KGS Home List of Computer Contributions

Kansas Geological Survey, Computer Contributions 41, originally published in 1970


FORTRAN IV Program for Sample Normality Tests

by D.A. Preston

Shell Development Company

small image of the cover of the book; goldenrod paper with brown text.

Originally published in 1970 as Kansas Geological Survey Computer Contributions 40.

Introduction

The improvement of prediction is among the most important goals of any scientific investigation. Yet the results of an investigation will be subject to considerable uncertainty if the supporting data are drawn from a population whose characteristics are incompletely understood. This limitation is particularly common for geologic investigations where populations are, in general, only partially accessible so that their exact nature cannot be determined. Useful numerical approximations of their probability distributions may be derived, however, if the data are drawn in such a manner that they constitute a representative sample of the parent population (see Griffiths, 1967; Griffiths and Ondrick, 1968). Statistical analysis of such data then may improve the predictability of further sampling by suggesting an appropriate a priori probability model for the parent population from which predictive inferences can be drawn. Moreover, once an appropriate model is established for a population, the quality of any sample drawn from that population may be determined easily.

A distinction must be preserved for possible purposes for which samples are statistically analyzed. On one hand, if samples are from a population whose probabilistic nature is known, statistical analysis yields a measure of sample error created by departure from randomness, operator error, and the like. The main problem is identifying the sources of error. On the other hand, if samples are analyzed in order to derive a population distribution model, the sample must be constructed painstakingly to minimize sample error. Poorly constructed samples from populations of unknown distribution will yield completely ambiguous statistical results.

The complete text of this report is available as an Adobe Acrobat PDF file.

Read the PDF version (8.4 MB)


Kansas Geological Survey
Placed on web Sept. 11, 2019; originally published 1970.
Comments to webadmin@kgs.ku.edu
The URL for this page is http://www.kgs.ku.edu/Publications/Bulletins/CC/41/index.html