topic in the EA community. The table below compares this demographic identification of this survey to the 2019 EA survey.

Edit: The EA Survey has closer to ~20% respondents who would identify as non-white race or ethnicity, although these numbers are uncertain. See David Moss's comment below for more detail.

To contextualise these numbers, I’ve included this table of potential reference classes for EA. There are likely relevant reference classes missing.

Please see the full report for the reference class sources.

To keep the survey short and accessible, I didn’t ask for any other demographic data. However, people mentioned some other demographic details in their comments, summarised below:

  • At least 3 (4%) people worked at or used to work at EA organisations
  • At least 7 (7%) people had attended EA Global
  • 2 people (2%) mentioned university groups and 8 people (8.5%) mentioned that they were in city groups.
    • Number of people who mentioned city groups: London (3), Sweden (3), Germany (1), Singapore (1).

Survey Questions & Data

You can see the survey questions here. You can see the scrubbed and coded data here.

Terminology

Throughout this report, the term “non-white” will be used to discuss respondents who identify as such. This is to reflect the phrasing of the question, which asked for identification with non-white race or ethnicity. It’s not an ideal term, and I do not intend to set a norm or default.

There is currently no consensus on an ideal term. The term “people of colour” does not cover people outside a Western context. Other terms that may be used are “ethnic minority” or “global majority”.

Limitations of the data

  • This was an initial, exploratory survey. The sample size is only 94 participants, of whom only 27 identified as a non-white race or ethnicity. It didn’t ask about all the possible variables in a person’s experience that could help eliminate confounders.
  • This survey only went out to people who engage on the EA Forum and EA Facebook groups, which might exclude certain groups of people. I am unsure how this biases the sample.
  • It is unclear how strongly either of the following biases skew the results:
    • People who left the movement due to negative experiences. This group will be difficult to track since so it is difficult to assess how many such people there might be and therefore how they might change the results. It seems that of the non-white people currently in the EA community, there don’t seem to be many people wish to leave. I have anecdotally heard of a few people who’ve left or reduced involved in the community, but not particularly those who are non-white.
    • People who care more strongly about ethnic diversity are probably more likely to take the survey. I am fairly confident in this bias given that it appears to be true across a range of topics.
    • People who may have been turned off by lack of diversity in the movement are unlikely to have seen or taken this survey. It is likely there is a large group of people who may have had valuable insights but were turned off EA initially and would be unlikely to fill out a survey about this. Therefore, this survey does not provide good information on how EA might turn people off when they first come across it.
    • Anecdotal or other observations on any of these biases would be helpful.

Results

For the questions with open-ended answers, I have tried to include a representative and fair sample of both commonly expressed sentiments and individual’s experiences that seemed particularly salient. In general, I’ve erred on the side of including more quotes rather than fewer, I’ve added summaries and bolding to help the reader.

Q5. Why do you think ethnic diversity is important?

This was a multiple choice question which 86 (91%) of respondents answered. Respondents could select multiple answers and/or suggest their own reasons. The graph includes shortened versions of the answer choices; the full text of each answer choice is below, in addition to respondent comments.

The graph shows the top reasons stated, and the proportions of each answer by identification. A total of 53 people answered this question, 36 white and 27 non-white and all 4 people in the “complicated” category.

Description of Reasons

For 59 (68%) respondents, the top two concerns about lack of diversity were missing out on potential members who would have otherwise joined the movement, and missing out on important perspectives that could influence our beliefs or actions. The next 3 reasons were selected by between 49 (56%) and 53 (62%) of respondents. Although the difference is not large, it suggests that on average, respondents were most concerned about the interpersonal issues relating to welcomingness and losing out on the diverse perspectives and input of potential members. They were slightly less concerned about the ability of EA as a movement to have an impact.

Interesting observations

  • The average percentage of non-white respondents across all the original options (excluding “PR issues or will be highly criticized”) was 26.7%. Some outliers include:
    • Non-white respondents were more likely to be concerned about “Missing opportunities for impact”, with 34.7% of them picking this option.
    • Only 12.5% of respondents (1 person) who selected this option “I don’t think it’s important” were non-white.
  • No non-white respondents raised the issue of PR issues, although several white respondents independently raised this point. However, since it was not one of the selectable options, I would expect more people in general to have chosen this option if it were offered.

