Data Scatter and Statistical Considerations

Abstract

If five specimens of the same material, size and surface finish were subjected to the same traction or fatigue test five different results are likely to be obtained. If the test pieces were ten, then ten different results are likely to be obtained. Increasing the number of specimens will not change this general outcome, but will probably yield some new lower or higher value, as well. Therefore, also the spread between the maximum and the minimum value will increase, albeit most values will appear closely-spaced. But the search of the reasonably lower value when not of the lowest possible, which indeed is the target of the designer, cannot be done by just increasing over and over the number of test specimens. The problem of inferring what might be the target value in the fatigue analysis can be solved with the available limited number of data using statistical analysis. This chapter provides the most used and recent statistical tools today available to that purpose.