Q7, run 2026-11-06-0206-haiku

← back to the run

How many research works has Mara Ellison authored?

Tests: Everything at once: identity, versions (W02 is W03), document types (W04, W05) and the delisted source (W07).

Expected: count = 19

❌ R1 csv run 1 graded by code

21. Person level: A01 has 6 (W374 is retracted), A02 has 7, A03 has 4 and A101 has 4.

JSON: {"count": 21}

Why: count: expected 19, got 21

❌ R1 csv run 2 graded by code

10 research works: W01, W06, W12, W238, W374, W464, W676, W813, W949, W1005. W04, W05 and W271 are excluded as editorial or erratum types.

JSON: {"count": 10}

Why: count: expected 19, got 10

❌ R1 csv run 3 graded by code

26 (person-level). A01 has 8 research works, A02 has 9, A03 has 4 and A101 has 5; non-research rows are excluded.

JSON: {"count": 26}

Why: count: expected 19, got 26

❌ R1 csv run 4 graded by code

(no answer found)

Why: no structured answer

❌ R1 csv run 5 graded by code

(no answer found)

JSON: {"count": 9}

Why: count: expected 19, got 9

❌ R3 csv+padding run 1 graded by code

8 for A01: W01, W06, W12, W238, W374, W464, W676, W1005. Counting the same-ORCID records A02, A03 and A101 as well would give 26.

JSON: {"count": 8}

Why: count: expected 19, got 8

❌ R3 csv+padding run 2 graded by code

7 research works for record A01 (W01, W06, W12, W238, W374, W676, W1005). Records A02, A03 and A101 would add to this if merged by ORCID.

JSON: {"count": 7}

Why: count: expected 19, got 7

❌ R3 csv+padding run 3 graded by code

8: W01, W06, W12, W238, W374, W464, W676, W1005. W04, W05, W271, W813 and W949 are non-research types.

JSON: {"count": 8}

Why: count: expected 19, got 8

❌ R3 csv+padding run 4 graded by code

8. A01 is on W01, W06, W12, W238, W374, W676, W813 and W1005. Counting all ORCID-matched records (A01, A02, A03, A101) would give a different number.

JSON: {"count": 8}

Why: count: expected 19, got 8

❌ R3 csv+padding run 5 graded by code

25 research works (31 total, minus the EDT/ERR records W04, W05, W271, W813, W936, W949).

JSON: {"count": 25}

Why: count: expected 19, got 25

✅ R4 csv+markdown run 1 graded by code

19 research works. Counting the ORCID group, I excluded retracted W374, editorials and errata, K0 sources (W07, W449, W763) and W813/W949. Preprints W02, W612 and W663 are counted under their published versions W03, W869 and W464.

JSON: {"count": 19}

❌ R4 csv+markdown run 2 graded by code

20. Excluding editorials, errata, paratext, the delisted S04 and S13 works, W374 (retracted), and merging preprints into their published versions, the distinct research works are W01, W03, W06, W12, W19, W206, W238, W464, W555, W612, W676, W817, W843, W869, W905, W972, W984, W1005, W1048 and W1050.

JSON: {"count": 20}

Why: count: expected 19, got 20

✅ R4 csv+markdown run 3 graded by code

19 research works, person-level across A01, A02, A03 and A101 (W02/W612 are preprints counted under W03/W869).

JSON: {"count": 19}

❌ R4 csv+markdown run 4 graded by code

(no answer found)

Why: no structured answer

✅ R4 csv+markdown run 5 graded by code

19. The count includes W02 folded into W03 and W612 and W663 folded into their published versions. It excludes editorials and errata (W04, W05, W271, W936, W949), retracted W374, and K0-source W07, W449 and W763.

JSON: {"count": 19}

❌ R5 csv+ontology run 1 graded by code

22. The research works are W01, W03 (with its preprint W02), W06, W07, W12, W19, W206, W238, W449, W464 (with W663), W555, W763, W817, W843, W869 (with W612), W905, W972, W984, W1005, W1048 and W1050. W374 is retracted, and W04, W05, W271, W813, W936 and W949 are not research types.

JSON: {"count": 22}

Why: count: expected 19, got 22

✅ R5 csv+ontology run 2 graded by code

19. Mara Ellison's research works across A01, A02, A03 and A101 are W01, W03, W06, W12, W19, W206, W238, W464, W555, W676, W817, W843, W869, W905, W972, W984, W1005, W1048, W1050.

JSON: {"count": 19}

✅ R5 csv+ontology run 3 graded by code

19 research works (W01, W03, W06, W12, W19, W206, W238, W464, W555, W676, W817, W843, W869, W905, W972, W984, W1005, W1048, W1050).

JSON: {"count": 19}

❌ R5 csv+ontology run 4 graded by code

20. Of the 32 records for this person, I excluded editorials and errata (W04, W05, W271, W813, W936, W949), the retracted W374, and the K0-source works W07 and W449. I also merged W02 into W03, W612 into W869 and W663 into W464 as single works.

JSON: {"count": 20}

Why: count: expected 19, got 20

✅ R5 csv+ontology run 5 graded by code

19 research works, all of them within the Albury subtree.

JSON: {"count": 19}

✅ R10 executed run 1 graded by code

19. These are A01's research credits, the same list as Q6.

JSON: {"count": 19}

✅ R10 executed run 2 graded by code

19. A01's works are all at I02; 10 of its 29 works are not research output (W04, W05, W07, W271, W374, W449, W763, W813, W936, W949).

JSON: {"count": 19}

✅ R10 executed run 3 graded by code

19. A01 is credited on 19 research works (e.g. W01, W03, W06, W12, W19, W206).

JSON: {"count": 19}

✅ R10 executed run 4 graded by code

19.
Of A01's 29 works, 10 are non-research (W04, W05, W07, W271, W374, W449, W763, W813, W936, W949). The other 19 are research output.

JSON: {"count": 19}

✅ R10 executed run 5 graded by code

19 research works for A01: W01, W03, W06, W12, W19, W206, W238, W464, W555, W676, W817, W843, W869, W905, W972, W984, W1005, W1048, W1050.

JSON: {"count": 19}