Comments on question phrasing

  • See the survey for how the question was phrased.
  • The first reason, phrased as “We could be missing a lot of potential members who would otherwise have joined the movement” could have been stated more broadly so as not to restrict who we are missing to just those who identify as members of the EA community. A better way to phrase this would have been “potential talent” or “potential talent pool”.
  • In this question, I gave a list of answer choices so that it would be quicker for people to fill out. However, I forgot to add “PR” reasons, so it’s likely that more people would have chosen that if it had been an option. This survey platform also didn’t have an option to random shuffle the order of answer choices, which is one way to reduce response bias. That may have changed the frequencies of responses. Items listed first are sometimes selected more often just because of ordering effects; however, since there isn’t a large difference in the popularity of different options by order of appearance, I think this wasn’t an issue for this question.
  • I also didn’t ask respondents to rate the relative importance of the different reasons why diversity is important, to keep the survey short. With a rating, it’s likely some options would have been more popular than others.
  • Someone commented that this question was an example of “push polling”. I agree that I could have phrased the question more neutrally. If I redid the survey I’d say “Do you think ethnic diversity is important”, and then provided options for both “yes” and “no”, or I would give both positive and negative reasons and ask people to state which of these they agreed with.
    • Another option would have been to conduct focus groups or a series of interviews to better inform the answer choices offered on the survey. However, due to time constraints this was not feasible.

Q4. Experiences

57 out of 91 survey respondents shared comments about their experiences related to ethnic diversity. 41 (72%) of these comments were from respondents who identified as white, and 16 (28%) were from respondents who identified as non-white or “complicated”. I will consider these latter two groups together, which comprises 13 who identified as non-white and 3 who were classified as “complicated” and refer to them as “non-white” in this section.

Experiences shared by non-white respondents

The following is a subset of the thoughts people shared. All bolded emphasis is added. Quotes that begin and end with an ellipses are excerpts from 4 longer quotes, which you can read in full in the appendix.

One common thread in the responses from the 14 non-white respondents was the idea that ethnic diversity was an important issue, but they experienced few negative experiences themselves:

I’ve never really felt out of place due to my race in EA settings, although I guess I have noticed that CEA is all/mostly white. Some other orgs like OpenPhil seem to be more diverse. I don’t personally feel too much discomfort about being a POC in EA, but the lack of diversity (not just in terms of race but socio-economic class, educational background, geography, etc. does concern me a bit (with respect to our epistemics, how we think about global EA strategy, understanding of issues related to oppression, colonialism, marginalization, and other topics. Lived experience (or lack thereof) `can significantly impact our ability to understand/perceive various societal/global dynamics which may be quite important to think about in the context of EA as a global movement.”

“I have not personally felt any sort of exclusion on the basis of race. However, I generally do not experience many things like that and I would not be very surprised if my experience were not representative.”

“...I do want to say that I'm grateful that I've never had any explicitly negative kinds of experiences within the community (something that sadly cannot be said about other places or daily life), and that virtually all EAs I have ever interacted with have been warm or at least "neutrally nice".

“Sometimes I notice when I'm the one non white person (or one of few) at an EA event, but that's not happened very often in the university groups I have been / am now part of. I've also been mostly been insulated from any serious negative experiences as someone running an EA group. I was worried for a while though that the US college group I was running seemed to skew white, or otherwise East / South Asian -- I think it's better now but it felt hard to actively encourage ethnic diversity and I still don't really know how to do it in EA without potentially making underrepresented groups feel uncomfortable / singled out.

One person did not feel race was an important part of their life.

“I am technically not white, though for me ethnicity has never been an important concept, either for defining my identity or for explaining my interactions with other people. / I'm open to changing my mind on this issue, but at the moment I feel that ethnic diversity is not particularly important, especially in comparison to other forms of diversity, such as cognitive diversity.”

Another theme was discomfort raising dissenting or unique viewpoints if you were the only minority in a group:

I have generally had positive experiences when discussing diversity issues with EAs in Singapore and London. My biggest concern is that being a very visible ethnic minority in a group creates a chilling effect where I feel diffident about offering differing views, for nervousness that I might be regarded as the guy who simply “doesn’t get it”. If it’s a major problem I have no issue speaking up. But on less serious (but still important) topics, a vital perspective may be lost because the cost of disagreeing as an “outsider” isn’t worth the lost social capital I know I need to build in this community.” / "I have personally not had any negative experience with this, and have always been heard out with respect whenever voicing this perspective out. Nonetheless, there will always be many competing priorities and this ought to be one."

“EA London have tried various times to have discussion groups on the issue, one of which I lead, but it didn't really go anywhere, and I didn't feel comfortable being representative of people of colour generally so was reluctant to make too much noise about it. EAL [sic, EA London] is overwhelmingly white though especially when compared to the population of e.g. London universities”

Negative Experiences

3 non-white respondents shared personal negative experiences (37.5% of all non-white respondents who shared experiences). All three worked at EA organisations. This could be because for those not working at EA organisations, EA is not as big a part of their daily experience, and so negative experiences in the community are both less frequent and less salient to them. It’s possible these experiences are not unlike other experiences at similar organisations or companies.

