A Study Of The Research Designs And Statistical Methods Used In Five Internal Journals Of Otorhinolaryngology

Objective: To assess research designs and the frequency of statistical techniques used in five internal otorhinolaryngological journals in the years from 2000 to 2002.Methods: The types of study design and the statistical methods used in all original articles (n=1331) published in these journals in the period were evaluated.Results: A total of 1331 articles were studied, of which 1035(77.8%) were classified as clinical research. For research design, over 64. 0% were classified as the â…£ type of research design, and only3.1% were belonged the â…  and â…¡ types of research design, or prospective designs. The commonly problems were The lower utilization rate or inappropriateness of the using of randomization and blind trial in the treatises published in these journals, and sample sizes were commonly not enough. The utilization rate of the statistical analytic method used in the studied treatises having reached 52.0 % in the three years, the rate of which in clinical research was higher than that in Nonclinical research. The proportion of correct use of statistical analytic method in the five journals were within the period from 48.7% to 72.7% , and with an average rate of 56.5%. The most commonly usedstatistical methods were t tests (37.9%), contingency tables ( 28.2%) and ANOVAs (14.3%). The most common errors were presentation of P values without specifying the test used, use oft tests instead of ANOVAs in the comparison among three and more groups, and use of unpaired t tests when paired tests were required.Conclusion: The quality of research designs used in five internal otorhinolaryngological journals is relatively needed being improved. Both the utilization rate of the statistical analytic method and the proportion using appropriate methods were in rather highly level, but the error or inappropriate use remains a serious problem.