Premium
Communicating numerical results correctly
Author(s) -
KIVELÄ T
Publication year - 2010
Publication title -
acta ophthalmologica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.534
H-Index - 87
eISSN - 1755-3768
pISSN - 1755-375X
DOI - 10.1111/j.1755-3768.2010.1342.x
Subject(s) - decimal , nonparametric statistics , statistics , test (biology) , value (mathematics) , standard deviation , confidence interval , parametric statistics , section (typography) , statistical hypothesis testing , computer science , range (aeronautics) , mathematics , arithmetic , paleontology , materials science , composite material , biology , operating system
Purpose To highlight some pearls and pitfalls in reporting numerical data related to the eye with special reference to the instructions of the EVER Journal, Acta Ophthalmologica. Methods Personal experience of the author as a writer, reviewer and editorial board member is used to highlight common issues specific for reporting statistics. The rules and the reasoning behind them are explained using abundant real life examples. Results Statistics and mathematical analyses applied should be described in the methods section, but the software used need not be referenced unless it is specific for the test used. The following rules regarding reporting are highly recommended: 1. Report proportions if the number of subjects is smaller than 10; percentages in integers if the number of subjects is less than 100; if the number of subjects is larger, one decimal place can be given but is seldom necessary. 2. Report summary statistics of normally distributed variables as mean with standard deviation; other variables as medium with range. 3. Use parametric and nonparametric statistical tests accordingly. 4. Give exact p‐values (e.g. p=0.034); if p‐value is less than 0.001, report p<0.001. 5. Give 95% confidence intervals for main findings. 6. Mention the statistical test used with the p‐value (e.g. p=0.015, paired t‐test) if multiple methods are used. Conclusion Efficient reporting and analysis of data related to the eye is not difficult as long as one is aware of the basic rules. Putting them into action helps to ensure that the conclusions the readers draw from your study are proper.