Not feeling taken seriously because I'm a woman of color. I don't appear as nerdy and I feel that my intellectual ability is in question. This is further complicated by men finding me attractive and it can be hard to find a purely professional connection. My manager at an EA [sic] for example would see my criticisms to the effectiveness of the org as me being sensitive, rather than engage with the logic of the argument. I was called out by him for wearing leggings.”

“One person commented to a friend that he assumes women who speak English with an accent are less intelligent.
One person said to me that diversity in workplaces is like "giving bonus points" to people who aren't as good on their own merit.
One person said to me he thinks brown women are less competent.
One person said to me that he thought the job he was advertising for looked for skills that were predominantly found in men and so it was just more likely he would hire a man.
In general, I find the public conversation on racial diversity in online EA forums pretty unhelpful and too combative. I avoid engaging solely because I don't have the energy for the confrontation.
In general, I find conversations on diversity in online EA forums demonstrates some pretty poor understanding of basic issues and I don't want to engage and have to teach people basics like: people of colour have different experiences than white people.

“I am non-white, have been part of the EA community for a couple of years and have worked at CEA in the past. The most common kind of experience relating to ethnic diversity that I've had within the EA community was simply this feeling of "out-of-place"-ness when the vast majority of faces you see are white. This sentiment is at times exacerbated when you're in a more prominent role, e.g. when being in an organizer or facilitator role and a sea of white faces looks at you, which just made this kind of homogeneity within EA all the more visible.”

Empirical Observations

Many made empirical observations about the lack of diversity in EA -- either via observations in a local group, EAG, or online. A good portion of these observations were purely descriptive with no additional context. It is likely that the respondent is making a negative observation, however, without additional context I do not want to overinterpret these statements.

Descriptive observations without additional context:

“I have noticed little ethnic diversity in the EA community.” (several similar)

Descriptive observations with more context:

“There is not much of it! I remember being at EA Global in SF last year and it was one of the whitest gatherings I've ever been at in the Bay Area”
“In EA Sweden (at least at Stockholm events) there is medium to high variety in which countries people are from, but nearly no-one belonging to an ethnic minority. Although that might be because personal identity/background is rarely brought up during meet-ups (neither one's own nor others') and therefore my estimation of diversity or lack thereof mostly comes from differences in appearance and name.”
“The majority of EA things I've attended have been non-representative in the sense that most people are white or Asian. No experiences beyond that, though.”
“I ran a local group in the UK for two years and we had more ethic [sic, ethnic] diversity in our group than the background population.

Many of the above comments also highlight the intersection of ethnic diversity with other issues like socioeconomic class, education and gender. Here are some more quotes on these intersections:

“I am a white male, but live in a relatively poor country in eastern Europe. Oftentimes, I feel excluded from the broader community life and discussion.
“There's literally no diversity in our student group - a few girls and mostly boys who are all white and fairly well off.
“Whenever I've been to EA events (EA Global London, EA London meetups, Giving What We Can meetups), the community appeared to be dominated my [sic, by] young white men.” / "The same problem with low levels of ethnic diversity is felt my [sic, by] many organisations, including the Quaker community that I'm a part of.
Being mindful of the ethnic diversity of the EA community is very important for the future of the movement because it will need to learn to encompass a wider demographic rather than remaining a niche tribe which I would argue it currently still is."
“It’s a monoculture of Oxbridge white men. Feels a bit like cultural imperialism, especially in global poverty focused on Africa, to have few black EAs.”
“...When I hear people in EA talking about issues related to global health and poverty, I often worry about the language being used to describe the people who will benefit from a programme - I commonly hear words like 'help' or 'make them' which suggest a strong power dynamic where the recipient is passive and not in control. This is a slightly separate issue but I feel it ties in a lot with interactions between people of colour and white people via the relationships between the global south & north. I worry that it might make a person of colour feel unwelcome in a discussion. ..”

Other negative experiences relating to ethnic diversity included examples of what respondents deemed to be inappropriate discussion topics, ignorance when talking about the developing world, one comment on the negative impacts of diversity efforts, and other anecdotes:

“Extremely inappropriate discussions about white/black IQ, Derailing of discussions about ethnic diversity issues to 'diversity' issues of political persuasion ('but do we discriminate against right-wing people?'), language ('should we translate EA stuff into e.g. German?')”
“… [on IDInsight’s new GiveWell report on value of life in Africa] “More to the point, it draws attention the fact that a group of a dozen or so generalist researchers with limited adult life experience in one of the wealthiest parts of the world are LITERALLY ASSIGNING DOLLAR VALUES TO THE LIVES OF POOR PEOPLE IN AFRICA. And while I understand that is an unavoidable value judgment that we all implicitly make anyway, it nevertheless highlights the highly problematic nature of not making more of an attempt to incorporate the perspectives of those affected in these calculations, and how much reputation risk (=impact risk) EA is exposing itself to if these underlying issues are not addressed”

“I've also seen several instances of a white person in EA confusing two people of colour eg thinking they have already met someone or have had a particular conversation with one person when they actually spoke to a completely different person who happened to be of the same ethnic minority. I realise this is anecdotal but a friend from an ethnic minority told me that it's a recurring pattern which he experiences a lot in EA.”

