Complete Statistical Evidence Calculator

2×2 Table Calculator

Enter your trial data to calculate complete statistical evidence (significance, fragility, and robustness).

A clinical trial comparing a new treatment to standard care:

• Group A (Treatment): 15 events, 85 non-events
• Group B (Control): 25 events, 75 non-events

Enter: a=15, b=85, c=25, d=75

Why Complete Evidence?

Traditional reporting provides only p-values and confidence intervals — this is partial evidence. Complete statistical evidence requires three complementary dimensions:

• Significance (p): Compatibility with null hypothesis
• Fragility (fr): Classification stability (GFQ, or MFQ when the trial is too large for the exact GFI search)
• Robustness (nb): Distance from neutrality (RQ, with NDI as its optional count form)

Only when all three dimensions align do you have truly convincing, replication-ready evidence. Clinical decisions also need the absolute effect size, such as the absolute risk reduction or number needed to treat: the triplet plus effect size is complete evidence.

The calculator also reports resampling fragility — the probability that the significance classification would reverse in a new sample of the same size under a stated replication model — as its own output category. Resampling fragility is a form of data fragility, but because it depends on a replication model it sits outside the model-free p–fr–nb triplet.