Momentary and Partial Interval Time Sampling: A Reply to Harrop, Daniels and Foulkes (1990)

Abstract

Momentary time sampling (MTS) is described as an unbiased method of estimating time spent engaging in target behaviour. Conclusions that MTS with long sampling intervals or for short sessions results in large error are based on studies which use too few observations (a too-small sample size) to expect small sampling error. Partial interval (PI) sampling is a biased method, and the degree of bias changes as the characteristics of the behaviour change, a particular problem for intervention studies.