D2904-97 – Standard Test Method Technical Guide

🎯 Purpose and Scope of D2904-97

ASTM D2904-97 (Reapproved 2002), formally the “Standard Practice for Interlaboratory Testing of a Textile Test Method that Produces Normally Distributed Data,” provides a comprehensive framework for designing and executing interlaboratory studies. The primary objectives are to generate the data necessary for calculating the required number of specimens to determine average textile quality (per Practice D 2905) and for developing precise statements on precision and bias (per Practice D 2906).

The standard’s scope explicitly covers three progressive scales of experimentation: Single-laboratory preliminary trials for initial variance estimation, Pilot-scale interlaboratory tests for protocol validation, and Full-scale interlaboratory tests for definitive precision determination. Critical guidance is provided for data transformations prior to analysis, the handling of missing data, and the identification and treatment of outlying observations.

📊 Data Normality and Statistical Methodology

The analytical procedures in D2904-97 are rigorously applicable to test methods measuring continuous variates from normal distributions or from distributions amenable to normalization via mathematical transformation. The core of the analysis is the Analysis of Variance (ANOVA), used to estimate the specific variance components that characterize repeatability and reproducibility.

For test methods that produce data not meeting these criteria, the standard provides explicit direction. Users with continuous but non-normal data, or discrete data (such as ratings on