Open Thread 4

I can think of many things I no longer do but I'm not sure that's a direct result of my EA involvement. I'm a busy person, so activities that offered a small benefit naturally gave way either to more productive things, or things that offered a bigger benefit.

I think I probably drink (and therefore spend less on) alcohol. I only watch a few selected TV shows and don't re-watch old episodes. I spend less time at the gym but more outdoors cycling with my boyfriend (kills multiple birds with one stone).

I think as we age our priorities naturally shift and our activities naturally change. Nothing in my personal life has changed ONLY as a result of EA ideas.

Since 2000 I've abandoned TV, videogames, celebrity gossip, musical ability, knowledge about bands, politics, theater classes, dancing classes, handball, tennis, reading fiction, reading parts of Facebook, maintaining contact with groups X and Y of friends, newspapers, magazines and comics.Those were not easy choices, each comes with a cost, a sadness, and a feeling that something valuable has been lost. The richness of flavors of life got somewhat poorer.

This is a bit tangential but I don't know if there's a single EA that smokes cigarettes.

What about one of Bostrom's books - either Global Catastropic Risks or Superintelligence? Both are excellent.

Here are some reasons to think AMF could still be worth giving to:

AMF net distributions come in large (and varied) sizes, and they generally only arrange distributions after they have the money to cover them. This means that a larger cash reserve may open up extra opportunities for distributions that they wouldn't have with a smaller cash reserve; it's therefore not lossless to just wait until they run down the cash reserves before giving. Of course the opportunity value created by a marginal dollar in their cash reserves may well be lower now than when the cash reserves were lower, but it's not obvious how big this effect is (it depends on the distribution of prices of possible net distribution opportunities).

Also note that incentives as a private donor differ from the incentives GiveWell has in making recommendations (at least as far as risk-aversion goes). GiveWell is in the business not just of making good recommendations, but of building a reputation for making good recommendations. A small chance that money going to AMF now will be wasted should act as a small discounting factor on the expected value for a private donor -- but rears much larger than this for GiveWell, for whom the bad outcome translates not just to the waste of donations, but a reduction in trust from donors who say with hindsight that they should have known better.

I don't personally donate to AMF (I think top animal charities are more effective), but I've heard a few people who still do. They are confident that AMF will successfully mobilize its current funds within the next couple of years or so.

As of December 1, 2014, GiveWell has reinstated AMF as a top charity. GiveWell believes that AMF now has room for more funding. See http://blog.givewell.org/2014/12/01/our-updated-top-charities/

This sounds like a great idea. I'd love to participate. Sounds like it could be a great way of creating content and discussion that would be available to refer to in future.

What topic would you suggest?

Per Bernadette, getting good data from these sorts of project requires significant expertise (if your university is as bad as mine, you can get student media attention for attention-grabbing but methodologically suspect survey data, but I doubt you would get much more). I'm reluctant to offer advice beyond 'find an expert'. But I will add a collection of problems that surveys run by amateurs fall into as pitfalls to avoid, and further to provide further evidence why expertise is imperative.

1: Plan more, trial less

A lot of emphasis in EA is on trialling things instead of spending a lot of time planning them: lean startups, no plan survives first contact, VoI etc. But lean trial design hasn't taken off in the way lean start-ups have. Your data can be poisoned to the point of being useless in innumerable ways, and (usually) little can be done about this post-hoc: many problems revealed in analysis could only have been fixed in original design.

1a: Especially plan analysis

Gathering data and then analysing it always suspect: one can wonder whether the investigators have massaged the analysis to satisfy their own preconceptions or prejudices. The usual means to avoiding it is specifying the analysis you will perform: the analysis might be ill-conceived, but at least it won't be data-dredging. It is hard to plan in advance what sort of hypotheses the data would inspire you to inspect, so seek expert help.

2: Care about sampling

With 'true' random sampling, the errors in your estimates fall as your sample size increases. The problem with bias/directional error is that its magnitude doesn't change with your sample size.