“In the Bay Area, I'm always a bit uncomfortable with how white the spaces / gatherings are. Even though I'm white, I imagine if I were to introduce friends who were people of color to the EA events, I would feel awkward because I imagine they might feel uncomfortable (and judgmental) given how white the events are.”

“...One minor thing of relevance to ethnic diversity is that at times I felt people who grew up in Western countries were quite ignorant when talking about the developing world (for lack of a better word), i.e. people talking about extreme poverty – with the best of intentions – but just using ignorant language or just not really knowing much about how things really are on the ground. …”

There are conflicting thoughts on whether EA is unusually unrepresentative for its reference class or not. It seems it depends on what the reference class the person is using. A number of people mentioned there are a lot of Asians involved in EA, possibly comparable to the total population of Asians in the US. But compared to the fields EA usually recruits from where there are more Asians than in the US overall, Asians may be underrepresented.

“I think there is unfortunately little [diversity], but that mostly reflects broader social trends (which should change,) rather than attitudes within the community (which I find to be fairly pro-diversity.)”

“My experience of EA is that it is full of brilliant individuals but also extremely insular and surprisingly ignorant of relevant initiatives, communities, and debates that are adjacent to EA's interests but don't use the keywords "effective altruism." For both general reasons and in my specific role as [organiser of an] interest group within EA, I have a strong motivation to bridge these gaps and increase the profile of EA ideas and thinking styles within more mainstream practice areas across the social sector. I currently find that very difficult to do, however, with EA's ongoing lack of diversity with respect to visible minorities. I engage with roughly a dozen analogous professional communities each year (going to conferences, following listservs, following people on Twitter, etc.) and EA is a stark outlier in terms of representation, a fact that is impossible to avoid as soon as one goes to one of the EA Global events or takes time to peruse the headshots on any movement-affiliated website. In turn, that creates a structural pipeline challenge that makes it more difficult for me, as [EA interest group] community co-leader, to recruit interested and engaged volunteers who are bought in to EA values and will represent our sub-movement well. More importantly, it leaves the broader EA community extremely vulnerable to critique and dismissal from the outside, which sharply hamstrings the potential that its ideas have to change the world. This is all the more true as US and international philanthropy have moved sharply in recent years to align with social justice movement norms and values.”

One comment expressed a criticism, but highlighted that EA leaders seem to care about this issue.

“...I also unfortunately get the feeling that lots of people in EA don't think issues of diversity are important (whether racial, gender, sexual/gender identity or others) e.g. there were 'anti-pc' and 'anti-woke' interest tags on the EAG 2019 app. This makes me personally feel uncomfortable bringing up such issues from a gender perspective and I don't want to project but could imagine the same being true for people of colour in terms of how comfortable they feel discussing these issues in a white-dominated space. I feel the 'powers that be' really care about this eg noticed a concerted effort to get a more diverse range of speakers at EAG 2019, to have diversity-related meetups and to raise the profile of ethnic minorities, but I worry that the same is not true of the 'average' person in EA. …”

Criticisms of diversity attempts

Two people shared comments criticizing diversity attempts by CEA and in the animal advocacy space.

CEA had soft speaker race/gender quotas for an event I planned; I'm seeing affinity-specific EA groups starting up online” (Context: for recommendations, this person suggested “Maybe CEA shouldn't be endorsing the creation of race/gender specific groups within EA.”)

I’ve invited Sky Mayhew to comment on CEA’s policies on this topic in the comments section.

“I have noticed an over emphasis on diversity in animal advocacy possibly to the actual detriment of orgs working in that area. The same occurs with some student groups where minority students are kinda seen as more valuable members when that really should not matter.” / The end-line impact is what matters and diversity and inclusion are an instrumental goal towards that, not intrinsically good.

Q6. Recommendations

The breakdown of recommendations is shown in the table below. There were a total of 49 respondents who gave suggestions, with some respondents giving multiple. 9 respondents stated they didn’t know what to recommend.

I have also added quotes from the “additional comments” section if they were recommendations below. The table above is not adjusted to include those responses.

