Premium
Pitfalls in statistics and how to avoid them ‐ part 1
Author(s) -
KIVELÄ T
Publication year - 2009
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.1755-3768.2009.1231.x
Subject(s) - categorical variable , computer science , statistical hypothesis testing , statistical analysis , statistics , test (biology) , action (physics) , data science , psychology , mathematics , paleontology , physics , quantum mechanics , biology
Purpose To highlight some pearls and pitfalls in reporting summary statistics and in using common statistical tests. Methods Personal experience of the author as a writer, reviewer and editorial board member especially for the EVER Journal, Acta Ophthalmologica, is used to highlight some of the most common possibilities for improvement in using basic statistics. Results The items discussed, based on examples from the literature, include: 1. how to avoid giving a false impression of precision when reporting key results and how to communicate effect size; 2. how to differentiate statistical from clinical significance and how to communicate statistically nonsignificant results when writing the manuscript; 3. recognising the common problem of making multiple comparisons and adjusting for them; and 4. differentiating ordered and unordered categorical data and choosing the proper statistical test accordingly. Conclusion Clear reporting and analysis of basic statistics is not difficult as long as one is aware of a few basic rules, which will be discussed in this talk. Putting them into action may improve your manuscripts.