Documentation overview
What DataPipe does, what it deliberately does not do, and where to go next.
How it works
Connect a storage provider — Google Drive, Dataverse, or Zenodo — to your DataPipe account, create an experiment, and add a few lines of code to the experiment your participants run. Google Drive and Zenodo connect in one click; Dataverse asks for an API token from your institution's installation. The getting started guide covers all of it, in order.
There are three moving parts:
- The experiment your participants run, hosted wherever you put it, sends each participant's data to a DataPipe endpoint with the experiment ID and a filename.
- DataPipe checks the submission against the settings on that experiment — whether it is still accepting data, whether the file validates, whether the session limit is reached — and passes it on.
- Your storage provider receives the file into your own account — the Drive folder, Dataverse dataset, or Zenodo deposition DataPipe created for this experiment.
Which provider you connect changes where files land and what happens when two of them share a name. Choosing a provider
What DataPipe does not do
It does not host your experiment.
You need a separate service to put the experiment your participants run online — GitHub Pages, Netlify, or your university's web hosting. DataPipe only moves data to your storage provider, so you never have to set up a server of your own.
GitHub Pages is a free option. Select "project site" and "start from scratch" in their guide.
Publishing an experiment on GitHub Pages is walked through here. Getting started guide, step 6
It does not keep your data.
Under normal operation, DataPipe routes each participant's data to the storage provider you connected but does not keep a copy. The one exception is an upload that fails: DataPipe queues that submission, retries it automatically, and lets you download it from your experiment dashboard in the meantime.
The full account of what is held, for how long, and who can read it. What DataPipe stores
The page to hand your IRB: architecture, encryption, retention, and jurisdiction in one place. Privacy & information for IRBs
It does not analyze your data.
DataPipe never interprets, summarizes or scores what a participant submitted. The only thing it can add is Psych-DS metadata — a description of your dataset and its variables, written alongside the data when you turn metadata on.
What the metadata files contain and how they are built. Psych-DS metadata
Where to start
These pages are in reading order, but each one answers its own question — arrive at any of them directly.
Sets up one experiment end to end, from choosing a provider to your first test run. Start here if you have not collected data through DataPipe before.
Storage providers
Collecting data
Your data