Most statistical techniques are based on a sample of fixed size n. This may be because n is chosen beforehand or because it is dictated by the circumstances. Sometimes, however, the members of the sample occur one after the other and we can then use sequential methods.

A sequential sampling scheme gives us a ‘stopping rule’ for deciding, after observing each successive member of the sample, whether we can stop sampling and reach a decision or whether we must examine another member.

As an example take the problem of deciding between hypotheses H0 and H1 that a parameter θ takes one of two values θ0 and θ1 respectively, using Wald's Sequential Probability Ratio (SPR) test.