Mean | Use enough decimal places to give either the SD to two significant digits,7 or the SE to one significant digit | 3320 g |
3.32 kg |
Percentage | Integers, or one decimal place for values under 10%. Values over 90% may need one decimal place if their complement is informative. Use two or more decimal places only if the range of values is less than 0.1% | 0.1% |
5.3% |
27% |
89% |
99.6% |
Mean difference | Use enough decimal places to give the SE to one or two significant digits. For a standardised mean difference use one or two decimal places | |
Regression coefficient | As with the mean difference. | |
Correlation coefficient | One or two decimal places, or more when very close to ±1 | 0.03 |
−0.7 |
0.89 |
0.999 |
Risk ratio | Round to two significant digits if the leading non-zero digit is four or more, otherwise round to three (the rule of four11). Alternatively use one/two significant digits rather than two/three. For ORs very close to 1 (eg, in logistic regression with a continuous variable) use three decimal places or else report the log OR×100 as the percentage odds to one decimal place13 | 0.0321 |
0.062 |
0.76 |
1.05 |
4.2 |
11.3 |
55 |
1.042 |
4.1% |
SD | One or two significant digits7 | 570 g |
0.57 kg |
9 mm Hg |
2.5 mL |
SE | One or two significant digits | |
CI | Use the same rule as for the corresponding effect size (be it mean, percentage, mean difference, regression coefficient, correlation coefficient or risk ratio), perhaps with one less significant digit | |
Test statistics: t, F, χ2, etc | Up to one decimal place and up to two significant digits | |
t=−1.3 |
F=11 |
χs=4.1 |
p value | Round up to one significant digit, within the limits shown in the examples. The lower limit may be smaller than 0.001, but never 0.000. For genome-wide association studies use the power of 10 format | >0.9 |
0.4 |
0.1 |
0.08 |
0.05 |
0.003 |
<0.001 |
6.10−9 |