sky's Shortform

post by sky · 2020-03-28T21:04:11.779Z · EA · GW · 5 comments


Comments sorted by top scores.

comment by sky · 2020-09-05T20:48:57.903Z · EA(p) · GW(p)

Quick thoughts on turning percentages back into people

Occasionally, I experiment with different ways to grok probabilities and statistics for myself, starting from the basics. It also involves paying attention to my emotions, and imagining how different explanations would work for different students. (I'm often a mentor/workshop presenter for college students). If your brain is like mine or you like seeing how other people's brains work, this may be of interest.

One trick that has worked well for me is turning %s back into people

Example: I think my Project X can solve a problem for more people than it's currently doing. I have a survey (N=1200) which says I'm currently solving a problem for 1% of the people impacted by Issue X. I think I can definitely make that number go up. Also, I really want that number to go up; 1% seems so paltry.

I might start with:Ok, how likely do I think it is that 1% could go up to 5%, 10%, 20%?

But I think this is the wrong question to start with for me. I want to inform my intuitions about what is likely or probable, but this all feels super hypothetical. I know I'm going to want to say 20%, because I have a bunch of ideas and 20% is still low! The %s here feel too fuzzy to ground me in reality.

Alternative: Turn 1% of 1200 back into 12 people

This is 12 people who say they are positively impacted by Project X.

This helps me remember that no one is a statistic. (A post which may have inspired this idea to begin with). So, yay, 12 people!

But going from 1% to 5% still sounds unambitious and unsatisfying. I like ambitious, tenacious, hopeful goals when it comes to people getting the solutions they're looking for. That's the whole point of the project, after all. Sometimes, I can physically feel the stress over this tension. I want this number to be 100%! I want the problem solved-solved, not kinda-solved.

At this point, maybe I could remind myself or a student that "shoulding at the universe" is a recipe for frustration. I love that concept, and sometimes it works. But often, that's just another way of shoulding at myself. The fact remains that I don't want to be less ambitious about solving problems that I know are real problems for real people.

I try the percents-to-people technique again:

  • Turn 5% of 1200 back into 60 people. Oh. That's 48 additional people. Also notice: it's only 60 people if we're talking about 48 additional people, while losing 0.
  • Turn 10% back into 120 people. 108 additional people, while losing 0.
  • Turn 20% back into 240 people. 228 additional people, while losing 0.
  • So, an increase of 5% or 20% is the difference between 48 or 228 additional people reached. I know about this program because I work on it, and I know how much goes into Project X right now to reach 12 people. I'm sure there are things we could do differently, but are they different enough to reach 228+ additional people?

Now this feels different. It's humbling. But it piques my curiosity again instead of my frustration: how would we attempt that? Could we?

  • What else do I need to know, to figure out if 60 or 120 or 240 (...or 1000, or 10000) is anywhere within the realm of possibilities for me?
  • Do I have a clear idea about what my bottlenecks or mistakes are in the status quo, such that I think there are 48 more people to reach (while still reaching the 12)? What processes would need to change, and how much?
  • This immediately brings up the response, "That depends on how long I have." (Woot, now I've just grokked why it's useful to time-bound examples for comparison's sake). We could call it 1 year, or 3, or 10, etc. I personally think 1-3 years is usually easier to conceptualize and operationalize.
  • Also, whatever I do next, it's obviously going to take notable effort. I know I can only do so much work in a day. (I probably hate this truth the most. This is definitely where I remind myself not to should at the universe). Now I wonder, is this definitely the program where I want to focus my effort for a while? Why? What if there are problems upstream of this one that I could put my effort toward instead? ...aha, now my understanding of why people care about cause prioritization just got deeper and more personally intuitive. This is a topic for another post.

To return to percentages, here's one more example. Percentages can also feel daunting instead of unambitious:

  • Going from 12 to 60 people is a 400% increase. (Right? I haven't miscalculated something basic? Yes, that's right; thank you, online calculators). 400%! Is that madness?
  • Turn '400% increase' back into 4 additional people reached, for every 1 person reached now.

That may still be daunting. But it may be easier to make estimates or compare my intuitions about different action plans this way.

If you (or your students) are like me, this is a useful approach. It gets me into the headspace of imagining creative possibilities to solve problem X, while still grounding myself within some concrete parameters rather than losing myself to shoulding.

comment by sky · 2020-09-01T22:33:53.969Z · EA(p) · GW(p)

Webinar tomorrow: exploring solutions journalism [for EA writers]:

If EA journalists and writers are planning to cover EA topics, I think a solutions journalism angle will usually be the most natural fit.

The Solutions Journalism Network "train[s] and connect[s] journalists to cover what’s missing in today’s news: how people are responding to problems."

The Solutions Journalism Network is having a webinar tomorrow:

Solutions journalism

"Can be character-driven, but focuses in-depth on a response to a problem and how the response works in meaningful detail

  • Focuses on effectiveness, not good intentions, presenting available evidence of results
  • Discusses the limitations of the approach
  • Seeks to provide insight that others can use"

This is still a less common media angle. The quality of coverage will clearly still vary a lot depending on one's research, editorial input, etc, but this is a better fit than many other media angles one could take to cover topics of interest to you in EA.

More info on this type of journalism:

comment by sky · 2020-03-28T21:04:12.017Z · EA(p) · GW(p)

Should reducing partisanship be a higher priority cause area (for me)?

I think political polarization in the US produces a whole heap of really bad societal/policy outcomes and makes otherwise good policy outcomes ~impossible. It has always seemed relatively important to me, because when things go wrong in the US, they often have global consequences. I haven't put that many of my actual resources here though because it's a draining cause to work on and didn't feel that tractable. I also suspected myself of motivated reasoning: I get deep joy from inter-group cooperation and am very distressed by inter-group conflict.

Then I read things like the thread below and feel like not paying more attention to this is foolish, like I've gone too far in the other direction and underweighted the importance of this barrier to global coordination. I imagine others have written about similar questions and I would be interested in more thoughts.