How to document data

Well-documented data are better understood and can be more easily reused. The guidelines provided below, also based on the CESSDA guidelines, help in producing an accurate documentation of your data, which is essential to increase the visibility of the author and to expand the possibilities for their reuse.

It is important to provide a thorough description of how the data have been handled, documenting its construction process and the whole life cycle, including: data collection, organization, and manipulation. Therefore, it is necessary to provide information on:

• study context: specifying research objectives, the methodology used, information on techniques and instruments used, sampling information, etc.
• data creation: describing how variables were constructed, labels, coding system, software, packages and scripts used for data management and analysis, transcription procedures, and any other information relevant to understanding the data creation process.

This information can be provided by depositing documents such as publications, methodological notes, research projects, guidelines, instructions, or other research materials such as questionnaires, interview transcripts, scripts, etc.

Data documentation is done through metadata. Metadata refers to all information that provides a detailed description of the data and the context in which it is created, following a predefined model. DASSI has created its own metadata model (DaMM - DASSI Metadata Model), which uses metadata according to the international standard DDI Codebook 2.5. Furthermore, DaMM operates in accordance with the CESSDA Metadata Model (CMM), using controlled vocabularies, thesauri, and other tools shared with CESSDA to make metadata interoperable. The information necessary for constructing metadata is collected by DASSI through the deposit form, a brief questionnaire for collecting the key information required by DASSI. In particular, the deposit form collects the following information:

• general study information: relevant information to identify the study within which the data were created, such as the study language, title, or a brief abstract.
• study participants: information describing those who managed and created the data to be shared, namely the authors and collaborators. Participant information also includes their affiliation and, if available, a persistent identifier (such as OrcID or similar).
• study scope: information describing the characteristics of the context in which the data were collected, such as keywords, study topics, data collection period, geographical context, etc.
• methodological aspects: information about the methodology and research techniques employed, such as units of analysis, reference universe, sampling procedures, description of any sampling weights, data collection methods, etc.
• additional study information: for example, information about any funding used for the study or a list of publications in which the data have been used.