Executive summary: The author provides detailed feedback on Animal Charity Evaluators’ (ACE) cost-effectiveness analysis (CEA) methods, suggesting ways to systematically estimate years of impact, probability of success, uncertainty modeling, and the evaluation of speculative interventions, while also critiquing the Suffering-Adjusted Days (SADs) metric.
Key points:
Estimating years of impact: The author supports modeling corporate campaigns and legislative reforms as accelerating change but acknowledges the difficulty in systematically estimating this effect. Expected benefits should account for both success and failure probabilities.
Probability of success estimation: The author proposes a weighted reference class approach to estimate success probability, favoring a logistic regression model over direct guessing to improve accuracy.
Modeling uncertainty: While Monte Carlo simulations are common, the author prefers maximizing expected welfare through improved modeling rather than focusing on uncertainty estimates, emphasizing the importance of unbiased point estimates.
Assessing speculative and long-term interventions: The author advocates for more quantification in animal welfare CEAs and suggests modeling research, policy, and fundraising interventions as accelerating beneficial changes, similar to GiveWell’s approach.
Final unit for measuring animal suffering: The author critiques AIM’s Suffering-Adjusted Days (SADs) for underestimating intense pain, arguing that it undervalues high-impact interventions like shrimp welfare and proposing alternative pain intensity estimates.
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.
Executive summary: The author provides detailed feedback on Animal Charity Evaluators’ (ACE) cost-effectiveness analysis (CEA) methods, suggesting ways to systematically estimate years of impact, probability of success, uncertainty modeling, and the evaluation of speculative interventions, while also critiquing the Suffering-Adjusted Days (SADs) metric.
Key points:
Estimating years of impact: The author supports modeling corporate campaigns and legislative reforms as accelerating change but acknowledges the difficulty in systematically estimating this effect. Expected benefits should account for both success and failure probabilities.
Probability of success estimation: The author proposes a weighted reference class approach to estimate success probability, favoring a logistic regression model over direct guessing to improve accuracy.
Modeling uncertainty: While Monte Carlo simulations are common, the author prefers maximizing expected welfare through improved modeling rather than focusing on uncertainty estimates, emphasizing the importance of unbiased point estimates.
Assessing speculative and long-term interventions: The author advocates for more quantification in animal welfare CEAs and suggests modeling research, policy, and fundraising interventions as accelerating beneficial changes, similar to GiveWell’s approach.
Final unit for measuring animal suffering: The author critiques AIM’s Suffering-Adjusted Days (SADs) for underestimating intense pain, arguing that it undervalues high-impact interventions like shrimp welfare and proposing alternative pain intensity estimates.
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.