Skip to content

Errata in ISO 11200

This is the entry of the errata registry on ISO 11200:2014. The registry’s introduction says how each claim is checked against the printed page and what the status of an entry means.

ISO 11200:2014, Annex B (the two case studies compute the same standard deviation two different ways)

Section titled “ISO 11200:2014, Annex B (the two case studies compute the same standard deviation two different ways)”
  • Location: Annex B, Table B.1 on printed folio 27 (PDF page 33) and Table B.3 on printed folio 30 (PDF page 36). Both tables carry a row labelled identically, “Standard deviation of the three values measured, ”.

  • The print: Table B.1 lists the three readings 94,5 dB; 94,3 dB; 93,8 dB and gives dB. Table B.3 lists 79,0 dB; 80,2 dB; 82,9 dB and gives dB.

  • The problem: the two use different estimators. Equation (C.1), printed identically in ISO 11201:2010, ISO 11202:2010 and ISO 11204:2010, is the sample standard deviation,

    With the first triple gives 0.3606 dB, which rounds to 0.4, not the 0.3 the table prints; the second gives 1.9975 dB, which rounds to the 2.0 the table prints. With the first gives 0.2944 → 0.3, the printed value, and the second 1.6310 → 1.6, which is not printed. Table B.1 therefore divides by and Table B.3 by , in the same annex, under the same label, for the same quantity.

  • Consequence: it is not cosmetic, because the value propagates. Table B.1 goes on to print dB and dB from dB. With the 0.4 dB that Equation (C.1) gives, dB and dB, the coverage factor applying to the unrounded total rather than to the decibel it is reported as. A reader reproducing the example from the equations does not obtain the uncertainty the example publishes.

  • The likely mechanism: three readings is the smallest sample the equation admits, and it is exactly where the two divisors differ most: is a 22 % gap. A spreadsheet’s population-standard-deviation function reaches for by default, and at three points the slip is large enough to change the rounded decibel.

  • Evidence: both tables read from the printed page, not from extracted text. Verified on PDF pages 33 and 36 (printed pp. 27 and 30) of ISO 11200:2014, against Equation (C.1) on PDF page 32 (printed p. 26) of ISO 11201:2010.

  • Library behaviour: operating_standard_deviation implements Equation (C.1) as printed, with . It reproduces Table B.3 and deliberately does not reproduce Table B.1’s 0.3 dB; tests/emission/test_workstation.py pins both halves so the choice cannot drift.

  • Status: unreported.

Edit page

Published , updated