Oblique factors: A term used in factor analysis for common factors that are allowed to be correlated. [Everitt, B. S. and Dunn, G., 2001, Applied Multivariate Data Analysis, 2nd edn, Arnold, London.]

O'Brien and Fleming boundaries: Amethod of interim analysis in a clinical trial in which very small P-values are required for early stopping of a trial, whereas later values for stopping are closer to conventional levels of significance. See also alpha spending function. [Statistics in Medicine, 1994, 13, 1441–52.]

O'Brien two-sample tests: Extensions of the conventional tests for assessing differences between treatment groups that take account of the possible heterogeneous nature of the response to treatment, and that may be useful in the identification of subgroups of patients for whom the experimental therapy might have most (or least) benefit. [Biometrics, 1984, 40, 1079–89.]

Observational study: A general term for investigations in which the researcher has little or no control over events, and the relationships between risk factors and outcome measures are studied without the intervention of the investigator. Surveys and most studies in epidemiology fall into this class. The classic example of such a study is that by Doll and Hill, which uncovered evidence of a causal relationship between smoking and lung cancer. See also experimental study, prospective study and retrospective study. [New England Journal of Medicine, 1953, 248, 995–1001.]

Occam's razor: William of Occam's fourteenth-century dictum, 'entia non sunt multiplicanda praetor necessitatem', i.e. ‘the assumptions introduced to explain a thing must not be multiplied beyond necessity’.