The Proper Interpretation of Significance Tests in Risk Assessment

  • Stephen Spielman
Part of the Environmental Science Research book series (ESRH, volume 21)


In experimental and epidemiological assessments of potential carcinogens, two complementary techniques of data analysis are commonly employed. Data are cranked through an appropriate test of significance to determine whether or not the substance or factor is risk increasing. If the data are statistically significant, the magnitude of the risk to humans is estimated. Such estimates are highly controversial if the study involves extrapolation from data of a large dose animal study. In those cases where the substance tested is a food additive, the main burden falls on the first pattern of analysis under the so-called Delaney clause of the Food, Drug and Cosmetic Act. This is especially true when the substance is weakly, if at all, carcinogenic. In such cases, no clear dose response curve is exhibited by the data, and significance tests play a crucial role.


Security Level Relative Success Posterior Odds Prior Odds Failure Ratio 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Neyman, J. and Pearson, E.S.: On the problem of most efficient tests of statistical hypotheses. Philosophical Transactions of the Royal Society A, CCXXXI: 289–337 (1933)CrossRefGoogle Scholar
  2. 2.
    Neyman, J: “Inductive behavior” as a basic concept of philosophy of science. Ruvue. Inst. Int. de Stat. 25: 7–22 (1957)CrossRefGoogle Scholar
  3. 3.
    Neyman, J.: First Course in Probability and Statistics. Holt, Rinehart and Winston, Inc., New York (1950)Google Scholar
  4. 4.
    Fleiss, J.L.: Statistical Methods for Rates and Proportions. John Wiley and Sons, New York, Chapt. (1973)Google Scholar
  5. 5.
    Fisher, R.A.: Statistical Methods and Scientific Inference, 2nd Edition. Oliver and Boyd, Edinburgh, p. 43 (1959)Google Scholar
  6. 6.
    Berkson, J.: Experience with tests of significance: a reply to Professor R.A. Fisher. J. Amer. Stat. Assn. 38: 242–246 (1943)Google Scholar
  7. 7.
    Neyman, J.: Lectures and Conferences on Mathematical Statistics and Probability. U.S. Dept. of Agriculture, Washington (1952)Google Scholar
  8. 8.
    Fisher, R.A.: Design of Experiments, 8th Edition. Hafner, New York, p. 14 (1966)Google Scholar
  9. 9.
    Cancer Testing Technology and Saccharin, OTA, Congress of the United States (1977)Google Scholar
  10. 10.
    Hoel, D.: Low dose and species-to-species extrapolation for chemically induced carcinogenesis. In Banbury Report 1, Ed. by V. McElheny and S. Abrahamson. Cold Spring Harbor Laboratory (1979)Google Scholar

Copyright information

© Plenum Press, New York 1981

Authors and Affiliations

  • Stephen Spielman
    • 1
  1. 1.Department of Philosophy Lehman College and the Graduate Center of the CityUniversity of New YorkBronxUSA

Personalised recommendations