Collection level documentation

Providing high-level context, study design, and methodological information

Collection-level documentation provides high-level information about the research context and design underpinning a data collection. It explains how and why the data were created, how they were collected, and how they were prepared for analysis and sharing.

Without this contextual information, data may be difficult to interpret accurately. Collection-level documentation enables secondary users, and future users within the original research team, to understand the assumptions, methods and decisions that shaped the data collection.

Good collection-level documentation should allow someone with no prior knowledge of the project to understand how the research was carried out and what the data represent.

Collection-level documentation should include information on the following:

project history
aims, objectives and hypotheses
investigators and collaborating institutions
funders.

types of data (e.g. interviews, images, survey responses)
structure of data files
number of cases, records and files
relationships between files or components.

data collection methodology and protocols
sampling design and sample structure
workflows and instruments used
hardware and software used
data scale and resolution
digitisation or transcription methods where applicable
secondary data sources where applicable.

data collectors
geographic coverage
dates of data collection
temporal coverage of the phenomena studied.

tools and procedures used to edit, clean, code and classify the data
derivations or transformations applied.

anonymisation or disclosure control measures
for time series or longitudinal data, changes to methodology, variables, question text, labelling, measurement or sampling across waves.