Sajjad Bahariniya1, Farzan Madadizadeh2. 1. Health Services Management, School of Public Health, Shahid Sadoughi University of Medical Sciences, Yazd, Iran. 2. Center for Healthcare Data Modeling, Departments of Biostatistics and Epidemiology, School of Public Health, Shahid Sadoughi University of Medical Sciences, Yazd, Iran.
Abstract
BACKGROUND: Nowadays, statistical methods are used frequently in research articles. This review study aimed to determine the statistical methods used in original articles published in Iranian journal of public health (IJPH). METHODS: Original articles in the period 2015 to 2019 from volumes 44 to 48 and numbers 1 to 12 were reviewed by a 3-member committee consisting of a statistician and two health researchers. The statistical methods, sample size, study design and population, type of used software were investigated. Multiple response analysis (MRA), Kruskal-Wallis test and Spearman correlation coefficient were used to data analysis. All analyzes were performed in SPSS21 software. Significant level was set at 0.05. RESULTS: Statistical population in most of the articles were related to human samples at the field level (36% and 297 articles). 66.6% (549 articles) had the sample size less than 500 cases. Study design in most of them were analytical observational 56.2% (464 cases). Acceptance period was 115.5 ± 52.27 days. All the mentioned variables had no significant relationship with the acceptance period (P>0.05). Both among the total tests and the articles, the highest rate of use of statistical methods was related to descriptive statistical method (34.4%, 75.8% and 532 articles), also, the highest use of tests was related to chi square test and t-test( (29%(450 articles)). CONCLUSION: Study design in most of the articles were analytical, to increase thematic diversity, accepting different articles seems necessary. The statistical tests, which used in most articles, were simple, so accepting articles with advanced statistical methods is recommended.
BACKGROUND: Nowadays, statistical methods are used frequently in research articles. This review study aimed to determine the statistical methods used in original articles published in Iranian journal of public health (IJPH). METHODS: Original articles in the period 2015 to 2019 from volumes 44 to 48 and numbers 1 to 12 were reviewed by a 3-member committee consisting of a statistician and two health researchers. The statistical methods, sample size, study design and population, type of used software were investigated. Multiple response analysis (MRA), Kruskal-Wallis test and Spearman correlation coefficient were used to data analysis. All analyzes were performed in SPSS21 software. Significant level was set at 0.05. RESULTS: Statistical population in most of the articles were related to human samples at the field level (36% and 297 articles). 66.6% (549 articles) had the sample size less than 500 cases. Study design in most of them were analytical observational 56.2% (464 cases). Acceptance period was 115.5 ± 52.27 days. All the mentioned variables had no significant relationship with the acceptance period (P>0.05). Both among the total tests and the articles, the highest rate of use of statistical methods was related to descriptive statistical method (34.4%, 75.8% and 532 articles), also, the highest use of tests was related to chi square test and t-test( (29%(450 articles)). CONCLUSION: Study design in most of the articles were analytical, to increase thematic diversity, accepting different articles seems necessary. The statistical tests, which used in most articles, were simple, so accepting articles with advanced statistical methods is recommended.
Authors: Charles D Scales; Regina D Norris; Glenn M Preminger; Johannes Vieweg; Bercedis L Peterson; Philipp Dahm Journal: J Urol Date: 2008-08-16 Impact factor: 7.450
Authors: Nick R Parsons; Charlotte L Price; Richard Hiskens; Juul Achten; Matthew L Costa Journal: BMC Med Res Methodol Date: 2012-04-25 Impact factor: 4.615