CASES OF STATISTICAL ANALYSIS IN MATERIALS SCIENCE: DIFFICULTIES AND SOLUTIONS

Experimental characterization of new materials requires an effective application of statistical experimental design concepts and techniques. Ideally, these techniques will be considered from the planning stage. In reality, however, the decision to consider analysis techniques is often made when a project has already a certain degree of advance. This situation imposes limitations to what statistical analysis can accomplish. In this work, three case studies with varying degrees of advance at the moment of introducing statistical analysis are discussed in terms of its limitations as well as the possibilities to improve the experiment's capabilities.