Perfect probabilistic sampling is probably a platonic ideal - especially with voluntary surveys, the factors that make someone take the survey will probably change the sample from the population of interest along axis that aren't perfectly orthogonal to your responses. It remains an ideal worth striving for: significant sampling bias makes your results all-but-uninterpretable (modulo very advanced ML techniques, and not always even then). It is worth thinking long and hard about the population you are actually interested, the sampling frame you will use to try and capture them, etc. etc.

1. Questions can be surprisingly hard to ask right

Even with a perfect sample, they still might not provide good data depending on the questions you use. There are a few subtle pitfalls besides the more obvious ones of forgetting to include the questions you wanted to ask or lapses of wording: allowing people to select multiple options of an item then wondering how to aggregate it, having a 'choose one' item with too many selections for the average person to read, or sub dividing it inappropriately: ("Is your favourite food Spaghetti, Tortollini, Tagliatelle, Fusili, or Pizza?")

Again, people who spend a living designing surveys try and do things to limit these problems - item pools, pilots where they look at different questions and see which yield the most data, etc. etc.

3a. Too many columns in the database

There's a habit towards a 'kitchen sink' approach of asking questions - if in doubt, add it in, as it can only give good data, right? The problem is that false positives become increasingly difficult if you just fish for interesting correlations, as the possible comparisons increase geometrically. There are ways of overcoming this (dimension reduction, family-wise or false-discovery error control), but they aren't straightforward.

There are probably many more I've forgotten. But tl;dr: it is tricky to do this right!

I think a key challenge with this is how you intend to select your sample, so as to be truly representative. Getting interested students will select a certain type of participant; so will offering a payment. Could you get the University on board with distributing your survey via email to random student numbers, for example? Your results will only be powerful (and useful) if you can ensure random selection of participants.

Sounds like an interesting project Jorgen! It sounds like you already have a good plan, so my main survey-running tip would be to keep it short, and break it out into multiple pages if it reaches sufficient length so people don't have to complete the whole thing. We used LimeSurvey for the EA survey, which is a pretty nice piece of software - I'd be happy to answer any questions on that if you want to message me.

For overviews, I recommend: Preventing Human Extinction by Beckstead, Wage and Singer Existential Risk as a Global Priority by Bostrom Reducing the Risk of Human Extinction by Matheny GCR Survey by Sandberg and Bostrom

For book-length detail, Global Catastrophic Risks by Bostrom and Cikovic is good. Others, that I haven't read, are Our Final Hour by Martin Rees and Catastrophe: Risk and Response by Richard Posner

I haven't read it myself, but I believe that the book 'Global Catastrophic Risks' (edited by y Nick Bostrom, Milan Cirkovic, and Martin Rees) covers a broad range. Here are links to it (5% of Amazon purchases through them go to SCI): US; UK.

You can also read The Open Philanthropy Project's (previously GiveWell Labs) notes on the x-risks they've investigated.

Don't forget to check out "What Small Things Can an EA Do?"

Summary:

• Use GiveWell to inform your donation choices.

• Join Amazon Smile.

• Join some mailing lists.

• Set up an altruistic tip jar.

• Talk to friends.

• Run a Giving Game.

• Run a Fundraiser.

• Join or Start a Local Meetup Group.

• Try Vegetarianism (or Veganism!) for a Week

• Volunteer.

• Live more frugally.

• Donate more.

I told someone that Heiffer International was an ineffective charity, showing her the GiveWell review on the page. She ended up never donating to anything international ever again, preferring now to donate to the local opera.

Seeing as Heiffer probably beats the local opera, this was a net loss.

Moral: be careful about framing, who you approach, and how.

In my opinion, not very large. In my experience, I think telling people about EA has had high counterfactual value, but further work to sustain their commitment hasn't had much counterfactual value -- they either don't get engaged despite my work, or they stay engaged despite my being hands off.

To tease the results of the 2014 Survey of EAs, local groups were one of the least popular answers to the question "Which factors were important in 'getting you into' Effective Altruism, or altering your actions in its direction?". Does anyone know of other data on this?

