Draft version 1.7 beta — Revised October 31, 2026

← Return to the table in Chapter 4, § 5

Nonretention of data records

A topic page linked from Chapter 4, § 5, Examples of Reported Violations of Enforceable Ethics Codes.

Definition

Non-retention of data records occurs when the documents or specimens that would enable an independent investigator to reproduce an analysis are no longer available. "Non-retention" means some or all of these records have been discarded, or if they still exist they are in a state that renders them useless. In a laboratory analysis, the necessary records might include paper or digital records of, for example, an X-ray diffractometer's analytical output for a given specimen, along with the record of the most recent machine calibration would also be useful to someone auditing the results.

The goal is to retain information related to the primary data and the subsequent analysis so it could be fully audited and substantially reproduced if necessary to verify or validate the process and results. A typical critiique of scientific data might involve the central value of each datum as well as the uncertainty around that central value. Without sufficient preserved data records, neither the central values or their uncertainties can be independently evaluated.

In my work, the final report or publication is just one product of the research. I try to document the source of all primary data, exactly how the data were used in the analysis, the computational codes I developed or adopted for the analysis, the commercial or open-source software that was used (name, version, publisher, etc.), graphics codes or software used to generate the key graphics, and so forth. That documentary record is saved as an archive. The most likely use of this archive is as the seed of a future project that expands upon or refines the original project, or of a new project that will use the same process.

If the data records are not retained, there is no proof that the analysis and results are anything but the product of the imagination of a person or team regarding some aspect of how a world might work in some parallel universe. But true science is used to characterize reality, not to construct fantasies. The preserved data record is the necessary ingredient in any review process to test whether the work was done as reported.

Relevant provisions in geoethics codes

AAPG – § X; AEG – § X; AGI – § X; AGU – § X; AIPG – § X; ASBOG – § X; CAL – § X; GSA – § X.

Code Abbreviations

← Return to the table in Chapter 4, § 5