Umbrella ordering: A commonly observed response pattern in a one-factor design with ordered treatment levels in which the response variable increases with an increase in treatment level up to a certain point, then decreases with further increase in the treatment level. [Journal of the American Statistical Association, 1981, 76, 175–81.]

Unanimity rule: A requirement that all of a number of diagnostic tests yield positive results before declaring that a patient has a particular complaint. See also majority rule. [Statistics in Medicine, 1988, 7, 549–58.]

Unbalanced designs: Synonym for non-orthogonal designs.

Unbiased: See bias.

Uncertainty analysis: Synonym for sensitivity analysis.

Uncle test: A question that might be posed to doctors about to take part in a clinical trial to assess whether it is ethical for them to participate, e.g. ‘Would you be willing to randomize a close relative of yours, or even yourself, into any arm of the study?’ [Statistical Methods in Medical Research, 2002, 11, 1–22.]

Unidentified model: See identification.

Uniform distribution: The probability distribution of a random variable having constant probability over an interval. The most commonly encountered uniform distribution is one over the interval zero to one.

Uniformly most powerful test: A test of a given hypothesis that is at least as powerful as another for all values of the parameter under consideration, and more powerful for at least one value of the parameter.

Unimodal distribution: A probability distribution or frequency distribution having only a single mode.