Documentation of research data

Documentation of research data means to describe your data, its generation and processing. The documentation should be so detailed that people who access your data can correctly interpret it and reproduce the results. What your research data documentation looks like depends primarily on your discipline and the practices in your research field.

 

Below you will find some concrete examples for the TU Wien:

  • information from the lab journal: samples, history, calibration of the device, environmental conditions, e.g. room temperature, air pressure, time of day, solar radiation, etc., documentation/archiving of the original condition
  • information about how a monitoring system was validated
  • data collection context: structure and organisation of data fields, variable names and description
  • methodology of data collection: definition of codes for access to data and classification schemes
  • algorithms for transformation and processing of data: codes for converting raw data into data that can be physically interpreted
  • codes, e.g. for visualisation: algorithms that are needed to generate images and films
  • README files
  • metadata text files
  • codebooks (e.g. for surveys) with variable names and descriptions
  • theoretical description of the measurement data (e.g. with Asyst, LabVIEW)

 

Research data documentation is a supplement to the description by metadata. Metadata contains only a subset of the information collected in the documentation and is primarily used make your research data findable. If possible, you should also use norm data and controlled vocabularies to describe your data in the documentation. In many cases, English is recommended as documentation language.

 

The documentation helps all project participants and the funder, but also interested research communities and reusing parties. Documentation facilitates classifying the contents correctly and understanding the research work. In addition, documentation is also useful to you, because it helps you to keep track of your data and to clearly structure your activities.

A rough structure of your documentation should be available before the data collection and should be refined as the project progresses. Latest by the time the project is completed, the documentation should also be complete so that the details of data collection and processing are not forgotten. In this way, you can minimise the effort involved in creating the documentation and avoid delays in publishing the data.

Regardless of whether README files, data dictionaries, codebooks, electronic lab notebooks or a combination of all: the essential information about your data must be clearly documented and available together with the data.

The content of the documentation also varies according to project type and discipline. We have compiled some of the most important contents for you:

  • description of the context and the conditions of the experiments
  • description of the method used to collect and process the data, including the tools used (equipment and software)
  • contents of test protocols, field reports, laboratory books
  • device-specific information required for an interpretation of the data
  • description of the quality assurance measures and procedures carried out
  • information on technical standards and calibrations
  • documentation and explanation of the parameters, variables, abbreviations and codes, including column headings in data tables
  • specification of data source when using existing data (references, DOI)
  • documentation of people involved and their tasks
  • documentation of the framework conditions for the long-term storage and subsequent use of data (licenses, usage restrictions, embargo periods, deletion rules)
  • a list of all associated files and folders and a description of their formats and contents
  • links to publications in which the data is used or quoted
  • links to related or related documents and datasets
  • links to all publicly accessible data storage locations
  • recommended citation of the data

The term electronic lab notebooks (ELN) refers to software that helps researchers to document experiments, but also serves as a collaboration tool and for managing laboratory inventory. The software helps to simplify the workflow in the laboratory digitally.

For example, an electronic lab notebook can be used instead of a paper lab book or journal for the documentation of experiments and investigations. It is considered a legal document just like the classic paper laboratory book in patent proceedings or legal disputes about intellectual property.

With the help of an electronic lab notebook you can organise and store your experimental procedures, notes and protocols. One advantage of this type of data documentation is that metadata is automatically created in the software.

Further advantages of the ELN software:

  • use on PCs, tablets and mobile phones
  • multiple users and collaborative work possible
  • direct transfer of data from analysis devices
  • data integrity secured, verifiability by time stamp
  • increased security through access controls
  • easy saving, copying, importing, exporting and linking of data
  • connection to other systems (API)
  • full text search in all contents
  • management of the laboratory inventory

 

At the TU Wien, Jupyter notebooks are offered for testing ELN.

If you are interested, please send a message to datalab@tuwien.ac.at.

 

Jupyter Notebook as electronic lab book

At the TU Wien, Jupyter Notebooks are offered as electronic lab notebooks for testing.

If you are interested, please send a message to datalab@tuwien.ac.at.