Improve global community building efforts and increase outreach (5)

“My current thought is that the (comparative) lack of effort to increase outreach globally (e.g. creating new global EA hubs) has more to do with core EA being overwhelmingly British and American than we realize. Arguments around EA potentially not translating well to other cultures feels a bit misinformed (or rather it seems difficult to be confident in that opinion given the lack of diversity in EA). Given capacity/funding, exploring the development of other global EA mini-hubs to facilitate full time EA work for people unable/unwilling to move to the UK/Bay seems like a good idea for EA in the long run.”

"Include non-Western philosophy, ethics schools, social science and academic research in the mix when making conclusions for EA as a whole. This is perhaps best done by including academics, philosophers, cultural ambassadors, social scientists who are specialised in non-Western schools of thought and have worked primarily in non-Western institutions/locations."

"I was disappointed to see that post on the EA Forum about how little CEA has done to support EA leaders in other countries who want to talk about customizing EA messaging for their culture and stuff like that. ... [quote on Motivation from the post] ... There's a lot we could be doing for diversity, but let's start by making sure we are doing the basics well first. Before proactively recruiting POC who could be a good fit for the EA movement, let's focus on making sure we are already serving the POC group leaders who are coming looking for advice and support. Also, working with people in other countries has a lot of potential to improve EA's impact in an international context (world peace type stuff), and in my opinion that is probably the biggest selling point of ethnic diversity.”

It could well be that after all this effort, our ethical conclusions are 100% the same as before (I would be surprised if that were the case). But even if so, at least this would be an informed conclusion. And our confidence in EA recommendations would be that much stronger. Sadly I am not academically trained so may not be able to articulate this the best. But I can think of several examples where this might change ethical conclusions, e.g. subjective quality of moment of death."

Create spaces for discussions (5)

Regular meetups/ethnic diversity group for those interested”

Talk more about systemic oppression e.g. racism, decolonising from an EA perspective starting with EA itself as a community. Give more space to people from other ethnic diversity (talks at EA global for example or in EA organisations)”

“I've seen affinity groups at EA conferences (eg people-of-colour in EA) which seem like a good idea.” (From Additional Comments)

Communicate EA ideas in a more accessible way (4)

“Change of message away from privately educated coding, improved outreach and making efforts to expand into more diverse places. It seems like a systemic selection issue that then leads to a positive feedback cycle.”

“To make the language more accessible, use plain English and avoid academic economics terms where possible (ie. counterfactuals, priors, etc).”

“I care about increasing access more than changing the way we work with people who find us. More EA materials should be translated into other languages, and we should be more clear about the wide variety of paths in life that are compatible with being a part of the movement (you don't have to study topic X, you don't have to subscribe to philosophical view Y -- if you live an ordinary life and give to good causes, you are one of us).”

“Re: ""non-native English speakers"": Native speakers should try to speak a bit more slowly and clearly, whenever possible. Subtitles for EAG videos (could be a volunteer project). Reminders on the EA Forum that for many readers English isn't their first language. Easy, informal and low-hanging ways to ask for help with language problems (""no dump [sic, dumb] questions""), without having to post in a large Facebook group. It would be great to have (more) EAG-facilitated meetups of people from a particular language region too.”

I've invited Sky from CEA staff to comment on subtitling in EAG videos.

Strong commitment to representation (4)

Quotas (2)

“I would like a commitment from the community that for every X people of white background speaking then X people from diverse backgrounds should also be invited and welcomed to speak. I would also recommend identifying Community Champions to lead the way in this area.”

Greater representation in EA leadership. (2)

“More diversity in figureheads - hosts of major podcasts, people writing books about EA, keynote speakers at conferences. That's not an easy change to make quickly, as it takes a combination of experience, personality, speaking skills, cred for doing something besides being a figurehead, etc. But something we should work towards over time”

“I don't think it's necessary to point out that among EA organizations, a look at the team page often shows a pretty homogenous picture in terms of ethnicity/race (and other things). There is some awareness around this already, so I won't write more here, though I will point out that it seems to me that the ""CEA leadership search committee"" (I'm referring to the email sent to CEA supporters a week ago [October 2019]) was, as far as I know, 100% white. Sure, you can't always tick all the boxes but with such an important position and such an important topic that (ethnic) diversity is, which EA hasn't exactly looked great on historically, I felt this was a missed opportunity.”

Targeted Outreach (3)

“Keep spreading EA things, esp. to non-elite places?”

“Do some targeted recruitment focused on STEM programs for people of color: orgs like Black Girls Code, the GEDI program of the American Evaluation Association, etc. It'll have to be handled carefully and would be best coming from people who already have a foot in both EA and the target community. But I think people in STEM programs are a natural audience for EA ideas.”

