Random assignment is the procedure of placing participants into experimental conditions by chance. Each person has an equal opportunity to be in any group.

The method is the cornerstone of true experiments. By distributing both known and unknown participant characteristics evenly across conditions, random assignment makes the groups comparable at the outset. Any subsequent difference in the dependent variable can then be attributed to the independent variable rather than to pre-existing differences between people. Coin flips, random-number tables, or computer algorithms are common tools.

Random assignment is distinct from random sampling. Sampling concerns how participants are selected from a population; assignment concerns how those participants are distributed among conditions. An experiment can use random assignment even if the sample itself is not random. When assignment is not random, the design becomes quasi-experimental and causal claims are weakened.

Chance is used deliberately so that systematic bias does not favor one condition over another.