Outcome vs. Impact

[lintool]

Hi Eugene, I very much like your analysis here: https://researchsetup.github.io/researchmodel

I have a few comments/questions, but I'll start with this one: In your mind, what's the difference between impact and outcome?

You write:

E[Impact|paper] = P(paper) x E[Outcome|paper]

Why is "outcome" necessary as a mediating variable?

You write:

We can imagine a possible outcome o, which could represent a product, another project, social change; its value could measure profits, social equality, happiness, etc.

Isn't that just the impact? Is it because you want to separate P(o|paper) from P(o)? But then you factor out that term anyway...

[sirrice]

Sorry. I completely missed this since May. Thanks for asking.

I'll up front say that it's a minor point. I wanted to distinguish between concrete events (outcomes) from impact, which is intangible and personal. So it's probably better to say

E[Impact|paper] ~ P(paper) x E[Outcome|paper]

It's also so later on, I don't have outcome on the left and right sides of the "equations".

I think outcome is at the paper level -- what