Online Approaches (2)

“Moderate FB groups a bit more aggressively. I understand there’s a balance here, but it feels like it’s often just one or two people who are loudly creating an environment that’s not super welcoming for others.”

“...Why I think online should be prioritized over offline for now: Online content and discussion can get read for years whereas, say, a single conference is over at the end of the weekend (sure, sometimes there are videos but they don't get many views). Also, EA is a global community that connects online, so I think it's so much the more important that the architecture online is as good as can be. Finally, I find the problem of representation to be particularly "in your face" online: If you, for instance, look through the most frequent (and thus visible) commenters on the EA Forum, I'd say that it's clear that the vast majority are men, the vast majority of whom are white... Re: Ethnic diversity online: Really difficult and I don't think there are easy solutions. But I think this issue should be prioritized when it comes to community health/community building, which I feel has not been the case so far.”

Improving awareness and training of social justice (2)

“I wish more people had training in / awareness of some social justice basics. And that people were more willing to discuss topics that might matter a lot to someone with lived experience of discrimination / racism / classism / etc. in the U.S., instead of only talking about, e.g., the effects of income inequality between people in "developed" countries and "developing" countries, as though all suffering is geographically far from us.”

“... 2. Invest in racial equity training of some kind for CEA staff and perhaps selected EA community organization leaders. I've done these kinds of training sessions twice as a participant and have co-facilitated others. They are often presented as the monolithic, not-to-be-questioned ""truth"" and I think it would be important not to frame it that way for this community. Instead, I see it as valuable for EA leaders to understand the arguments that other social sector leaders are being exposed to and increasingly accepting, so that they can understand the broader environment in which they are operating.
3. Consider a racial equity track or focused discussions at EA Global conferences. The W. K. Kellogg Foundation has funded similar themes at other social sector convenings and they seem to have had an impact on shaping the discussion. Note: it is *crucial* IMO that any such discussions involve white people, as they are mostly the ones within this community that need to buy in to change. I do understand the desire to have closed affinity groups for POC; what I'm talking about is in addition to that.
Part of the reason why I hesitate on these last two in particular is that EA is a naturally skeptical and inquisitive community, and I wouldn't want those qualities to get swallowed up or repressed by church-of-social-justice dogma, or worse, promote a backlash that tears the movement apart. My ideal version of EA is that it is a community that can credibly push back on, e.g., social justice norms that harm values of truth-seeking in ways that matter for the world. But in order to have that credibility, EA must first show that it is not primarily a constituency of overeducated ambitious young straight white men from English-speaking countries. So that's why I am focused on representation first, introspection second.”

Reflect (1)

“I think to ensure the principles of effective altruism are being applied, rather than people drawing on associations. I would certainly look more closely at things like moral uncertainty and how we consider including people to cover the scope of issues or excluding people in order to make decision making "easier", or perhaps to avoid discomfort.”

Things not to do (4)

“Maybe CEA shouldn't be endorsing the creation of race/gender specific groups within EA.”

"I need to acknowledge that I come from an ethnically diverse country, which almost doesn't care about diversity. So my views might be influenced by that.
A. Honestly, I think [sic, feel] uneasy about diversity in EA. It seems to me that focusing on the ethnic diversity of the community (especially in the US/UK) is not as important as, say, building a stronger presence in Europe.
B. Also, I am unsure whether highly capable people of color care as much about ethnic similarity as similarity of world views/intelligence and other traits. If EAs would onboard/engage/headhunt member of minority from, say, a big 3 consulting, ugh, I have hard time believing that we would be less ethnically diverse than his or her previous workplace."

"No. I'm not sure there is any simple policy change that could fix this issue, I think this is just EA unfortunately reflecting broader trends in society. I'd fear that a heavy handed policy change might be harmful or impede effectiveness."

“Many diversity initiatives are not effective at accomplishing their goals, so I would like to see whatever is tried have (1) clear goals, and (2) measurable outcomes.

Additional Comments

These are the additional comments weren’t related to the previous questions:

“I believe that there is also a real lack of diversity and inclusion in other areas too e.g. gender and sexuality.”

“This is an important topic and i am pleased more consideration is being given. However, i would say that people raising issues are perhaps taking something of a risk in relation to their position in the movement.”

“I think there is a danger of conflating ethnic diversity and things correlated with it, e.g. diversity of geography/experience/language/class that would enable EA to serve developing countries better and have a better grasp of the real world in general.”

Additional Data

As a result of the EAG London meetup, we created a closed Facebook group for non-white EA community members to share experiences with each other, similar to the Woman and Non-Binary Facebook group that already exists. The group now has ~100 members and runs monthly calls. Upon creation of the group, I reached out to about 12 new members to ask them why they joined the group, to make sure the group was in line with what community members wanted. Many of them stated that they hadn’t had any negative experiences themselves, but wanted to see what was happening in this space. It’s unclear how many of those 12 also filled out the survey.

