This paper presents a theoretical discussion about implications of statistical analysis performed by a specialist, who overlooks the hypothesis test about the normality of the collected data and accepted processes that do not really have the ability to meet the proposed goals. The discussion focuses on the analysis of the results of desertion rates research carried out between 2000–I and 2009–II at the Universidad Tecnologica de Pereira. The paper proposes a direct analogy between industrial and social processes in terms of validating the ability of the process and explains how to troubleshoot statistical normality by a selection of a transformation or distribution wish presents the best fit of analyzed process data.