Statistical methods and software used in nutrition and dietetics research: A review of the published literature using text mining.
Statistical methods and software used in nutrition and dietetics research: A review of the published literature using text mining.
复制标题
DOI：
10.1111/1747-0080.12678
复制
发表时间：
2021-07
期刊：
影响因子：
--
通讯作者：
Beck EJ
中科院分区：
文献类型：
--
作者：
Coenen A;Batterham MJ;Beck EJ
营养师必须具备统计学知识，才能有效地解释支撑该学科的科学文献。尽管如此，还没有进行研究，客观地确定常见的统计方法和软件包，具体到目前的营养和饮食学文献。本研究的目的是确定在营养和饮食学研究中经常使用的统计方法和软件。使用词袋方法的文本挖掘方法应用于从SCImago Journal和Country Rank门户网站中的“营养与饮食学”主题类别的所有期刊中获得的随机样本，并于2018年发表。编制了一份229个统计术语和19个统计软件包的清单，以确定要挖掘的搜索术语。将纳入文献方法部分的统计信息提取到Microsoft Excel（2016）中进行数据清理。在R（版本3.6.0）和Microsoft Excel（2016）中进行统计分析。纳入了757篇期刊文章。数值描述性统计是最常见的统计方法组，出现在83.2%的文章中（n = 630）。其次是特定假设检验（68.8%，n = 521）、一般假设概念（58.4%，n = 442）、回归（44.4%，n = 336）和ANOVA（30.8%，n = 233）。IBM SPSS统计软件是最常见的统计软件包，在纳入的文章中占41.7%。这些研究结果为教育工作者评估目前的统计课程和开发继续教育短期课程提供了有用的信息。他们也可以作为一个起点，营养师教育自己的典型统计方法，他们可能会遇到。
Dietitians must be statistically literate to effectively interpret the scientific literature underpinning the discipline. Despite this, no study has been conducted that objectively identifies common statistical methods and packages specific to current nutrition and dietetics literature. This study aimed to identify statistical methods and software frequently used in nutrition and dietetics research. A text mining approach using the bag‐of‐words method was applied to a random sample of articles obtained from all journals in the ‘Nutrition and Dietetics’ subject category within the SCImago Journal and Country Rank portal and published in 2018. A list of 229 statistical terms and 19 statistical software packages was developed to define the search terms to be mined. Statistical information from the methods section of included articles was extracted into Microsoft Excel (2016) for data cleaning. Statistical analyses were conducted in R (Version 3.6.0) and Microsoft Excel (2016). Seven hundred and fifty‐seven journal articles were included. Numerical descriptive statistics were the most common statistical method group, appearing in 83.2% of articles (n = 630). This was followed by specific hypothesis tests (68.8%, n = 521), general hypothesis concepts (58.4%, n = 442), regression (44.4%, n = 336), and ANOVA (30.8%, n = 233). IBM SPSS statistics was the most common statistical software package, reported in 41.7% of included articles. These findings provide useful information for educators to evaluate current statistics curricula and develop short courses for continuing education. They may also act as a starting point for dietitians to educate themselves on typical statistical methods they may encounter.
登录
查看更多内容
影响因子：
0.5
作者：
King, Carrie;Byham-Gray, Laura;Roberts, Melissa M.
通讯作者：
Roberts, Melissa M.
DOI：
10.1016/s0002-8223(00)00336-9
发表时间：
2000-10-01
影响因子：
--
作者：
Slawson, DL;Clemens, LH
通讯作者：
Clemens, LH
影响因子：
3.3
作者：
Howard, A. J.;Ferguson, M.;Campbell, K. L.
通讯作者：
Campbell, K. L.
影响因子：
3.3
作者：
Harrison, JA;Brady, AM;Kulinskaya, E
通讯作者：
Kulinskaya, E
影响因子：
3.1
作者：
Allman-Farinelli, Margaret
通讯作者：
Allman-Farinelli, Margaret