Abstract
Background: Statistical methods (SM) are a ubiquitous tool in research. This study aimed to review SM used in original article published in the Archives of Iranian Medicine (AIM) and assess their effect on article acceptance period.
Methods: The original articles published in the period 2015–2019 from volumes 18 to 22 and issues 1 to 12 of the AIM were reviewed and six items such as SM, study design, statistical population, sample size, software and acceptance period were extracted. Mean (SD), frequency (percentage) and multiple response analysis (MRA) were used for description. The Kruskal-Wallis test and Spearman correlation coefficient were used for data analysis in SPSS 26 with significance level at 5%.
Results: During the study period, 423 original articles were reviewed. The statistical population in most of them was patients (38.8% and 164 articles), and most studies (51.5% and 218 articles) had a sample size of less than 500 people. The study design in most of the articles was analytical-observational (55.1% and 233 articles), and 79.7% (337 articles) used SPSS for data analysis. The median (IQR) acceptance period was 194 (134.25). MRA results showed that the highest rate of use of SM was related to descriptive statistics (277 articles, 30.3%) and Chi square test (130 articles, 14.2%). In the last two years, the acceptance period had a declining trend. There was no significant relation between mentioned variables and acceptance period (P>0.05).
Conclusion: Contrary to the researchers’ misconceptions, the acceptance period was not affected by SM, study design, statistical population, sample size, or type of software.