ABSTRACT
Good statistical design is a pivotal factor in animal research. However, replication, randomization and blinding, which are key components of good statistical design, are less often used in animal research (Kilkenny et al., 2009). Hess (2011) reviewed statistical design given in 100 articles on animal experiments published in Cancer Research in 2010. In 14 of the 100 articles, the number of animals used per group was not reported. In none of the 100 articles the method used to determine the number of animals per group was reported. Among the 74 articles in which randomization seemed feasible, only 21 reported that they had randomly allocated animals to treatment groups. None of these articles described how the randomization was carried out. Selection of appropriate statistical tools is very crucial in the analysis of data obtained from toxicological and pharmacological studies. Selection of a non-appropriate statistical tool during the design of a study or using a different statistical tool from that mentioned in the study plan with improper justi¿ cation may lead to misinterpretation of the data (Kobayashi et al., 2011).
