THE STATISTICAL VALIDATION OF TESTS

THE STATISTICAL BEHAVIOR OF DATA AND THE USE THAT CAN BE MADE OF A KNOWLEDGE OF THIS BEHAVIOR FOR THE VALIDATION OF TESTS IS DISCUSSED AND THE ROLE OF EXPERIMENTAL DESIGN IN OVERCOMING DIFFICULTIES INHERENT IN THE XPERIMENTAL SITUATION IS DESCRIBED. BASIC TO THE EVALUATION OF TEST METHODS IS AN UNDERSTANDING OF THE BAHAVIOR OF MEASUREMENTS, AND PARTICULARLY OF THE DISTURBANCES CAUSED BY RANDOM AND SYSTEMATIC EXPERIMENTAL ERRORS. THE CONCEPT OF STATISTICAL CONTROL IS DISCUSSED IN TERMS OF MEASURING PROCESSES. FINALLY, THE ROLE OF STATISTICAL MODELS AND DESIGNS IS ILLUSTRATED BY AN EXAMPLE DEALING WITH CERTAIN PHYSICAL CHARACTERISTICS OF A BLOOD EXPANDER. STATISTICAL ASPECTS OF A MEASURING PROCESS, STATISTICAL POPULATIONS OF MEASUREMENTS, SYSTEMATIC EFFECTS, REPEATABILITY AND REPRODUCIBILITY, QUALITY CONTROL OF MEASUREMENTS AND STATISTICAL DESIGN ARE DETAILED.

• Supplemental Notes:
  • PP 8-13,62
• Authors:
  • Mandel, J
• Publication Date: 1972-9

Media Info

• Serial:
  • Materials Research and Standards
  • Publisher: American Society for Testing and Materials (ASTM)

Subject/Index Terms

• TRT Terms: Design; Experiments; Mathematical models; Measurement; Statistical analysis; Testing; Validation
• Uncontrolled Terms: Models
• Subject Areas: Data and Information Technology; Design; Geotechnology; Highways; Materials;

Filing Info

• Accession Number: 00217287
• Record Type: Publication
• Files: TRIS
• Created Date: Sep 17 1973 12:00AM