Theory Of Change As A Hypothesis: Choosing A High-Impact Path When You’re Uncertain

by lynettebye3 min read28th Nov 20205 comments

19

Personal DevelopmentCareer Choice
Frontpage

This post appeared first on the EA Coaching blog.

Some life paths are well mapped out. If you want to be a doctor, there is a straightforward path through undergrad, med school, and residency (at least in the US). These goals may require a lot of work, but it’s relatively easy to figure out what your next step is. 

However, many cause areas that effective altruists focus on don’t have clear paths to success. You need to figure out what to do for yourself if you’re doing global priorities research, or working on pandemic preparedness, or founding a new charity. Similarly, if you don’t yet know what you want to work on, you have to figure out what on earth you’re going to do (or off earth, if that’s your jam). 

Really excelling means needing to figure out your own path to success. If it was easy, then many people would follow that path. With that much competition, you would be hard pressed to be one of the best. Hence, you’re more likely to have an impact in neglected career paths because it’s less likely that someone else would have done whatever you did. Similarly, within careers where the steps are fairly straightforward, it often takes doing something unusual to have a large impact. 

This creates a catch 22. If you know exactly what needs to be done to have an impact, then so do most other people with similar or higher levels of experience. So someone else is more likely to already be doing it. And your counterfactual impact is lower, particularly if you’re fairly junior in your field. 

If you don’t know what needs to be done to accomplish some high-level goal, then counterfactually other people are also less likely to, but you’re guessing in the dark about what you should be doing.

So what can you do? 

The first step is to form a hypothesis about what is valuable for you to do. What is your best guess about how to cause the change you want to see in the world? I call this hypothesis your theory of change. It is your best guess about what you should do to actually try moving directly towards your goal. You can see examples of theories of change here and here

You don’t need to get your theory of change perfect right away (and you probably won’t). In fact, it’s best to lightly hold several hypotheses when you’re highly uncertain. You can start with your best guess and then get more information to improve these hypotheses. 

For example, imagine you’re a college freshman who wants to work on climate change and renewable energy. Without a theory of change, you probably do basically what others around you are doing. Maybe you read a lot about climate change in college. If you’re lucky, your university has a climate club that you can join. When you graduate, you apply to some cool sounding climate activist orgs. Unfortunately, you’re competing against a dozen applicants with nearly identical resumes and you don’t get your top choice. 

Let’s imagine instead you plan out your best guess at how to have an impact. You come with two options based on what you know so far: 

Option 1: Study policy in college, do an internship in DC, graduate, work as a congressional staffer for a couple years, then move up to legislative work as a staffer. As a legislative staffer, you can make an impact by writing policy related to climate change. Perhaps later you can run for a political office. 

Option 2: Study business in college, start doing entrepreneurial things on the side, intern at a clean energy startup, graduate, work full time for a solar power company, start your own company that produces solar panels more cheaply. Your cheap panels will make solar power more accessible. 

“But those plans won’t work!”

You’re right. There’s very little chance a college freshman with no prior experience will be able to map out a successful startup or plan a successful political career. 

You don't expect to neatly follow the theory of change. Instead, it’s a good starting point. If you don't have a hypothesis, you don't know which information is important to gather. So you need a first guess, then you can gather information and make a better plan.

Reality will look messier. That freshman doesn’t know what they will do after college. But because they have these possible theories of change, they know that interning at a solar company, taking a policy class, and job shadowing a current staffer would be good ways to get more information. 

If a path seems promising after some exploring, they can go deeper. Maybe the freshman starts a blog walking homeowners through new solar technology, which requires them to read about all of the new innovations in solar technology, interview local solar companies, and help their neighbor install solar panels. Suddenly, they have a ton of experience before they even graduate college. 

Based on that theory of change, they can be testing things, exploring, and generally bumping into the world. In other words, even if they are uncertain now, they can work their way to a better plan. (Related: Dive In

That’s the purpose of a theory of change - you want to have a plausible, concrete hypothesis for how your top options could accomplish your goals. Then you can gather useful information to improve your hypothesis.  

For planning your own theory of change, here is a template spreadsheet that includes a real example.

 

Enjoyed the piece? Subscribe to EA Coaching’s newsletter to get more posts delivered to you.

19

5 comments, sorted by Highlighting new comments since Today at 12:27 AM
New Comment

Thanks for the post! This is just the type of thinking I wanted to do this morning, and I'm finding it and the spreadsheet template a useful motivator.

The given "Dive In" link is broken; I think the correct one is http://mindingourway.com/dive-in-2/

Thanks for writing about this, I like seeing more on this topic.

I can imagine if a person has very low priors on how to have the most impact (or many lightly held hypotheses as you put it) that preserving option value would become important. Do you think it is good to advise this to such people on the margin, and do you see any side effects (e.g. at a group level) from doing so?

I agree that option value is important, but I think there's a trap where preserving option value means never testing one path. I lean toward trying to rapidly and cheaply test multiple paths, while preserving option value.