25 lines
591 B
Markdown
25 lines
591 B
Markdown
---
|
|
title: "Randomization"
|
|
type: concept
|
|
tags: [statistics, experimentation]
|
|
sources: [project-management-experiment-tracker]
|
|
last_updated: 2026-04-20
|
|
---
|
|
|
|
## Definition
|
|
Randomization is the assignment of units to experimental variants by chance to reduce bias and support causal inference.
|
|
|
|
## Core Principles
|
|
- Randomly assign at the correct unit level
|
|
- Keep assignment stable during the experiment
|
|
- Avoid contamination between variants
|
|
- Validate balance across key segments
|
|
|
|
## Related Concepts
|
|
- [[A/B Testing]]
|
|
- [[Hypothesis Testing]]
|
|
|
|
## Related Entities
|
|
- [[Experiment Tracker]]
|
|
|