BenjaminTodd

That's a good point. We do that elsewhere in the career guide, but I might add it to the intro text.

https://80000hours.org/articles/the-meaning-of-making-a-diff...

You can alter each of your questions responses individually to see what effect it's having on the results.

We found these questions were the most useful filters when doing one-on-one advice with people, but I can see they'll probably have to change as our age range expands.

We do quite a few interviews within the profiles. Is that what you have in mind?

We're focused on social impact, which means enabling other people to live more flourishing lives. https://80000hours.org/articles/the-meaning-of-making-a-diff...

That's why startup and software jobs reasonable scores for direct impact. https://80000hours.org/career-guide/top-careers/profiles/sof... https://80000hours.org/career-guide/top-careers/profiles/tec...

There's some more detail in the profile: https://80000hours.org/career-guide/top-careers/profiles/thi...

That's really interesting. Originally we just had a list, and I had this concern with the test. However, as the list of paths we've reviewed gets longer, it seems useful to provide a quick way of filtering it.

It's true that it depends a lot on what you do within the career.

On stats behind earning to give, of the people we know, 0% have quit, meaning carrying on in a high-earning career but stopped donating at least 10%. Most are still earning to give, and the others have left high-earning careers though to make an impact through research, nonprofits etc instead. I think this is due to the strength of the community - if you were earning to give alone it would be much harder. Also, we only have 3 years of data, so it's too early to get a good measurement.

That means they're all tied.

What responses did you put in?

Ok, thanks for taking the time to read through.

I agree another big part of the equation is what causes to focus on. We have separate content on that but it needs a lot of work:

https://80000hours.org/articles/cause-selection/

If you want to focus on education as a cause, then I agree working as a teacher at some point makes a lot of sense.

I think it'll be hard to work out what works just from that though, because the outcomes of education take many years and are often counterintuitive. You'd need to eventually do randomised controlled trials.

I've written some more here: https://80000hours.org/2015/06/5-reasons-not-to-go-into-educ...

I'd also recommend checking out sources like this: https://educationendowmentfoundation.org.uk/about/what-works...

I'd suggest considering going back to tech/startups, but trying to join one that's doing positive innovation in an important cause, like edtech or online education.

You could use a think tank or econ phd position to research policies with potential for disruptive change. We also recommend startups highly. What paths do you think we should look into more?

We do consider teaching, and it could be a good path if you're a very good fit, but we don't in general recommend it if you want to maximise your social impact. https://80000hours.org/career-guide/top-careers/profiles/tea...

Thanks, that's useful feedback.

Click through to the individual profiles to see the reasoning.

Fair. Bear in mind: (i) it's aimed at talented graduates (ii) it's finds which of our top recommended careers you should most strongly consider, rather than which career you'd be best at.

Ha! I agree it's an issue that people might not assess their skills and preferences correctly, but that seems hard to avoid in a short test. Later we could add more objective measures of skills etc.

All those options are a little different.

OpenPhil is planning to give to substantially different charities from GiveWell recommended charities: http://www.open-philanthropy.org/

Giving to GiveWell or Giving What We Can means your money is spent on research and advocacy, probably resulting in a greater total impact: https://80000hours.org/2015/06/donating-to-giving-what-we-ca... Giving to GiveWell also has the effect of increasing the funds available to OpenPhil, since OpenPhil (to some extent) makes up the short fall in GiveWell's funding.

The Giving What We Can Trust is just a convenient way to have a Donor Advised Fund. The money goes to whichever charities you like. https://www.givingwhatwecan.org/trust

I was focusing on what to do with money in the article. I agree that sometimes time is needed instead.

It's definitely hard to compare the effects of different humanitarian efforts. The problem is that you have to choose something. Whenever you spend money to you're acting based on an implicit ranking. Rather than throw up our hands with "this can't be quantified" and give randomly, or give according to your whims, I think we should try to make the best decisions we can.

That's true it's another effect, but I doubt it means hybrid cars win e.g. it seems you could have a greater impact by speeding up the development of hybrid cars just by giving the $20,000 in cash directly to Toyota.

If you're a young personal at the start of your career, then I agree it could easily make more sense to invest it in yourself (e.g. education, trying out a startup, graduate school).

GiveWell is great, and I seriously considered making "give to GiveWell recommended charities" the recommendation of the article. However, I think it's even better to give to the Open Philanthropy Project (a collaboration between GiveWell and Good Ventures) or to give to GiveWell itself, rather than their recommended charities (which has the effect of increasing OpenPhil's funds). It depends on the extent to which you share GiveWell's values though.

The Open Philanthropy Project is aiming to fund research. I'd prefer to let them do the research working out where to donate, rather than picking scientists myself.

That's a tricky philosophical question, but quick answer would be the long-term flourishing of sentient beings, treating everyone as equal.

https://80000hours.org/career-guide/framework/the-meaning-of...

The question was framed in terms of helping the "common good". If you want to support your local community that's a different question.

My reasoning is that if you accept something like the efficient market hypothesis, investing in startups is higher risk than investing in the S&P500, so the return should be higher. You could also just buy the S&P500 with leverage.

I've also seen some empirical evidence in support, though I haven't vetted it: http://sites.kauffman.org/pdf/angel_groups_111207.pdf

I agree micronutrients are a very promising intervention, along with some other cheap, proven health interventions like malaria nets (I mention Against Malaria Foundation in the post). Here's a list: http://www.givewell.org/international/technical/programs/201...

You also need to find a competent group to implement the intervention. That's basically what the charity evaluators GiveWell and Giving What We Can do.

Here's a promising group specialising in micronutrient supplementation in particular: https://www.givingwhatwecan.org/top-charities/project-health...

I don't agree with (b) low overhead ratio, mainly for these reasons: http://overheadmyth.com/

What are the other problems with the data? We're keen to hear.