RSS

Coun­ter­fac­tual reasoning

TagLast edit: 29 Mar 2022 18:18 UTC by Pablo

Counterfactual reasoning involves scenarios that will occur if an agent chooses a certain action, or that would have occurred if an agent had chosen an action they did not. For instance, we can consider a counterfactual scenario in which the effective altruism community was called ‘effective giving’ rather than effective altruism.

When we rank actions, we generally want to consider not just how good an action is, but how good it is relative to the alternatives. This is implicitly assumed by the framework of idealized decision-making, but it is useful to state it explicitly.

One related heuristic is replaceability: it may be the case, for instance, that if you do not take a certain action, then someone else will take it instead.

Unfortunately, counterfactuals are often difficult to evaluate. Even after an action is taken, there will in many cases remain substantial uncertainty about what would have happened if one had acted otherwise. This means that we will often be unsure about whether we have acted in the best possible way.

Further reading

Ord, Toby (2014) Drones, counterfactuals, and equilibria: Challenges in evaluating new military technologies, Future of Humanity Institute, University of Oxford.

Sempere, Nuño (2019) Shapley values: Better than counterfactuals, Effective Altruism Forum, October 10.

Related entries

impact assessment | markets for altruism

Shap­ley val­ues: an in­tro­duc­tory example

Stan Pinsent12 Nov 2023 13:35 UTC
9 points
0 comments3 min readEA link

Malaria bed­net and medicine overlap

MichaelStJules5 Sep 2023 23:55 UTC
24 points
4 comments9 min readEA link

Shap­ley value, im­por­tance, eas­i­ness and neglectedness

Vasco Grilo5 May 2023 7:33 UTC
27 points
0 comments4 min readEA link

Co­op­er­a­tive or Com­pet­i­tive Altru­ism, and An­ti­so­cial Counterfactuals

Davidmanheim20 Mar 2023 17:54 UTC
53 points
25 comments5 min readEA link

[Question] What do EA’s think about Bayesian in­fer­ences ver­sus other types?

Noah Scales4 Feb 2023 11:09 UTC
4 points
0 comments1 min readEA link

The Challenges with Mea­sur­ing the Im­pact of Lobbying

Animal Ask31 Oct 2022 16:32 UTC
57 points
1 comment33 min readEA link

How can I tell whether my fundrais­ing pro­ject is harm­fully can­ni­bal­iz­ing ex­ist­ing dona­tions v.s. ac­tu­ally in­spiring new peo­ple to donate?

PerseusPrior31 Aug 2022 5:01 UTC
30 points
6 comments1 min readEA link

[Question] What should I ask Alan Há­jek, philoso­pher of prob­a­bil­ity, Bayesi­anism, ex­pected value and coun­ter­fat­u­als?

Robert_Wiblin1 Jul 2022 13:23 UTC
25 points
12 comments1 min readEA link

Gen­eral equil­ibrium thinking

Karthik Tadepalli11 Jun 2022 19:27 UTC
27 points
4 comments3 min readEA link

Coun­ter­fac­tual im­pact when your cowork­ers share your values

Kirsten16 Oct 2021 11:48 UTC
25 points
4 comments5 min readEA link

[Question] What is the marginal im­pact of a small dona­tion to an EA Fund?

BrownHairedEevee23 Nov 2020 7:09 UTC
12 points
4 comments1 min readEA link

Shap­ley Values II: Philantropic Co­or­di­na­tion The­ory & other mis­cel­lanea.

NunoSempere10 Mar 2020 17:36 UTC
35 points
1 comment29 min readEA link

Shap­ley val­ues: Bet­ter than counterfactuals

NunoSempere10 Oct 2019 10:26 UTC
141 points
54 comments14 min readEA link

The coun­ter­fac­tual im­pact of agents act­ing in concert

John G. Halstead27 May 2018 10:54 UTC
11 points
23 comments5 min readEA link

Triple count­ing im­pact in EA

Joey26 May 2018 23:00 UTC
56 points
19 comments4 min readEA link

Match­ing-dona­tion fundraisers can be harm­fully dishonest

BenHoffman12 Nov 2016 3:30 UTC
19 points
4 comments15 min readEA link