Fig. 1

Analogy between a diagnostic test with continuous results and statistical inference tests of hypothesis. A Distribution of the density functions (area under each curve is equal to 1) of a biological marker measured in a group of disease-free people (solid curve) and those with a certain disease (dashed curve). The vertical dot-dashed orange line depicts the cut-off value for the marker. The light red-shaded area corresponds to the false-positive results; light blue-shaded region, false-negative results. B Distribution of a statistic (eg, Student’s t) density functions. Let under the null hypothesis (H₀), the mean statistic be zero (solid curve); under the alternative hypothesis (H₁), it is not zero (dashed curve). Given a set cut-off value for the significance threshold of the statistic (eg, the vertical dot-dashed orange line), the light red-shaded region corresponds to type I (α) error; light blude-shaded region, type II (β) error. The curves are drawn based on the results presented in our case study