Statistics Guidelines

Statistics

(Revised, 7/2017)

JARO recommends that authors obtain the advice of a statistician prior to starting a study, to aid in the study design. Any statistical analysis must be checked for accuracy by the authors; if statistical software is used, the source (including version number) of the tools must be listed in Materials and Methods; this information should be provided in the final paragraph. Authors are strongly encouraged to seek the assistance of a statistician at their institution or elsewhere; if they seek such advice, the resource should be identified in the letter of submission, and in the acknowledgements. • Statistics should be fully reported in the manuscript/article. This includes stating the statistical test(s) used, the exact value of N (sample size), and the definitions of the distribution center (mean, median) and measures of variability (SD, SEM, and confidence intervals) reported in the text and shown in figures. • All legends should include specific “N” for each treatment group and a description (or brief description) of statistics used for each experiment. The definition of a sample should be made clear (N cells, N subjects). If data are pooled from multiple observations on individual subjects, the operations involved in computing the pooled value must be stated. • Statements that indicate the presence of “significant effects” should be supported by an appropriate statistical test. • State the type of test that was used, including the degrees of freedom, the resulting test value, and the exact P-value (to 2 significant figures) that the result occurred at chance under the null hypothesis. For t-tests, always state whether the test is for a one-sided or two-sided hypothesis. • Format: The format of the description of the statistical results should indicate the degrees of freedom, the statistic value, and the P value, as in these examples:

F(3,21) = 5.62, P = 0.0054
t(7) = 4.582, P = 0.0025
r^2(9) = 0.77, P = 0.0004

(where: ^ implies superscript; _ implies subscript).

  • To avoid ambiguities, all statistical variables should be italicized (F, t, r, P).
  • The type of post-hoc tests used when following any ANOVA with multiple comparisons should be identified.
  • Reporting P values with inequalities should be limited to data is grouped in tables or figures, or for post-hoc tests (multiple comparisons) if no exact value is reported by the software.

• Manuscripts that report results based on the analysis of large data sets, including (but not limited to) genomic sequencing studies and fMRI imaging studies are also required to specify in detail how the statistical analyses were done.

  • Although we do not recommend this as a regular practice, if the statistical analysis has been extensive, the results may be placed in a table, and if the table would be quite large and not suitable for a printed page, we will consider an authors’ request to place the table in a Supplemental material section. Such tables must provide all of the same information requested above, and must be appropriately cross-referenced to the text.

links