Results Interpretation Playground
Explore a hypothetical effect, its uncertainty, and a meaningful-difference threshold.
Your entries stay in this page’s memory and are not submitted to us. Downloads are generated locally. Reloading or leaving the page clears your entries. Do not enter patient information.
Hypothetical example: a normally distributed estimate of a difference, in arbitrary outcome units, with a supplied standard error. Change the uncertainty without changing the estimated effect.
Hypothetical result
95% interval: 0.040 to 3.960. Two-sided normal-approximation p-value: 0.0455.
The interval excludes zero under this model; this alone does not establish clinical importance. The interval is not wholly beyond the selected magnitude threshold.
Purple: effect and confidence interval. Dashed gray: zero. Dotted gold: ± your meaningful-difference threshold.
What this illustrates
The interval is estimate ± 1.959964 × SE. The p-value uses a standard normal null distribution and tests zero difference. These calculations are inappropriate for odds ratios entered on their raw scale, small-sample t inference, dependent observations without an appropriate SE, or estimates with a non-normal sampling distribution. A confidence interval is not a 95% probability statement about this fixed parameter.