Design a study. Pre-register it. Analyze with verified statistics on a tamper-evident audit trail.
Compare two groups and get a plain-English answer: is the difference real, or could it be luck? Every number is checked, and your data never leaves your browser.
Drop a CSV here or browse
One numeric column + one group column is all you need
One click runs the whole verified workflow: design, lock, analyze, detect, prove. Every number is computed and checked in your browser.
Every result is checked against published references and recorded on a tamper-evident trail. No vibes, no black boxes.
t, F, χ² and U values cross-checked to known published numbers.
Every action lands on a SHA-256 hash chain. One silent edit breaks it, visibly.
Every commit runs the full suite against those reference values.
ResLab is a lab notebook for research that wants results it can defend. You design a study, seal the plan with a SHA-256 checksum before you see the data, run verified statistics, and keep every action on a tamper-evident audit trail.
When the plan is locked before data collection, a result that is not significant is still a result. The checksum binds the plan, and the integrity detectors catch you if the analysis drifts from it.
Welch and Student t-tests, one-way ANOVA, Mann-Whitney U, and correlation. The engine checks normality and variance assumptions first and picks the test the data actually supports.
Nowhere. Analysis runs entirely in your browser, on your machine. Your CSV never leaves it.
Every action is appended to a hash chain. Each entry stores the hash of the entry before it, so a silent edit breaks the chain and verification fails visibly.
Drop a CSV into the analyzer to get a verified result in seconds, or open the pre-registration workspace to lock a plan first.