Datasets and versions
Define measurement or variable datasets in EnaResearch - resolution, time, participants, period, events - and see how every scope change creates a new version.
- Admin panel
A dataset defines what you export from a study. EnaResearch has two kinds: Measurements (long format), one row per measurement or time bucket, and Variables (participant × variable), one row per participant with one column per variable. Every change to a dataset's scope creates a new, fixed version, and every export is linked to a version, so a file can always be produced again.
Required role
- A member of the study team with the export permission (owners and researchers).
Create a dataset
Open the study → Datasets → New dataset. Enter a Title and an optional Description (the analysis it is for; no personal data), then choose the Dataset kind.
Measurements (long format)
| Field | Options |
|---|---|
| Measurement types | Only the study's permitted types |
| Resolution | Raw measurements (one row per measurement); Per minute or Per hour (UTC-aligned) and Per day (participant's local day): mean, minimum, maximum and count; totals for steps and energy; seconds per sleep stage |
| Time | Date and time (UTC and local time); Relative to participation start (no calendar dates); From the event (no calendar dates, negative before the event) |
| Include clinical records | Hospital stays and clinical events in a separate events.csv; notes and records entered in error are left out |
| Period | Calendar: optional from and to dates in the organization's time zone; or Around an event: an anchor event with days before and after |
| Include baseline data | Adds data from before each participant's start, up to the study's baseline days |
With Around an event, each participant's first record of that type (a completed event or a hospital admission) is used; participants without one are left out. Each participant's data is limited to their participation period in any case.
Variables (participant × variable)
- Variables: each becomes a column. The variable's current version is pinned; the dataset keeps using it even if the formula changes later.
- File formats: CSV and the codebook are always produced; add Excel, SPSS (.sav) and R (script) as needed. See Exports.
A variable dataset needs at least one variable; define them first on the Variables page.
Participants (both kinds)
- Participant status: withdrawn participants are included only if you select them.
- Participants (optional): pseudonyms separated by commas, for example
P001, P004. Empty means all participants with the selected status.
Versions
The list shows each dataset's current version. Changing the measurement types, participants, period, resolution or variables creates a new version; earlier exports stay linked to their own version. Changing only the title or description does not.
Click the version to open Dataset versions: each version is a fixed copy of the definition, with who created it, what changed and its exports. Export version N produces a file with that version's scope without changing the definition.
Warnings on a dataset
- "Source data changed after the last export: a new file may give different values." A clinical record or a measurement used by the dataset was corrected after the last export's cutoff.
- "Some variables have a newer version; edit the definition to use it." Turn on Use the latest variable versions when editing; this creates a new dataset version.
Archive removes a dataset from the list. Earlier exports and their manifests remain.
Related articles
Can't find what you need?
Our support team can help, from setting up a study to exporting a dataset. Please include your organization code.
Variables
Define EnaResearch variables without code: metric summaries in a window, time to threshold, length of stay, event duration and time between two moments.
Exports
Export EnaResearch datasets as pseudonymized files - CSV, Excel, SPSS and R - with a data cutoff, manifest and checksum so every file can be reproduced.