Statistical Problems In Manuscripts Submitted To Chinese Journal Of Laboratory

In order to improve the quality of research papers, it is very important to study the statistics problems in the research papers. Most of the previous reports are focusing on the published papers, which cannot completely reflect the actual statistics problems faced by the authors while they are preparing the manuscripts. In this thesis, the submitted manuscripts are studied and analyzed, and the types of statistical errors, the reason and the possible solutions are discussed in the view of an editor.3690 submitted manuscripts are selected from"Chinese Journal of Laboratory Diagnosis", which are received from 2007 to 2009. 2043 manuscripts have the contents of statistics, including 897 papers, which are correct in statistics, and 1146 manuscripts, 56 % of the 2043 manuscripts, which have at least one error in statistics.In this thesis, the reasons of statistics problems in the research papers and how to resolve the problem are discussed in the following three aspects:1. Realizing the important of statistics problems in the medical related research papers. The authors should keep in mind that statistics is unavoidable during medical research and preparing research papers. The peer reviewers and editors should have the knowledge of statistics and pay attention to statistics problems. The readers can also play important roles in the evaluation, improvement and supervision of the published paper's statistics. 2. Improving the educational courses and selecting course of statistics. The teachers should emphasize the actual research paper problems during the statistics course, starting from the"prototype"problem to teach the statistics. The medical journal office should improve the editor'knowledge of statistics, as well as encourage the authors and readers to contribute to the statistics education.3. Strengthening the requirement of the medical journal on the statistics. There are not only clear and specific statistics requirement in the journal's instruction for authors, but also the demands on the statistics in the daily manuscript treating process.I hope that the statistics errors in the submitted manuscript will diminish and the quality of the papers will improve under the endeavor of the authors, editors and readers.