# Errata in ISO 11200

> One page of the phonometry documentation, as Markdown. Index: https://docs.jmrp.io/phonometry/llms.txt

Canonical: https://docs.jmrp.io/phonometry/reference/errata/iso-11200/
Language: en
Alternate: https://docs.jmrp.io/phonometry/es/reference/errata/iso-11200/index.md
Version: phonometry 4.0.0
Install: pip install "phonometry[full]"
Author: José Manuel Requena Plens (https://orcid.org/0000-0003-1250-6212)
Published: 2026-10-28
Updated: 2026-11-16T21:36:14Z
License: MIT (https://github.com/jmrplens/phonometry/blob/main/LICENSE)
Summary: Errata in ISO 11200: 1 defect in the printed text, with the evidence and what phonometry does about it.

---

This is the entry of the [errata registry](https://docs.jmrp.io/phonometry/reference/errata/) 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)

- **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, $\sigma_{omc}$".
- **The print:** Table B.1 lists the three readings 94,5 dB; 94,3 dB; 93,8 dB
  and gives $\sigma_{omc} = 0{,}3$ dB. Table B.3 lists 79,0 dB; 80,2 dB;
  82,9 dB and gives $\sigma_{omc} = 2$ 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,

  $$
  \sigma_\mathrm{omc} = \sqrt{\frac{1}{N-1}
  \sum_{j=1}^{N} \left( L'_{p,j} - \overline{L'_p} \right)^2}
  $$

  With $1/(N-1)$ 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 $1/N$ 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 $N$ and Table B.3 by $N-1$, 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 $\sigma_\mathrm{tot} = 1.5$ dB and $U = 2.4$ dB from
  $\sigma_{R0} = 1.5$ dB. With the 0.4 dB that Equation (C.1) gives,
  $\sigma_\mathrm{tot} = \sqrt{1.5^2 + 0.4^2} = 1.552 \to 1.6$ dB and
  $U = 1.6 \times 1.552 = 2.48 \to 2.5$ 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: $\sqrt{3/2}$ is
  a 22 % gap. A spreadsheet's population-standard-deviation function reaches
  for $1/N$ 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`](https://github.com/jmrplens/phonometry/blob/main/src/phonometry/emission/workstation.py)
  implements Equation (C.1) as printed, with $1/(N-1)$. 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.
