DIN ISO 5479 2004-01 PDF

Full title and description

STB DIN ISO 5479:2004-01 — Statistical interpretation of data — Tests for departure from the normal distribution. This entry is the German national adoption (DIN) published in January 2004 of ISO 5479 (Edition 1997); it provides graphical and numerical procedures, tables and probability‑plot (probability‑net) tools to help decide whether a sample departs from the normal distribution.

Abstract

This standard gives guidance on methods and tests to determine whether the hypothesis that observations are drawn from a normal distribution should be rejected, assuming observations are independent. It includes both graphical procedures (probability plots/probability‑net) and computational goodness‑of‑fit tests together with tables and examples for practical interpretation. (ISO original: 1997; DIN adoption: 2004).

General information

• Status: DIN adoption — Withdrawn (national publication withdrawn); ISO original (ISO 5479:1997) remains published/confirmed.
• Publication date: DIN adoption: 2004-01; ISO original edition: 1997-05.
• Publisher: Deutsches Institut für Normung (DIN) — for the 2004 national publication; International Organization for Standardization (ISO) — original.
• ICS / categories: 03.120.30 (Statistical methods) and 17.020 (Metrology and measurement) — statistical interpretation and metrology contexts.
• Edition / version: DIN ISO 5479:2004-01 (national adoption of ISO 5479:1997, Edition 1).
• Number of pages: 75 (DIN adopted/translated document as offered by national sales channels); ISO original edition lists 33 pages.

Scope

The standard provides recommended procedures and tests (graphical and numerical) for assessing departures from normality in a dataset where observations are assumed independent. It is intended to support the statistical evaluation of data prior to applying methods that assume normality, by giving decision‑rules, example computations and reference tables for critical values.

Key topics and requirements

• Graphical methods: normal probability plots / probability‑net and guidance on visual assessment.
• Numerical goodness‑of‑fit tests: presentation of commonly used tests and criteria to decide rejection of normality (guidance for application and interpretation).
• Reference tables and critical values to support computation and decision making.
• Worked examples and recommendations on test selection depending on sample size and suspected type of departure (tails, skewness, kurtosis).
• Advice on assumptions (independence of observations) and limitations of normality testing.

Typical use and users

Applied statisticians, quality engineers, metrologists and researchers who need to verify the normality assumption before performing parametric inference (ANOVA, t‑tests, capability analysis, regression diagnostics). Also used by standardization bodies and testing laboratories preparing reports that require documented checks for normality.

Related standards

ISO 5479 is part of statistical guidance used alongside other ISO/TC 69 documents on statistical methods and metrology standards. National adoptions (DIN and others) reproduce the ISO content with national forewords; users commonly consult other standards and guidance on capability analysis and test methods when deciding on further analysis steps.

Keywords

normality test, normal distribution, probability plot, goodness‑of‑fit, statistical interpretation, ISO 5479, DIN adoption, probability‑net, reference tables.

FAQ

Q: What is this standard?

A: An ISO standard (ISO 5479:1997) giving methods and tests to decide whether sample data depart from the normal distribution; the entry STB DIN ISO 5479:2004-01 refers to a 2004 German national adoption of that ISO text.

Q: What does it cover?

A: It covers both graphical procedures (probability plots/probability‑net) and computational goodness‑of‑fit tests, includes tables of values, worked examples and guidance on interpretation and decision criteria for rejecting normality.

Q: Who typically uses it?

A: Statisticians, quality and metrology professionals, laboratories and researchers who must check the normality assumption before applying parametric statistical methods.

Q: Is it current or superseded?

A: The DIN national publication (DIN ISO 5479:2004-01) is marked withdrawn by national sellers; the underlying ISO standard (ISO 5479:1997) has been subject to periodic review and was confirmed as current by ISO in recent reviews (ISO record shows the 1997 edition remains published/confirmed). Users should check their national body for current availability or replacement.

Q: Is it part of a series?

A: It is a standalone ISO publication within the ISO/TC 69 body of standards on statistical methods; related ISO documents address broader statistical practice, capability analysis and measurement uncertainty. National adoptions replicate the ISO document as needed.

Q: What are the key keywords?

A: normality test; goodness‑of‑fit; probability plot; probability‑net; statistical interpretation; ISO 5479; normal distribution.