Data Quality: Importance of the ‘before analysis’ domain (Theory of Sampling, TOS)
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Data analysts/chemometricians are part of a scientific collegium covering three distinct domains: i) sampling – ii) analysis – iii) data modelling, which are collectively influencing ‘data quality’. There is much more to data quality than analytical uncertainty. There are many situations where analysis is to be made of heterogeneous materials/batches/lots/flowing streams, which need to be sampled appropriately before analysis, following an often long and complex pathway ‘from-lot-to-aliquot’. In most cases, sampling and sub-sampling will dominate the total Measurement Uncertainty budget (MUtotal). Left-out MUsampling contributions may easily overwhelm the Total Analytical Error (TAE) uncertainty by factors 5, 10, 25 or higher as a function of the specific heterogeneity characteristics of the materials and systems targeted, and of the sampling procedure used (grab vs. composite sampling). Focus is here on the consequences of unwittingly ignoring the uncertainties originating in these domains, which e.g. will influence adversely on bilinear component directions (reducing model accuracy) as well as RMSE estimates reflecting precision (analyte concentration prediction, classification, time series prediction) and along the way will also clear up an evergreen mistake: contrary to many beliefs, ‘more data’ will not automatically reduce the magnitude of an unsatisfactory performance RMSE. It is shown how the Theory of Sampling (TOS) is the only guarantor of representative sampling in the critical ‘before analysis’ domain. This article introduces the essential minimum TOS competence which must be mastered by stakeholders from all three domains. The conceptual elements in the TOS system can be visualised as a graphic overview: Kim H. Esbensen has been professor at three universities (National Geological Survey of Denmark and Greenland (2010–2015), Aalborg University, Denmark (2001–2010), Telemark Institute of Technology, Norway (1990–2000) and professeur associé, Université du Québec à Chicoutimi before switching to a quest as an independent consultant in 2015. He is a member of several scientific societies and has published widely across several scientific fields. He is the author of a widely used textbook in Multivariate Data Analysis (chemometrics), and in 2020 published: “Introduction to the Theory and Practice of Sampling”. He was chairman of the taskforce responsible for the world's first horizontal (matrix-independent) sampling standard DS 3077:2024 - Esbensen is the founding editor of: “Sampling Science and Technology (SST)” - https://www.sst-magazine.info/issues/ He can be reached at his homepage https://kheconsult.com/
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.031 | 0.026 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.002 | 0.017 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.004 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it