Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
D2906-97 – Standard Test Method Technical Guide
📜 Historical Background and Scope of D2906-97
The development of this practice began in August 1966, with the first recommendations issued as the ASTMD-13 White Paper in December 1968. It was subsequently published as Practice D 2906 – 70 T and refined through revisions in 1973, 1974, and 1984, when the term “accuracy” was replaced with “bias” to align with ASTM style guidelines. The scope (Section 1.1) frames D2906-97 as a guide for using information from Practice D2904 or other statistical techniques to prepare precision and bias statements for textile test methods under Committee D-13. The standard emphasizes that at least a single-operator precision statement be included in any new test method, with a complete statement expected at the next reapproval.
📊 Categorization of Test Results for Precision Statements
Per Section 1.3, the instructions in this practice apply to test methods where test results are derived from specific data types. The table below summarizes the four categories covered:
| 🟦 Type | 📏 Description | 📐 Basis of Test Result |
|---|---|---|
| 1. Variables | Measurement of physical properties via continuous scales | Arithmetic average of individual observations |
| 2. Success/Failure | Number of successes or failures in a specified number of observations | Count of successes or failures |
| 3. Defects/Incidents | Number of defects or incidents counted in a specified interval or amount of material | Count of occurrences |
| 4. Attributes | Presence or absence of a characteristic (go/no-go test) | Attribute test result |
⚙️ Key Recommendations for Precision Statements
Preparing valid precision and bias statements requires a general knowledge of statistical principles, including the use of components of variance estimated from an analysis of variance (ANOVA). Practice D2904 provides detailed instructions for these calculations. For methods where precision has not yet been established, an interim statement based on single-operator precision is expected. The practice also allows for extension of these principles to test results based on other data functions, such as standard deviations, with qualified assistance.
💡 Tip: Include a single-operator precision statement in every new or provisional test method. This is the minimum requirement before a complete statement is prepared at first reapproval.
⚠️ Note: The 1984 revision replaced “accuracy” with “bias” throughout the standard. Ensure all statements use “bias” to describe systematic error relative to a reference value.
❓ Frequently Asked Questions
🔍 What is the primary purpose of D2906-97? It serves as a guide for preparing precision and bias statements in textile test methods under ASTM Committee D-13.
💡 Is single-operator precision mandatory? Yes, for any new test method or any method proposed for 5-year reapproval without an existing precision statement.
⚡ Does this practice cover attribute (go/no-go) data? Yes, it includes test results based on the presence or absence of an attribute, as listed in Section 1.3.
📌 How should bias be reported for methods without a true value? Bias can be stated relative to a reference method or accepted reference value, following the 1984 terminology update.