Q47, run 2026-11-06-0855-haiku
Why do Kesting University and Rostval Technical University swap places in 2024 between full and fractional counting?
Tests: Counting method: why the ranking flips.
Expected: Rostval is on four 2024 works (W21, W27, W28, W29), but on each it has one author of four: 4 x 1/4 = 1. Kesting is on three, with larger shares: W22 alone, half of W20, two of the four authors of W21: 2.
csv+axioms run 1
Full counting favours the institution with more works (I09 has 4 vs 3). Fractional counting divides each work by its author count: Rostval's four 2024 works each have four authors, giving 1/4 each, while Kesting's W22 has one author and W20 has two, so Kesting gets more credit per work despite having fewer works.
csv+axioms run 2
Full counting gives Rostval the lead (4 vs 3) because each work counts once per institution. Fractional counting gives Kesting the lead (2.0 vs 1.0) because Rostval's 2024 works each have four authors, so each Rostval author gets only 1/4, while Kesting's works have fewer authors (W20 has two, W22 has one).
Why: Correctly says Rostval's works have four authors so each gets 1/4, but omits that Kesting holds larger shares via single-author W22 and half of W20 and W21 (mentions W20 two authors and W22 one, but not W21 or half-shares explicitly).
csv+axioms run 3
Rostval's four 2024 works are each multi-author (four authors) with one Rostval author, so fractional counting gives it only 1.0. Kesting has one single-author work (W22, worth 1.0) and two shared ones (0.5 each), so it gets 2.0. Under full counting each institution gets 1 per work, so Rostval's four beat Kesting's three.
csv+axioms run 4
Full counting gives I09 4 works against I07's 3. Fractional counting gives I07 2.0 against I09's 1.0, because Kesting's 2024 works have fewer authors and more of them are Kesting-affiliated (e.g., W22 is single-authored), while each of Rostval's four works has 4 authors with only one from Rostval (0.25 each).
Why: Covers Rostval's four works with four authors each and 0.25 share, and mentions W22 single-authored, but omits that Kesting holds half of W20 and W21.
csv+axioms run 5
Under full counting Rostval has 4 works against Kesting's 3. Under fractional counting Kesting's shares add to 2.0 (W20 1/2, W21 1/2, W22 1) against Rostval's 1.0 (four 2024 works with 1/4 each), because Kesting has a single-author work and half-shares on multi-author works.
csv+axioms+markdown run 1
Rostval (I09) has four 2024 works with four authors each (W21, W27, W28, W29), so each author's share is small, giving 1.0 under fractional counting. Kesting (I07) has fewer co-authors per work (W22 has one author), so its fractional share is larger at 2.0.
Why: Gives Kesting's larger share via W22 single author, but says Rostval's works have four authors each without stating only one is from Rostval (a quarter each), and omits Kesting's half shares of W20 and W21.
csv+axioms+markdown run 2
Under full counting, Rostval (I09) has four 2024 works to Kesting's three, because each work counts once for every institution credited. Under fractional counting, Rostval's works each have four authors with only one Rostval author, so each gives it only 0.25, while Kesting's W22 is single-authored and gives it a full 1.0.
Why: Covers Rostval's 0.25 shares and Kesting's single-author W22, but omits Kesting's half shares of W20 and W21.
csv+axioms+markdown run 3
Full counting gives Rostval 4 (W21, W27, W28, W29) and Kesting 3 (W20, W21, W22). Fractional counting gives Kesting 2.0 and Rostval 1.0, because each Rostval-credited author on a four-author work gets only 1/4, while Kesting's works have fewer authors (W20 has 2, W22 has 1).
Why: Gives Rostval's quarter shares (1/4 on four-author works) and Kesting's single-author W22 plus W20 shares, but doesn't say all four Rostval works have four authors with one Rostval author, and says W20 has 2 authors without mentioning W21's half share.
csv+axioms+markdown run 4
Kesting and Rostval swap because of author count. Under full counting, Rostval has 4 works in 2024 (W21, W27, W28, W29) against Kesting's 3 (W20, W21, W22). Under fractional counting, Kesting's works are split among fewer authors (W22 has one), giving it 2.0 against Rostval's 1.0, since each of Rostval's four-author papers gives it only 0.25.
Why: Gives Rostval's quarter shares and Kesting's single-author W22, but never says Kesting holds half of W20 and W21.
csv+axioms+markdown run 5
Under full counting, Rostval has 4 works in 2024 (W21, W27, W28, W29) against Kesting's 3. Under fractional counting, each of Rostval's four works has four authors with one from Rostval, giving 1.0, while Kesting gets 2.0 from its single-author W22 and two-author W20.
Why: Covers Rostval's quarter shares (four authors, one from Rostval) but says Kesting's 2.0 comes from W22 and W20 as two-author work; omits W21 half share and doesn't explain W22 as a single-author review with halves of W20 and W21 as stated in the checklist