Research Data Management Plan and Record-Keeping

Description

Decide and document before collection who owns the data, how it is stored, shared, retained and disposed of, and keep a dated, complete study record another researcher could reconstruct.

A data management plan settles ownership (who collected the data, under whose direction, and any obligation to assign it to an employer, client or funder), formats and metadata, storage and preservation, sharing and its limits (privacy, confidentiality, security, IP), roles and responsibilities including cover when people leave, retention periods and disposal. The companion record-keeping practice treats the study notebook as the core of the research record. NSF investigators judge records on completeness (all activities, not only successes), linkage to electronic files by name and location, regular supervisor review, accuracy sufficient for another researcher to reconstruct the work, and safekeeping with off-site backup. Errors are struck through once and explained, never erased; reinterpretations go in new cross-referenced entries. Electronic records need protected storage, encryption, read-only archives, time stamps or hashes and daily backup.

Attributes

Loss type
Siloed
Effort
Days

Steps

  1. Establish ownership of the data: who collects it, under whose direction, and any obligation to an employer, client or funder.
  2. Describe the data to be collected, its formats and metadata standards, and where every file will live.
  3. Specify storage, security and preservation during and after the project, including encryption of laptops and portable media and daily backup.
  4. Specify how and when data may be shared, and the privacy, confidentiality and IP protections that limit sharing.
  5. Assign roles for data management, including cover when key people leave, and set retention periods and disposal or archiving from legal and organisational rules.
  6. Keep a dated study log: purpose, methods and deviations, raw observations recorded immediately, locations of off-book files, discussion and an explicit conclusion.
  7. Never erase: strike errors with one line and explain; record decisions, correspondence and every contributor's role.
  8. Have records reviewed regularly (weekly or monthly) and route external data requests through a designated owner.

Sources

  • Scientific Integrity (Macrina, 2014)
  • Miles, M. & Huberman, A. M., Qualitative Data Analysis: An Expanded Sourcebook, 2nd ed. (1994)

Also known as

  • Research data management plan, ownership and sharing
  • Data management plan
  • DMP
  • Data ownership and retention policy
  • Data sharing plan
  • Research notebook
  • Laboratory notebook
  • Research log
  • Electronic laboratory notebook
  • ELN

Categories

  • Research Operations
  • Research Design & Rigor