Statisticians and data scientists
modelling, estimating what is missing, image and text analysis
Many of these problems come down to estimating what is missing, separating signal from noise, or testing whether a pattern beats chance. Statisticians and data scientists can design those tests and make the reasoning explicit and repeatable.
19 problems need statisticians and data scientists
- 01How many people lived in the Americas in 1492?
Combine the documentary table and archaeological proxies into regional estimates with uncertainty intervals, so the hemispheric total and its interval can be recomputed openly.
Demography · Americas · 1492–1650also needs demographers, archaeologists and geneticists - 02Who printed the anonymous and false-imprint books of eighteenth-century Britain?
Estimate false-match rates for ornament and damaged-type identifications on books with known printers, and set an evidence threshold that keeps wrong attributions below a stated rate.
Texts & authorship · Europe · 1695–1800also needs archivists and conservation scientists - 03How many people lived in China around 1600, and how many died between 1620 and 1680?
Flag outliers and internal contradictions in gazetteer household and land figures, and test each under-registration model against grain-price series and reclamation memorials.
Demography · East Asia · 1580–1680also needs demographers and GIS specialists - 04What new evidence can fix when and why five historically extinct species disappeared?
Benchmark sighting-record extinction estimators on graded databases that mix certain and doubtful records, using species with documented last individuals as controls.
Chronology · Global · 1500–1945also needs archivists, geneticists, philologists, archaeologists, computational linguists and GIS specialists - 05Were real wages in the Yangzi delta and Bengal comparable to England's before 1750?
Fit a model that pools wage and price quotations of different kinds, places and units, and propagate measurement and conversion uncertainty into each decade's welfare ratio.
Economy · Global · 1500–1850also needs economists and archivists - 06Did Gregorian chant descend from Old Roman chant, or did both come from a common ancestor?
Pre-register rival transmission models (Roman-derived, common ancestor, later Roman reworking of Gregorian) and score each by how well it predicts held-out readings in early pitched sources.
Texts & authorship · Europe · 680–1250also needs musicologists and philologists - 07How did Gutenberg cast his type?
Separate position-specific variation across B42 and DK-type copies from sort identity, and compare the observed shapes per letter with simulations of each casting process.
Material & visual · Europe · 1450–1470also needs conservation scientists - 08How many people were carried in the Indian Ocean slave trades, 1500–1850?
Compute documented minimum counts by route, carrier and decade from the transfer records, quantify the gaps, and state an all-carrier total with an explicit uncertainty range.
Economy · Indian Ocean · 1500–1850also needs archivists - 09Does the Indus script encode language, or a non-linguistic sign system?
Pre-register the classifiers and thresholds, then report error rates for the Indus corpus across alternative sign lists and segmentations.
Languages & scripts · South Asia · 2600–1900 BCEalso needs archaeologists and computational linguists - 10How many people were killed in the Irish rising of 1641?
Cluster reports of the same killing across depositions into a deduplicated Armagh register, and compare eyewitness and hearsay totals against the 1650s examinations.
Demography · Europe · 1641–1642also needs archivists and GIS specialists - 11Do khipus record more than numbers, and can their non-numerical signs be read?
Test whether proposed meanings predict unseen cords better than chance and better than the numerical baseline set by Medrano (2024).
Languages & scripts · Americas · 1400–1800also needs archaeologists and computational linguists - 12How many early printed editions have been lost without trace?
Fit Chao1 and Efron–Thisted estimators by format, language and genre on a past snapshot of ISTC or GW, and check their predictions against editions found since.
Texts & authorship · Europe · 1450–1600also needs archivists - 13Were Mendel's pea-hybrid data too close to expectation, and if so, why?
Encode every count in the 1866 paper and the Nägeli letters as one dataset, and test which documented procedures reproduce both Fisher's goodness-of-fit and 2:1 progeny-test anomalies.
Science & knowledge · Europe · 1856–1866also needs geneticists - 14Did Ptolemy observe the Almagest star catalogue or adapt Hipparchus's?
Rebuild the competing tests on one edition of the catalogue matched to Hipparcos and Gaia DR3, reporting results by constellation, magnitude and source subset with doubtful identifications flagged.
Science & knowledge · Mediterranean · 135 BCE–160 CEalso needs astronomers - 15How literate was the population of late imperial China?
Report the proportion of autograph signatures by decade, region, sex and role from a coded sample of several thousand contracts, and link scribes across documents.
Texts & authorship · East Asia · 1644–1911also needs archivists - 16How much American silver reached China between 1550 and 1800?
Convert every published figure to tonnes of fine silver with documented assumptions, then propagate the uncertainty to show where mint, treasury and shipping series double count.
Economy · Global · 1550–1800also needs economists and archivists - 17Who supplied the medical knowledge that Hans Sloane's catalogues attribute to unnamed people?
Design a stratified sample of anonymous entries for manual review so the share traced to named people can be estimated with stated uncertainty and tracked over time.
Science & knowledge · Global · 1680–1753also needs archivists, computational linguists and philologists - 18When did the thylacine go extinct in the wild?
Test extinction-date estimators that use graded sightings on species with known extinction or rediscovery dates, and report how the thylacine estimate shifts with each grading scheme.
Chronology · Pacific · 1888–2019also needs archivists, geneticists and GIS specialists - 19Did Vermeer use a camera obscura or other optical aids?
Test whether tracing from a projection, unaided camera observation or pin-and-string construction best fits the measured geometry across the paintings, and report the comparative fit.
Material & visual · Europe · 1655–1675also needs conservation scientists
Can you help?
Propose an approach or leave a comment on any problem above, or write to say what you could contribute.
Drafted by Claude Opus 5.5