SAMPLING TIME AND SAMPLING ERROR IN PEDESTRIAN COUNTS

Records of pedestrian flow per minute taken over a period of time contain secular trends and random variations. Because average figures are calculated from samples containing irregular fluctuations, all estimates of mean flow rates are subject to error, the magnitude of which varies with sampling time. To isolate the effects due solely to sampling differences, the trends are removed from flow data referring to a two-hour spell in mid-morning at 11 locations. Sampling theory is used to estimate the maximum errors expected in samples of any length from one minute to two hours for different levels of pedestrian activity. Increasing the sampling time from one minute produces initially large reductions in error, which are progressively lessened. A 15-minute sampling period, for example, used to estimate a mean flow rate per minute is likely to produce maximum errors in the range 8 to 14 per cent for moderately busy streets. (A) /TRRL/

  • Availability:
  • Corporate Authors:

    Printerhall Limited

    29 Newmart Street
    London W1P 3PE,   England 
  • Authors:
    • HAYNES, R
  • Publication Date: 1977-2

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00153926
  • Record Type: Publication
  • Source Agency: Transport and Road Research Laboratory (TRRL)
  • Report/Paper Numbers: Analytic
  • Files: ITRD, TRIS
  • Created Date: Sep 20 1977 12:00AM