Suggestions and Recommendations on Statistical Analysis and Research Methodology

An inappropriate use of statistical methods may be associated with incorrect results, capable of leading to erroneous clinical practices, which can sometimes be a disadvantage for patients. False positive results will hinder the implemen-tation of future studies that are used to restore the truth, ending up being difficult to be justified. On the other hand, false negative results may stimulate unnecessary research studies and a waste of resources. The problem is to determine what is good and what is bad, what is lofty and what is mediocre, what is probable and what is improbable. This is where statistics intervenes; if rigorously applied, it yields a precise value to the probability of occurrence of the studied effect.The Romanian Review of Laboratory Med-icine uses a working mechanism to monitor the statistical accuracy of scientific articles submit-ted to the editorial office [1]. Choosing the method of statistical analysis depends on the purpose and objectives of the study, the types of pursued variables and com-parisons that are likely to be made. A well con-ducted statistical analysis should provide the op -portunity to be firstly verified by reviewers, in order to decide whether the best statistical soft-ware respectively the best statistical tests were decided upon so as not to mislead the reader.There are situations identified in many pa-pers evaluated from a statistical point of view, when the statistical software and the statistical tests applied in statistical processing are not mentioned, or there is no reference concerning the threshold of statistical significance compared to which the results of inferential statistics are interpreted. Usually, in statistics the statistical significance threshold of 0.05 is chosen, which means that a risk of 5/100 chances was decided in order to reject the null hypothesis, even if it is true. The section from The Romanian Review of Laboratory Medicine that describes the policies of the journal and the instructions for authors contains aspects that must be observed strictly by the statistical analyses of data in studies [2-4].For submitted papers to be published, it is compulsory to mention in the