A comparison of work-sampling and time-and-motion techniques for studies in health services research

Health Serv Res. 1993 Dec;28(5):577-97.


Objective: This study compares results and illustrates trade-offs between work-sampling and time-and-motion methodologies.

Data sources: Data are from time-and-motion measurements of a sample of medical residents in two large urban hospitals.

Study design: The study contrasts the precision of work-sampling and time-and-motion techniques using data actually collected using the time-and-motion approach. That data set was used to generate a simulated set of work-sampling data points.

Data collection/extraction methods: Trained observers followed residents during their 24-hour day and recorded the start and end time of each activity performed by the resident. The activities were coded and then grouped into ten major categories. Work-sampling data were derived from the raw time-and-motion data for hourly, half-hourly, and quarter-hourly observations.

Principal findings: The actual time spent on different tasks as assessed by the time-and-motion analysis differed from the percent of time projected by work-sampling. The work-sampling results differed by 20 percent or more of the estimated value for eight of the ten activities. As expected, the standard deviation decreases as work-sampling observations become more frequent.

Conclusions: Findings indicate that the work-sampling approach, as commonly employed, may not provide an acceptably precise approximation of the result that would be obtained by time-and-motion observations.

Publication types

  • Comparative Study

MeSH terms

  • Bias
  • Confidence Intervals
  • Data Collection / methods
  • Health Policy
  • Health Services Research / methods*
  • Hospitals, Urban / organization & administration
  • Humans
  • Internship and Residency / organization & administration*
  • Job Description
  • Medical Staff, Hospital
  • Reproducibility of Results
  • Sampling Studies
  • Time and Motion Studies*
  • United States
  • Work*