统计报告中的问题:基础科学类期刊与临床实践类期刊的比较

The Internet Journal of Epidemiology Pub Date : 2008-12-31 DOI:10.5580/bba

Jaykaran, P. Yadav, P. Bhardwaj, J. Goyal

{"title":"统计报告中的问题:基础科学类期刊与临床实践类期刊的比较","authors":"Jaykaran, P. Yadav, P. Bhardwaj, J. Goyal","doi":"10.5580/bba","DOIUrl":null,"url":null,"abstract":"IntroductionFor any research paper most important part is method section. Results and conclusions derived from research articles are of no validity if methods are not appropriate. Statistics is very important part of method section. It is observed that many articles published in various journals have one or more statistical faults.MethodsWe appraised research paper published in Indian journal of pharmacology and Indian pediatrics in 2007 and 2008 on the basis of modified checklist.ResultsWe observed that calculation of sample size was not mentioned in any article in Indian journal of pharmacology. It was mentioned in 24% of the article of Indian pediatrics. Confidence interval was mentioned in only one article of Indian journal of pharmacology but it was mentioned in 6 (13%) article of Indian pediatrics. Distribution of data was analyzed in only three percentages of articles in Indian journal of pharmacology and 15% of article in Indian pediatrics. Average sample size for clinical trials was 18.9 in Indian journal of pharmacology whereas in Indian pediatrics it is 66.6. Appropriate statistical tests were used in 91% of the articles in Indian journal of pharmacology whereas 93% in Indian pediatrics. Study protocol was followed in 100% of articles and two tailed tests were used in 100% articles. Parametric tests were used in 89% of articles while nonparametric tests were used in 11% of articles in Indian journal of pharmacology. In Indian pediatrics these values are 84% and 15% respectively. ConclusionWe believe that awareness need to be generate regarding the use of confidence interval, distribution of data, adequate sample size and nonparametric statistics in both of the journals.","PeriodicalId":247354,"journal":{"name":"The Internet Journal of Epidemiology","volume":"83 1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Problems in reporting of statistics: comparison between journal related to basic science with journal related to clinical practice\",\"authors\":\"Jaykaran, P. Yadav, P. Bhardwaj, J. Goyal\",\"doi\":\"10.5580/bba\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"IntroductionFor any research paper most important part is method section. Results and conclusions derived from research articles are of no validity if methods are not appropriate. Statistics is very important part of method section. It is observed that many articles published in various journals have one or more statistical faults.MethodsWe appraised research paper published in Indian journal of pharmacology and Indian pediatrics in 2007 and 2008 on the basis of modified checklist.ResultsWe observed that calculation of sample size was not mentioned in any article in Indian journal of pharmacology. It was mentioned in 24% of the article of Indian pediatrics. Confidence interval was mentioned in only one article of Indian journal of pharmacology but it was mentioned in 6 (13%) article of Indian pediatrics. Distribution of data was analyzed in only three percentages of articles in Indian journal of pharmacology and 15% of article in Indian pediatrics. Average sample size for clinical trials was 18.9 in Indian journal of pharmacology whereas in Indian pediatrics it is 66.6. Appropriate statistical tests were used in 91% of the articles in Indian journal of pharmacology whereas 93% in Indian pediatrics. Study protocol was followed in 100% of articles and two tailed tests were used in 100% articles. Parametric tests were used in 89% of articles while nonparametric tests were used in 11% of articles in Indian journal of pharmacology. In Indian pediatrics these values are 84% and 15% respectively. ConclusionWe believe that awareness need to be generate regarding the use of confidence interval, distribution of data, adequate sample size and nonparametric statistics in both of the journals.\",\"PeriodicalId\":247354,\"journal\":{\"name\":\"The Internet Journal of Epidemiology\",\"volume\":\"83 1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2008-12-31\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"The Internet Journal of Epidemiology\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.5580/bba\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"The Internet Journal of Epidemiology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.5580/bba","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 2

摘要

对于任何研究论文来说，最重要的部分是方法部分。如果方法不合适，从研究文章中得出的结果和结论是无效的。统计是方法部分的重要组成部分。可以观察到，发表在各种期刊上的许多文章都有一个或多个统计错误。方法对2007年和2008年分别发表在《印度药理学》和《印度儿科学》上的研究论文进行评价。结果《印度药理学杂志》中未见有关样本量计算的文章。24%的印度儿科学文章提到了这一点。置信区间仅在印度药理学杂志的一篇文章中提到，但在印度儿科学的6篇(13%)文章中提到。仅分析了印度药理学杂志文章的3%和印度儿科学文章的15%的数据分布。印度药理学杂志的临床试验平均样本量为18.9，而印度儿科学杂志的临床试验平均样本量为66.6。《印度药理学杂志》91%的文章使用了适当的统计检验，而《印度儿科学杂志》93%的文章使用了适当的统计检验。100%的文章遵循研究方案，100%的文章采用双尾检验。《印度药理学杂志》89%的文章采用参数检验，11%的文章采用非参数检验。在印度儿科，这些数值分别为84%和15%。结论我们认为，在这两种期刊中，需要对置信区间的使用、数据的分布、足够的样本量和非参数统计产生认识。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Problems in reporting of statistics: comparison between journal related to basic science with journal related to clinical practice

IntroductionFor any research paper most important part is method section. Results and conclusions derived from research articles are of no validity if methods are not appropriate. Statistics is very important part of method section. It is observed that many articles published in various journals have one or more statistical faults.MethodsWe appraised research paper published in Indian journal of pharmacology and Indian pediatrics in 2007 and 2008 on the basis of modified checklist.ResultsWe observed that calculation of sample size was not mentioned in any article in Indian journal of pharmacology. It was mentioned in 24% of the article of Indian pediatrics. Confidence interval was mentioned in only one article of Indian journal of pharmacology but it was mentioned in 6 (13%) article of Indian pediatrics. Distribution of data was analyzed in only three percentages of articles in Indian journal of pharmacology and 15% of article in Indian pediatrics. Average sample size for clinical trials was 18.9 in Indian journal of pharmacology whereas in Indian pediatrics it is 66.6. Appropriate statistical tests were used in 91% of the articles in Indian journal of pharmacology whereas 93% in Indian pediatrics. Study protocol was followed in 100% of articles and two tailed tests were used in 100% articles. Parametric tests were used in 89% of articles while nonparametric tests were used in 11% of articles in Indian journal of pharmacology. In Indian pediatrics these values are 84% and 15% respectively. ConclusionWe believe that awareness need to be generate regarding the use of confidence interval, distribution of data, adequate sample size and nonparametric statistics in both of the journals.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

The Internet Journal of Epidemiology

自引率

0.00%

发文量