---
title: "A. Philip Dawid"
type: "entity"
entity_kind: "person"
status: "hub"
canonical_name: "A. Philip Dawid"
aliases: ["Philip Dawid","Alexander Philip Dawid"]
first_seen: "2026-08-13T00:00:00.000Z"
connects_to: ["Dawid-Skene model","crowdsourcing label aggregation","EM algorithm","forensic and legal statistics","Allan M. Skene"]
writer_model: "claude-sonnet-5"
seek_code_commit: "17d9798"
---


Bayesian statistician who, with Allan M. Skene, published "Maximum Likelihood Estimation of Observer Error-Rates Using the EM Algorithm" in 1979 — a method for reconciling disagreeing raters (in the paper's own worked example, five anaesthetists rating patients' fitness for surgery) that later statisticians named the Dawid-Skene model. He went on to a career as a leading Bayesian statistician at UCL and Cambridge, was elected a Fellow of the Royal Society in 2018, and is also known for work in forensic statistics (probabilistic evaluation of evidence in court) and causal-inference notation. Matters to this vault as the traced root of a real, two-hop citation lineage: the Dawid-Skene model is the explicitly credited ancestor of Li & Yu (2014)'s crowdsourcing weighted-majority-voting method, which RA-RAG (2025) explicitly extends for LLM retrieval — a 46-year, citation-verified inheritance rather than the vault's more common pattern of independent reinvention under a shared name.

## References
- 2026-08-13: Co-creator of the 1979 confusion-matrix/EM model that Li & Yu (2014) name as crowdsourcing WMV's direct ancestor, itself extended by RA-RAG (2025). ([[claim-li-yu-2014-credits-dawid-skene-1979-as-wmv-ancestor]], [[claim-dawid-skene-1979-worked-example-is-anaesthetist-fitness-ratings]])