Now I ask you to consider whether benefit (1) would in fact be the case for important elections (elections say where the elected will govern over 10 000 000 people). If 100 Worlds had an election decided based on one vote, which percentage of those would be surreptitiously biased by someone who could tamper with the voting? How many would request a recount? How many would ask it's citizens to vote again? Would deem the election illegitimate? Etc... Maybe some of these worlds would indeed accept the original counting, or do a fair recounting that would reach the exact same number, I find it unlikely this would be more than 80 of these 100 worlds, and would not be surprised if it was 30 or less.

It does indeed look hard to predict what will happen here exactly. Luckily it pretty much factors out of the analysis.

I'll demonstrate with a toy example. Say the election goes to whoever gets more votes, EXCEPT if the counts only differ by 1 vote. In that case there's a 100% chance that the election is deemed illegitimate and the military step in. Say you're voting for Party B rather than Party A. You know Party A has 1,000 votes, and you think Party B has about 1,000 votes.

If you move Party B from 1,000 to 1,001 votes you make no difference -- you get the election declared illegitimate either way. Instead your impact comes from the cases where Party B was getting 998 votes already (you move from Party A win to military takeover) or Party B was getting 1,001 votes already (you move from military takeover to Party B win).

If you're uncertain enough about the outcome that all of those possibilities for vote numbers look about equally likely to you, then the net expected effect of your vote is a small chance (the chance that the votes stand at any one particular number) to move from a Party A win to a Party B win -- just the same as if you disregarded all the possible complications.

My main response is that this is worrying about very little -- it doesn't take much time to choose who to vote for once or twice every few years.

But in particular,

2) The risk you incur in going to the place where you vote (a non-trivial likelihood of dying due to unusual traffic that day).

is an overstated concern at least for the US (relative risk around 1.2 of dying on the road on election day compared to non-election days) and Australia (relative risk around 1.03 +/- error analysis I haven't done).

Diego, I don't weight any of the 4 risks you've listed very heavily. I also think you've underestimated the benefits.

In regards to Benefit #1, a vote's relevance doesn't depend on the election being decided by a single vote. If you think probabilistically, then in any given election, your vote has a certain probability of affecting the outcome. You can weight that against how important you think it is for Party A to win over Party B. I think that given how little it costs to vote, it's usually clearly worth it to take a small action with a tiny probability of having large-scale consequences.

I think this is somewhat analogous to going vegetarian, in which case you're contributing to a larger cause even though your individual decision not to buy meat only has a tiny probability of being the non-purchase that causes the grocery store to order one less item next time.

Other benefits:

a) Your vote might cause other people to vote with you. In this case, you are no longer a single vote but a package of votes.

b) There's also something to be said for signalling an interest in politics and social issues.

c) In some elections, your vote might give the party you voted for more seats, funding, power and/or legitimacy, even if they ultimately lose the election.

d) The attention it takes to learn about politics can also have multiple benefits: being in touch with the people around you, learning about issues in society, learning about solving those issues, etc.

It seems that you missed a substantial cost: the time taken to physically go to vote (and queue if necessary). I'd expect this to be a bigger expected cost than the extra risk incurred.

I think voting is very valuable from a "moral trade" perspective. I want to convince other people to take my ideas of virtue seriously, but they won't if they see me doing something that's commonsense unvirtuous like not voting.

I can write [crosspost] for anything that I cross-post.

I don't know if this is an "I believe X and no-one also does." But I have noticed I have a tendency to poke everything in a counter-intuitive manner, So even something dumb like gamer gate - my instinct is to ask "well is there any kernel of truth in these criticisms?" rather than advocating on behalf of a cause. (annoyed EA types would see the same about my poking of X-risk cause)

I worry that this makes me quite "uncooperative" and reduces my effectiveness because this is my instinct even with super clear causes like climate change. But I am not sure how to resolve it.

I just don't find advocacy as interesting as argument/ debate - even though I can't say it works with my terminal values.

Working on getting a more useful skill but for now if ever someone needs some audio editing, perhaps for a potential EA podcast, I can do it.

Also this job board seems relevant as skills people have they might not think would be of use are in demand.

I do data science, which makes me a good fit for working on the forthcoming EA survey!