Discussion

This is not a comprehensive discussion, but rather a series of ideas and thoughts that might help clarify discussion around this topic in the future.

Some Open Questions

  • What are the correct reference classes for EA?
  • What are the ideal reference classes for EA?
  • Communication seems to be particularly promising - what results would A/B testing different ways of explaining EA ideas and concepts produce?
  • Seeing how many current people were interested in EA via personal contact vs the internet, and if there are patterns with how minorities are attracted to EA
  • How do we have conversations about diversity in general and account for the interactions of ethnic and racial diversity with gender and class in particular?

Possible Themes to Explore

The goal of these themes is to break down diversity issues into useful categories to help think about diversity in different contexts. I have developed these from the survey data and observations of the EA community. These categories are early stage and likely to change, but might make diversity discussions more precise.

Where the problems occur

The table below shows the places where a lack of diversity could impact the EA movement, drawn from the survey responses and broader discourse.

Types of Ethnic Diversity

There appear to be three main types of ethnic diversity (including ethnicity, race, religion and local community) which would be addressed in different ways. It might make sense to separate conversations around these efforts, as the solutions required for these different types of diversity could differ .

Appendix

Summary of All Question Responses

Long Responses

Please see the full report for summaries of four longer Responses.


Relevant Community Spaces

Diversity & Inclusion in Effective Altruism Facebook Group (open to all, to discuss any diversity and inclusion-related topis)

Ethnic Diversity in Effective Altruism Facebook Group (closed, for those who identify as a non-white race or ethnicity)

Comments9
Sorted by Click to highlight new comments since:

Thanks for the post!

Just a comment on the reference to the EA Survey numbers. As I discussed here, because the EA Survey's question about race/ethnicity was multi-select, the percentages of respondents selecting each category can't be straightforwardly converted into percentages "identif[ying] with non-white race or ethnicity." We used multi-select to allow people to indicate complex plural identities, without forcing people to select more fixed categories, but it doesn't allow for particularly simple bench-marking if you want a binary white/non-white distinction. In the next survey we'll consider adding a further question with more fixed options. It's more accurate to describe our data as showing that 13.1% of respondents did not indicate white identity at all, 80.5% exclusively selected white and a further 6.4% selected both white and other identities. Unfortunately, interpreting this last category in terms of an interest in a white/non-white binary is fraught, since it's unclear whether these individuals would identify as "mixed race", white, non-white or a "person of colour." Of note, despite Asian being the most common identity other than white selected for this question, the most common selection within this 'mixed' category, was White and Hispanic (and the relationship between Hispanic identity and ethnicity/race is not straightforward.

As such, in a more expensive sense, the total "non-white" percentage may be higher, up to around 20%.

Regarding the broader claim that: "The goal of the survey was to solicit the experiences of people from ethnic minorities in EA. The survey seems to have achieved this goal compared to the annual EA survey, a much higher proportion of respondents to this survey were non-white."

I agree the percentages of non-white/white respondents is a bit higher in the dedicated "Ethnic Diversity" survey, but you had around 10x fewer ethnically diverse respondents expressing views overall, so this is not a clear win. The percentage difference could be explained entirely by white EAs thinking "This survey isn't really for me." A survey specifically about ethnic diversity seems particularly likely to skew towards respondents (both white and non-white) with a particular interest in the topic too, which is probably of particular significance when we're dealing with only around 30-36 respondents. That said, I agree this is an important source of more qualitative data than we could gather with the EA Survey!

I would love for there to be an analysis of how demographically diverse core EAs are (high self-reported engagement and/or EA Forum membership).

(I also wrote this here.)

Thanks Jonas! We'll be discussing this in more detail in our forthcoming post on EA Engagement levels.

Hi David, thanks for the detailed information - I will make an edit to the text to clarify this point and adjust the wording accordingly so it's not misleading, upon looking over that phrasing I don't think I'd use that phrasing even if the percentages were the same.

Regarding your second point, I agree it's plausible the those who care more about the topic are likely to respond, I mention this in my limitations section and also mention my uncertainties about how much this skews the data.

This survey was not meant to be comprehensive or fully representative of the population in question (I don't think I make this claim anywhere in the survey, if I do please let me know where) but instead provide some qualitative data to improve the quality of diversity and inclusion discussions and hopefully. As I mention at the beginning, the initial goal was to share this with CEA to give community input on diversity issues, but I'll edit the text in the write-up to be more consistent with this.

