Answers are grounded in this paper and cite their sources.
Time-sampling thechniques have been used with considerable frequency in recent years in observational studies of behavior, particularly in studies of the social behavior of young children. In the earliest of these studies, behavior was recorded in terms of occurrence or nonoccurrence within the time sample, the measure of frequency being the number of time samples in which the behavior occurred. In later studies, the time sample was subdivided into finer units, behavior was recorded in terms of occurrence within these subunits and it became possible to deriver measures of variability as well as of frequency for the individuals observed. In a study of social talkativeness in young children which used a five-minute sample subdivided into five-second intervals, the distributions of the frequencies per sample for the individual children were found to be uniformly asymmetrical, for the most part J-type distributions. This finding prompted an intensive exploration of the data to determine whether the usual measures of average tendency and variability (means and standard deviations) adequately represented these irregular distributions. The results of various empirical tests indicate that the usual formulas are applicable to distributions of this sort except in cases where the behavior has been inadequately sampled.
No takes yet. Share an insight, caveat, or question.
Ruth E. Arrington (1937) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: