N = 1.

Studies focused on the behavior of only 1 individual are examined. Despite the limitation on generalizing inherent in such studies, selected examples attest their importance in the history of psychology. There frequency in modern psychology is noted, together with their distribution across major topics. The usefulness of an N of 1 in research is viewed as extending beyond the single-case studies of clinician and personologists. An N of 1 is seen as also appropriate when, for the function considered, intersubject variability is low, when opportunities for observing a given class of events are limited, and when a supposed universal relationship is questioned and the obtained evidence is negative. (51 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

N=1 Experiments: hypothesis testing.

Despite the value of one-subject experiments, they have generally been regarded as being incapable of providing statistical inferences about hypotheses. Statistical inferences about populations cannot be made on the basis of only one experimental subject because more than one subject is needed in order to get an estimate of the inter-subject variability within the population. But one-subject experiments do permit statistical inferences about treatment effects on that particular subject, provided there has been appropriate random assignment of treatment times to treatments. Procedures are described for testing hypotheses when only one subject is used in an experiment, and the validity of the statistical inferences is shown by reference to randomization tests. Analogous rank-order tests and parametric tests are specified. (PsycINFO Database Record (c) 2010 APA, all rights reserved)