Skip to main content
Figure 1 | Annals of Intensive Care

Figure 1

From: Do simple screening statistical tools help to detect reporting bias?

Figure 1

Illustration of the checking procedure. A Variable distribution. The variable FFP during surgery is described with a mean of 60 and a SD of 210. As shown in the left panel, if this variable was normally distributed, it should exhibit some negative values. Because negative values are impossible for such a variable, its distribution is necessarily asymmetric (right panel: example of a strictly positive variable characterized by a large SD). B P value distribution under the null hypothesis. Left panel represents the theoretical distribution of the p values under the null hypothesis, that is a uniform (0,1) distribution: p values are equally distributed on both sides of the middle line. As shown in the right panel, the observed p values are likely not to be distributed uniformly. C Distribution of the simulated p values corresponding to the comparison of two variables in the two groups across the 10,000 simulated datasets. The left panel shows a very high probability for the comparison of the urine output at 5 hours to be statistically significant, whereas it was reported as nonsignificant by the authors. The left panel shows the distribution of the simulated p values concerning the comparison of the PRBC during surgery: the black vertical line represents the 0.05 threshold of statistical significance, and the dashed line represents the p value that has been explicitly computed, given the observed mean and SD in the two groups.

Back to article page
\