{"items": [{"author": "Vivian", "source_link": "https://www.facebook.com/jefftk/posts/690624440962?comment_id=690626107622", "anchor": "fb-690626107622", "service": "fb", "text": "I think odinsdream has a more modest goal than you do.  I think odinsdream is looking for charities that do more good than harm, rather than charities that do more good than all other charities.  (Although the first goal is still more difficult than it seems.)", "timestamp": "1412452761"}, {"author": "Aleksandra", "source_link": "https://www.facebook.com/jefftk/posts/690624440962?comment_id=690734565272", "anchor": "fb-690734565272", "service": "fb", "text": "Jeff, I interpreted the statement very differently. I didn't think that the person was saying that the problem isn't difficult. Rather, I thought that they were expressing their dissatisfaction with the final results that Givewell produces. I've heard other people say things like this: \"Give me the data that Givewell analyzes, and I'm pretty sure I could produce comparable results, without the budget on which they operate.\" How would you respond to a statement like that?", "timestamp": "1412522560"}, {"author": "Tom", "source_link": "https://www.facebook.com/jefftk/posts/690624440962?comment_id=690743282802", "anchor": "fb-690743282802", "service": "fb", "text": "Indeed - one piece of evidence for the difficulty of this is that GiveWell's work has got substantially more valuable as they learnt more about evaluation, and build up a fairly sizeable team of full-time employees working on this.", "timestamp": "1412527150"}, {"author": "Julia", "source_link": "https://www.facebook.com/jefftk/posts/690624440962?comment_id=690871011832", "anchor": "fb-690871011832", "service": "fb", "text": "Aleksandra Taranov, re: \"Give me the data that Givewell analyzes, and I'm pretty sure I could produce comparable results, without the budget on which they operate.\"<br><br>- You can't *get* the data that GiveWell analyzes as a random person. I used to work in donor services at a large charity, and I can tell you most charities don't even collect the data. If you're a small donor, most charities won't give you the data they do have, because they haven't made a nice presentation out of it. If you're a large donor, they might find it worthwhile to package the data nicely if you ask for it. But if you're a random person combing through hundreds of charities, you will get no data from most of them beyond what's on their websites. (Not that charities are unreasonable to do this; it would take a lot of staff time to answer every request from every random person.) The data on their websites is basically limited to their 990 tax form, which doesn't give any useful information about what they are accomplishing with the money. It's more worthwhile for charities to collect and share data with an organizing that will recommend them to others than it is for random individuals. <br><br>- It takes time and skill to go through data if you're even able to get it. Most people don't have either in the quantities that GiveWell does, because they hire especially skilled people to work at this full time.<br><br>- GiveWell has a track record of noticing things that other people didn't (e.g. that the deworming figures were off by a factor of 10).<br><br>- GiveWell is solving a coordination problem. A bunch of random people deciding to investigate charities will largely duplicate each other's work rather than dividing up the research to get a wide spread. <br><br>- Their operating budget comes to about $100 per year per person donating through their website. (I imagine a lot of others are swayed by their findings but don't donate on their actual website or otherwise report \"I am a GiveWell donor.\") That's not counting the foundations that use GiveWell findings to make major donation decisions. We could quibble about whether $100 per donor is a bit too high or too low a budget, but I think it's in the right neighborhood.", "timestamp": "1412591892"}]}