Q14, run 2026-11-06-0206-sonnet
How many research works did GB institutions publish in 2024, under full counting and under fractional counting (each work split equally among its authors, each author's share going to their institution's country)?
Tests: Counting method at country level.
Expected: full = 46 fractional = 42
csv run 1
Full counting gives 49 works and fractional gives about 44.58, for 2024 works with a GB author.
JSON: {"full": 49, "fractional": 44.58}
Why: full: expected 46, got 49; fractional: expected 42, got 44.58
csv run 2
Full 48, fractional 44.0.
JSON: {"full": 48, "fractional": 44}
Why: full: expected 46, got 48; fractional: expected 42, got 44
csv run 3
Full counting gives 48 works and fractional counting gives 44, from my hand scan of the 2024 records.
JSON: {"full": 48, "fractional": 44}
Why: full: expected 46, got 48; fractional: expected 42, got 44
csv run 4
Full 45, fractional 41.5.
JSON: {"full": 45, "fractional": 41.5}
Why: full: expected 46, got 45; fractional: expected 42, got 41.5
csv run 5
Full 47, fractional 43.0.
JSON: {"full": 47, "fractional": 43}
Why: full: expected 46, got 47; fractional: expected 42, got 43
csv+padding run 1
Full 49, fractional 44.75.
JSON: {"full": 49, "fractional": 44.75}
Why: full: expected 46, got 49; fractional: expected 42, got 44.75
csv+padding run 2
Full 48, fractional 44.
JSON: {"full": 48, "fractional": 44}
Why: full: expected 46, got 48; fractional: expected 42, got 44
csv+padding run 3
Full 47, fractional 43. This counts 2024 research works with at least one GB-affiliated author. Fractions below 1 are W28, W29, W273, W385, W833 and W984 (0.5 each) and W304, W308 and W787 (2/3 each).
JSON: {"full": 47, "fractional": 43}
Why: full: expected 46, got 47; fractional: expected 42, got 43
csv+padding run 4
Full 49, fractional ≈44.67. The 49 comes from 29 works with an I02-unit author, 12 with an I05-unit author (W28 and W29 among them) and 8 more with an I13-only author.
JSON: {"full": 49, "fractional": 44.67}
Why: full: expected 46, got 49; fractional: expected 42, got 44.67
csv+padding run 5
Full 50, fractional 45.75. Fifty 2024 works have a GB-affiliated author, and each author's share goes to their institution's country.
JSON: {"full": 50, "fractional": 45.75}
Why: full: expected 46, got 50; fractional: expected 42, got 45.75
csv+markdown run 1
Full 47, fractional about 42.67 (128/3). I counted 2024 research works with at least one credited GB institution, and summed the GB-credited authors' share of each work.
JSON: {"full": 47, "fractional": 42.67}
Why: full: expected 46, got 47
csv+markdown run 2
Full 46, fractional 42. Forty-six 2024 research works have at least one GB-credited author. Nine of them are partly GB, and the rest are fully GB.
JSON: {"full": 46, "fractional": 42}
csv+markdown run 3
Full 45, fractional 41.0. These count 2024 research works with at least one credited GB institution; I06, I13 and I21 count as GB.
JSON: {"full": 45, "fractional": 41}
Why: full: expected 46, got 45; fractional: expected 42, got 41
csv+markdown run 4
Full 44, fractional 40. This was counted by hand over a large set, so it carries some risk of a missed or extra work.
JSON: {"full": 44, "fractional": 40}
Why: full: expected 46, got 44; fractional: expected 42, got 40
csv+markdown run 5
Full 46, fractional 42. This counts 2024 research works with a credited GB institution, including the affiliated institutes I06 and I21.
JSON: {"full": 46, "fractional": 42}
csv+ontology run 1
Full 48, fractional about 43.67 (131/3). This covers GB-credited research works from 2024.
JSON: {"full": 48, "fractional": 43.67}
Why: full: expected 46, got 48; fractional: expected 42, got 43.67
csv+ontology run 2
Full 45, fractional 41. Fractional is 36 whole works plus nine partial ones: 0.5 for W28, W29, W273, W385, W833 and W984, and 2/3 for W304, W308 and W787.
JSON: {"full": 45, "fractional": 41}
Why: full: expected 46, got 45; fractional: expected 42, got 41
csv+ontology run 3
Full 45, fractional about 40.67 (122/3). That is 45 distinct works with at least one GB-credited author in 2024.
JSON: {"full": 45, "fractional": 40.67}
Why: full: expected 46, got 45; fractional: expected 42, got 40.67
csv+ontology run 4
Full counting gives 46 works; fractional counting gives about 40.67 (122/3).
JSON: {"full": 46, "fractional": 40.67}
Why: fractional: expected 42, got 40.67
csv+ontology run 5
Full 47, fractional about 42.67 (128/3). I counted 2024 works with at least one GB credited institution (I01–I06, I13, I19, I20, I21) and summed the GB authors' shares.
JSON: {"full": 47, "fractional": 42.67}
Why: full: expected 46, got 47
executed run 1
Full counting gives 42 works. Fractional counting gives about 38.83.
JSON: {"full": 42, "fractional": 38.83}
Why: full: expected 46, got 42; fractional: expected 42, got 38.83
executed run 2
Full 46, fractional 42 (2024 research works with GB-credited authors).
JSON: {"full": 46, "fractional": 42}
executed run 3
Full 47, fractional 43. The GB institutions are I02, I03, I04, I05, I06, I13, I19, I20 and I21.
JSON: {"full": 47, "fractional": 43}
Why: full: expected 46, got 47; fractional: expected 42, got 43
executed run 4
Full 46, fractional 42. I counted 2024 research works with at least one GB-credited author, and summed the GB share of authors per work; I scanned the list by hand, so a miss is possible.
JSON: {"full": 46, "fractional": 42}
executed run 5
Full 46, fractional 42. Forty-six 2024 research works have at least one GB-credited author. Of these, 37 are entirely GB-credited and 9 are partly GB (shares 0.5 or 2/3), giving 37 + 5 = 42.
JSON: {"full": 46, "fractional": 42}