Tutorials in clinical research: part VII. Understanding comparative statistics (contrast)--part A: general concepts of statistical significance.

摘要

OBJECTIVES/HYPOTHESIS: The present tutorial is the seventh in a series of Tutorials in Clinical Research. The specific purpose of the tutorial (Part A) and its sequel (Part B) is to introduce and explain three commonly used statistical tools for assessing contrast in the comparison between two groups. STUDY DESIGN: Tutorial. METHODS: The authors met weekly for 10 months discussing clinical research studies and the applied statistics. The difficulty was not in the material but in the effort to make the report easy to read and as short as possible. RESULTS: The tutorial is organized into two parts. Part A, which is the present report, focuses on the fundamental concepts of the null hypothesis and comparative statistical significance. The sequel, Part B, discusses the application of three common statistical indexes of contrast, the chi2, Mann-Whitney U, and Student t tests. CONCLUSIONS: Assessing the validity of medical studies requires a working knowledge of research design and statistics; obtaining thisknowledge need not be beyond the ability of the busy surgeon. The authors have tried to construct an accurate, easy-to-read, easy-to-apply, basic introduction to comparing two groups. The long-term goal of the present tutorial and others in the series is to facilitate basic understanding of clinical research, thereby stimulating reading of some of the numerous well-written research design and statistical texts. This knowledge may then be applied to the continuing educational review of the literature and the systematic prospective analysis of individual practices.