Interview
#2 - Prof David Spiegelhalter on risk, stats and improving understanding of science
80,000 HoursProf David Spiegelhalter, Robert Wiblin, Jess Whittlestone, David Spiegelhalter, Keiran Harris
- A trial episode recorded two years ago is anticipated for release if the abandoned original podcast project allows, while plans include interviewing David Spiegelhalter to discuss his global impact, charitable donations, career paths, and high-impact causes.
- Spiegelhalter predicts a societal tendency to misestimate probabilities by overestimating risks like terrorism while underestimating familiar threats such as heart disease and cancer, noting that measuring the success of his public engagement efforts remains difficult.
- In the UK, public engagement with science is described as strong relative to other nations, with academics viewing it as a duty and Cambridge University taking it seriously for promotion, alongside an upcoming Royal Statistical Society initiative to train statistics ambassadors.
- High demand for statistical skills is expected in agencies monitoring global welfare goals, though academia remains tough; metrics like the value of human life and environmental goods are seen as successful transparency tools despite imperfect measurement capabilities.
- The National Institute for Health and Care Excellence (NICE) evaluates interventions based on costs below £20,000 per Quality-Adjusted Life Year (QALY) and renegotiates those between £20,000 and £30,000, currently applying a 3.5% discount rate that values a life year in 25 years at half its present worth.
- Spiegelhalter argues the current discount rate is too high for long-term policy issues like climate change and suggests that in deep uncertainty, individuals should build flexibility and resilience rather than optimize mathematically, utilizing mixed strategies for careers and giving to avoid putting all eggs in one basket.
- Risk is quantified through micromorts (one in a million chance of dying, e.g., skydiving at seven) and microlives (half-hour reduction in life expectancy from chronic risks), with smoking identified as offering the biggest health return, though QALYs are difficult to apply outside health to areas like education.
- Well-being indices show cultural variation in life satisfaction scores, ranging from 6–8 in the UK and US to higher scores in South America and lower scores in Africa and parts of Asia, while Spiegelhalter warns against over-metricizing activities like academic impact factors and advocates for understanding statistical strengths and weaknesses over treating them as absolute truth.