I don't think this survey is the most valuable piece of research to be done on this topic, but I asked questions that are not asked on the EA survey, which to my knowledge have not been asked in an EA survey before. So I believe that it is still valuable to get some data, both on experiences of non-white respondents but also general attitudes and beliefs on this topic, rather than relying purely on anecdotal evidence while having diversity discussions.

Finally, I hope this survey might help further efforts in this space be a little more structured or productive, which is why I suggested "Possible Themes To Explore".

Thanks Vaidehi. I agree that this is still useful information, I was simply responding to your direct comparison to the EA Survey ("The survey seems to have achieved this goal [solicit the experiences of people from ethnic minorities in EA] compared to the annual EA survey, a much higher proportion of respondents to this survey were non-white.").

By the way if you have specific questions that you would like us to include in the EA Survey please let us know (though no hurry).

“… [on IDInsight’s new GiveWell report on value of life in Africa] “More to the point, it draws attention the fact that a group of a dozen or so generalist researchers with limited adult life experience in one of the wealthiest parts of the world are LITERALLY ASSIGNING DOLLAR VALUES TO THE LIVES OF POOR PEOPLE IN AFRICA. And while I understand that is an unavoidable value judgment that we all implicitly make anyway, it nevertheless highlights the highly problematic nature of not making more of an attempt to incorporate the perspectives of those affected in these calculations, and how much reputation risk (=impact risk) EA is exposing itself to if these underlying issues are not addressed”

To be clear, unless this was referring to something else, this was a study where IDInsight researchers (including professional developmental economists) went to Ghana and Kenya and set up experimental designs where they asked locals how much they value additional years of life versus cash transfers, among other questions. Saying that they should have "ma[de] more of an attempt to incorporate the perspectives of those affected in these calculations," without explaining how, seems somewhat unfair to me.

EDIT(2020/08/05): Possible conflict of interest: I'm friends with and had some volunteering engagements with the lead author.

Thanks for doing this work!

I've thought of the "Improving awareness and training of social justice" point a bit in the past when thinking about gender diversity and find it difficult. I am a bit worried that it is extremely hard or impossible without everyone investing a substantial amount of time:

My impression is that a lot of (ethnic/gender/...) diversity questions has no easy fixes that some people can think about and implement, but would rather benefit a lot from every single person trying to educate themselves more to increase their own awareness, esp. community builders, and high profile people that get a lot of attention. One example that I think is hard to improve otherwise: I noticed in Toby Ord's The Precipice the following sections:

"Indeed, when I think of the unbroken chain of generations leading to our time and of everything they have built for us, I am humbled. I am overwhelmed with gratitude"

I know this doesn't detract from his overall point of what the generations of the last hundreds of thousand years have done "for us", but I can't help to wonder how reading this must feel like for some people that primarily associate history with their ancestors being fucked over by colonialism or being enslaved, and them still paying the price for this. The Precipice actually mentions this later but is clearly written from the perspective of the people descending from those inflicting injustice, and not receiving it:

"Consider that some of the greatest injustices have been inflicted [...] by groups upon groups: Systematic persecution, stolen lands, genocides. We may have duties to properly acknowledge and memorialise these wrong; to confront the acts of our past. And there may yet be ways for the beneficiaries of these acts to partly remedy them or atone for them."

While I'm a POC, I'm certainly not from an ethnicity that suffered the most from the historic (and ongoing) actions of the elites in primarily white countries. But I can imagine that many people whose families are or have been on the receiving end of the injustice might be alienated by this section, which reads a bit like it's a given that readers are on the other side of the coin.

I don't want to slander the book or the person, I very much enjoy the book, and don't assume any negative intentions (and think the section can also be read more charitably, but I think it's important that it can be read in an alienating way by POC). I just think this is a good example of the problem and that it is really hard to be aware of such things when you didn't spend substantial time to understand underrepresented groups - that you are not part of - better.

While, in an ideal world, I would like everyone in the world to do so, it is a big time sink and feel reluctant about recommending everyone to invest this time, especially when the opportunity cost are so high. I'm not sure how to remedy this; whether investing the time is clearly worth it; there are better ways to make progress that are less time intensive; or whether we should only aim for low hanging fruit; or something entirely else. I would be very curious to hear other people's thoughts, I'd gladly notice that I'm totally off the mark and I worry more than warranted :)

(I also feel weary of openly saying that investing into understanding underrepresented groups might not be worth the time, as I just did, because I think it can be very hurtful and dehumanizing.)

This was very interesting for me to read! I would also be very curious to learn if some groups have found successful practical ways to work with improving diversity/making everyone feel welcome and comfortable.

Very interesting to see this data- thanks so much for collecting it and writing it up! I hope future versions of the EA Survey will adopt some of your questions, to get a broader perspective.

Curated and popular this week
Relevant opportunities