A survey platform built around experimental research — factorial designs, condition-aware pages and questions, and built-in data-quality checks. Also works well for plain surveys.
Conditions are first-class. Use a factorial design when your manipulation has a clean structure, or just list out the conditions you need. Either way, condition assignment is handled with block randomization and per-cell quotas.
List your independent variables and their levels; ThisStudy generates every cell of the factorial automatically. Set per-cell quotas if you need balance — participants are assigned with concurrency-safe block randomization, so no cell over-fills.
Each participant carries a condition signature, which lets you rebuild conditions mid-study without losing the data you've already collected.
Not every design is factorial. When you need a one-off set of conditions — a control plus three primes, a between-subjects vignette set, an unbalanced design — list them out directly. Same assignment, same quotas, same per-condition targeting downstream.
You can also combine the two: factorial mains plus extra conditions that sit alongside.
Every page and every question has a "show for" setting. Tap the conditions that should see it; leave the rest off. No scripting, no expression language, no separate logic tab.
The same condition picker is used everywhere conditions are referenced, so the control doesn't change shape as you move around the study.
Preview runs are kept separate from real data, and preview links work without an account — so you and your collaborators can walk through the study as many times as it takes.
The preview inspector has a condition cycler — step to any cell of your design and walk the study exactly as a participant assigned to it would see it. No faked assignments, no extra accounts.
The inspector also shows what's hidden on each page and why, so you can check your show-for rules as you go.
Send a collaborator link for the full inspector view, or a pilot link for a clean participant view. Either way, testers can leave notes tagged to the page they're on — no account needed.
Every note lands in one review list on the study, so feedback doesn't scatter across email threads.
Validated scales can be dropped into any study from a built-in catalog. Consent forms, end pages, and other reusable content can be saved as personal templates and used across studies. The catalog grows over time.
Pick from the catalog and a scale arrives pre-built: the items, the response options, the question naming, and the codebook labels are all in place.
Need a scale we don't have? Save it once as a personal template — it's available to every study in your account from then on.
…and more, with new scales added on request.
A dashboard for the lifecycle of the study — recruitment, condition balance, completion, data quality — broken out per condition, with a funnel view of where participants drop out.
Total enrollment, completion rate, mean duration, and quality flags are summarized for each condition. Drill in to see individual participants, response histories, and per-page timing — or filter the whole view down to one cell of your factorial.
The same page also surfaces the study-wide settings — recruitment, redirects, quotas, condition rebuild, sharing — so there's one place to manage the study while it's running.
A dropout funnel shows where people go: arrived, screened out, eligible, in progress, completed. Below it, the participant log lists every session as it happens.
Drill into any participant for their full response history and per-page timing.
Each response is checked against a handful of quality signals. Honeypot fields, no-JS verification, and duplicate-text checks run on submit. Speeder verdicts, click-through patterns, and IP-cluster signals are computed at read time.
Nothing is auto-deleted — flags surface in the dashboard so you decide what to do with them.
The features that probably aren't why you'd switch, but matter once you have.
ThisStudy is run by academic researchers. There's no subscription, no per-response fee, and no plan to add one. We're in beta while we stabilize things. If you want to try it on a real study, ask.
Request access