pubmed.info
Drag tiles by their title, resize at the bottom edge.

Randomisation list generator – simple, block and stratified

Create a reproducible allocation list for your trial: up to six groups, any ratio, blocks of varying size and separate lists per centre. Free, no account; the list is created in your browser and not transmitted.

More about this tool

What you can do

  • Simple randomisation: each participant is allocated independently, weighted by the chosen ratio.
  • Block randomisation: the ratio is reached after every complete block; with several block sizes (e.g. 4 and 6) one is chosen at random for each block, so the next allocation is harder to guess.
  • Stratified randomisation: a separate list per stratum, for example per centre or severity grade.
  • Use 2 to 6 groups with any whole-number ratio, such as 1:1 or 2:1, and up to 5,000 participants.
  • Reproduce the list at any time from the seed.

How it works

  • Method: block (recommended) or simple.
  • Group names, separated by commas, and the ratio.
  • Block sizes: each must be a multiple of the ratio sum (1:1 → 2, 4, 6; 2:1 → 3, 6, 9).
  • Total participants, or with stratification one line per stratum with name and number.
  • Seed: leave empty for a random one, or enter a number to recreate an earlier list.

The tile warns about fixed block sizes and about simple randomisation in small samples.

Output and export

You get the distribution per group (and per stratum) and the list with number, stratum, block and group. Download it as a CSV file, whose first line records method, ratio, seed and N, or copy it as a tab-separated table, e.g. into Excel. Record the settings and the seed in the study protocol: the same settings and seed always produce the same list.

Method and sources

Random numbers come from a seedable generator (Mulberry32); the automatic seed is taken from the browser's cryptographic random generator. Blocks are shuffled with the Fisher–Yates algorithm. Background: Schulz and Grimes, Lancet 2002; Kernan et al., J Clin Epidemiol 1999.

Data and privacy

The list is created in your browser and is not transmitted. Your input is remembered only in this browser. The generator is for guidance and does not replace a validated randomisation system or statistical advice; drug trials and other regulated studies require validated randomisation, e.g. through a clinical trials unit.

Frequently asked questions

Simple or block randomisation?

Simple randomisation is unpredictable but can produce clearly unequal groups in small studies. Block randomisation keeps group sizes roughly equal; use varying block sizes in blinded studies.

How do I keep the allocation concealed?

Those enrolling participants must not have access to the list, for instance by using sealed, numbered envelopes or a central system. Anyone who knows the seed and the settings can recreate the whole list.

Are minimisation or adaptive methods available?

No. The generator covers simple, block and stratified randomisation only.

pubmed.info sets no tracking cookies and builds no user profiles. Visits are counted anonymously (cookieless, shortened IP addresses). Ads from Google appear only according to your consent choice.