User Tools

Site Tools


terms

Terminology

Significant

The authors describe that they “hypothesized that there is a significant difference between A and B”. I assume that the word significant here refers to clinically significant, not statistically significant because the latter depends on study-specific conditions, such as sample size, related to study design rather than the studied disorder. Please clarify what characterizes a clinically significant difference.

Independent t-tests

The statistics section describes that the statistical testing has been made using “independent t-tests”. Several such tests have, however, been developed, e.g. Student’s, Satterthwaite’s, Welch's, Prien’s, Yuen’s, and Pitman’s t-tests, and even to Hotelling’s T-test. Which one was used in this study?

Univariate – multivariate

The term multivariate is used incorrectly, see e.g. Hidalgo B, Goodman M. Multivariate or multivariable regression? Am J Public Health 2013;103:1-3.

Peters TJ. Multifarious terminology: multivariable or multivariate? univariable or univariate? Paediatric and Perinatal Epidemiology 2008;22:506.

Quartiles

Four quartiles are presented. According to the statistical terminology (see The International Statistical Institute. The Oxford Dictionary of Statistical Terms. Oxford University Press, New York 2003) only 3 quartiles exist, the middle one also being the median.

Quintiles

The authors refer to 5 quintiles, but only 4 can be defined. See Altman DG, Bland JM. Statistics Notes: Quartiles, quintiles, centiles, and other quantiles. BMJ 1994;309:996.

terms.txt · Last modified: 2020/02/16 18:51 by ranstam