This paper presents both parametric and nonparametric procedures for obtaining design allowable values from small sets of material strength data. The allowable represents a material structure design number defined as the 95% lower confidence bound on the specified percentile of the population of material strength data. The percentiles are the first and tenth for the A and B allowables. The proposed methods were evaluated in order to determine an acceptable procedure that reduces the penalties associated withsmall sample allowable computations. This involves accurately maintaining the definition requirements and reducing variability in the estimate. Application of very small samples will obviously reduce costs in testing and manufacturing which is the primary motivation for this study. In the evaluation process, two nonparametric methods were introduced, one a Bayesian and the other a modified Hanson-Koopmans (MHK) application. The methods involved using the Diriichlet prior in the Bayesian case and the log-convex function for MHK. Weibull, normal models, and a pooling process used by the aircraft industry were also applied in the study. Results showed the MHK method was superior when compared to the other methods in determining small sample allowables. The values obtained from the MHK application consistently met the coverage requirement (95% of values less than a specified percentile of the population of all test data) for a relatively wide spectrum of data sets.