Completely Randomised Design (Unequal Replication)

A completely randomised design where each treatment can appear a different number of times. Treatments are randomly allocated to experimental units with no blocking structure.

Design Reference — what this design is, parameters explained, output guide

Completely Randomised Design — Unequal Replication

A completely randomised design where each treatment can appear a different number of times. Treatments are randomly allocated to experimental units with no blocking structure.

Use when: Some treatments require more replication than others (e.g. a control treatment that needs higher precision), and there is no known blocking structure.


Parameters

Parameter Type Default Min Max Description
experiment_name text "" — — Name of the experiment
treatment_factor text "Variety" — — Label for the treatment factor column
treatment_count integer 2 2 500 Number of distinct treatments
per_treatment_reps list of integers [2, 2] 1 100 Number of replicates for each treatment — one value per treatment, comma-separated
unit_label text "Plot" — — Label for each experimental unit
treatment_names list None — — Optional custom names for each treatment (comma-separated)
seed integer 0 — — Random seed — 0 for random, or set a value for reproducibility

Constraints

  • per_treatment_reps must contain exactly treatment_count integer values
  • Each value in per_treatment_reps must be between 1 and 100
  • Total units = sum(per_treatment_reps) must not exceed 5,000

Output

The output is a single list (no layout view).

Column Description
Unit Sequential unit number
unit_label (e.g. Plot) Plot number within the randomised sequence
treatment_factor (e.g. Variety) The treatment allocated to this unit

Total units: sum(per_treatment_reps)


Algorithm

  1. Build a treatment list: treatment i repeated per_treatment_reps[i] times
  2. Shuffle the entire list using the seeded RNG
  3. Assign sequential unit numbers

Example

With treatment_count = 3, per_treatment_reps = [3, 1, 2], treatments T1/T2/T3:

  • T1 appears 3 times
  • T2 appears 1 time
  • T3 appears 2 times
  • Total = 6 units, randomly ordered
📄 Static Preview — This is a read-only snapshot of the EDGAR interface hosted on GitHub Pages. Design generation and downloads are not available here. Run the app locally to generate and download experimental designs.
Enter comma-separated integers.
Enter comma-separated names, or leave blank for defaults.
Use 0 for a random seed, or set a specific value for reproducibility.