Evidence file for an audit of measurement-condition reporting in terahertz tissue studies, 2003–2026.

Description

This dataset is the complete evidence file for a reporting audit of terahertz measurements of biological tissue. Twenty-four tissue studies published between 2003 and 2026 were assessed against six criteria describing the conditions under which each measurement was made: measurement geometry, path length or penetration depth, usable bandwidth, referencing procedure, sample hydration state, and measurement temperature. Each of the 144 study-by-criterion cells was coded as reported or not reported, and every cell carries the verbatim sentence from the source paper on which the decision rests, or the words NOT STATED where the condition is absent, together with an assessor note recording near-misses and the reason they do not satisfy the criterion. The corpus was coded independently by two assessors. The second coded blind, working from the criteria and the ambiguity rules alone and without sight of the first coding. The dataset contains both codings before adjudication, the adjudicated result, and a live calculation of inter-assessor agreement, so that the reported agreement of 87.4 per cent across 143 cells (Cohen's kappa 0.75) can be reproduced and any individual decision contested. Contents, one Excel workbook of six sheets: • Coding — the adjudicated assessment of all 144 cells with supporting quotations. • Summary matrix — the 24 × 6 result with column totals, formula-driven. • Protocol — the corpus definition, the six operational criteria, the ambiguity rules and the coding procedure. • Judgement calls — the decisions resting on a judgement at the margin, each with the effect of deciding it the other way. • Independent codings — every cell showing the first assessor's original mark, the second assessor's blind mark, the adjudicated mark, and whether it was revised. • Agreement — percentage agreement and Cohen's kappa per criterion, computed from the codings sheet. Prepared as Supplementary Material for the manuscript "Reporting of measurement conditions in terahertz biomedical spectroscopy and imaging: a critical review of comparability across twenty-four tissue studies".

Institutions

• Universiti Malaysia Terengganu

  Terengganu, Kuala Terengganu

Categories

Licence