Pedigree matrix for radiological data entries in milk contamination model
(7968 reads) 
Pedigree matrix for radiological data entries used in study of
uncertainty in a model describing milk contamination after Tsjernobyl accident
(Funtowicz and Ravetz, 1990)
|
Score
|
Type
|
Source
|
Set-Up
|
|
4
|
Constants |
Reviewed |
Universal |
|
3
|
Deduced |
Refereed |
Natural |
|
2
|
Estimated |
Internal |
Simulated |
|
1
|
Synthesized |
Conference |
Laboratory |
|
0
|
Hypothetical |
Isolated |
Other |
Type represents the relative strength of the inferences by which the data are derived.
Modes:
- Constants. (physical or mathematical).
- Deduced. The data sources are empirically rich. direct measurements; or validated numerical operations on measurements, or derived from established theories.
- Estimated.. The data are inferred from sources of general relevance. Judgements need to be made as to reliability of particular values.
- Synthesized. Only weak data sources are available. Models must be used for production of input data, and these cannot be validated or tested.
- hypothetical. The data are "assumed". They are based on pragmatic considerations, conjecture and perhaps even on speculation. The processes they depict have not been measured and perhaps cannot even be observed.
Source. represents an evaluation of the reliability of the origin of the data in terms of the quality assurance of particular sources. It makes explicit the traditional scholarly evaluations of peer-review.
Modes:
- Reviewed. The data entry is the outcome of a review procedure across a wide range of sources. It may be obtained from a review article in the literature, or elicited from experts with the requisite knowledge.
- Refereed. The data entry is obtained from an article in a reputable scientific journal, for which a full refereeing procedure can be assumed.
- Internal. The data entry is an in-house result which has been given institutional scrutiny.
- Conference. data entry was presented in conference proceedings. lt has passed the quality control of the conference organizers, but has not necessarily been subjected to specific vetting, nor to a full refereeing procedure.
- Isolated. Information which has not undergone any known quality control procedure. Neither the publication, the institution, nor the author has standing.
Set-up describes the conditions of the study which yielded the data entry.
Modes:
- universal. The study conditions are valid for every single situation of interest (e.g., radiological decay constants).
- Natural. The study conditions have no known significant differences from natural situations (e.g., measurements of radiation intake by cattle feeding under normal or representative conditions).
- Simulated. "Artificial" aspects are imposed for the achievement of greater control (e.g., feedlot experiments with cattle), or for generality (e.g., averaging over several different study conditions).
- Laboratory. Experiments and simulations which are significantly different from natural conditions.
- Other. Miscellaneous studies.
Funtowicz, S.O. and Ravetz, J.R., 1990. Uncertainty and Quality in Science for Policy. Dordrecht: Kluwer. | 
