The significance level 
in hypothesis testing
is the prescribed probability of rejecting the null hypothesis when that hypothesis is true. It
is therefore the maximum allowed probability of a
type I error under the null model. A test based on
a p-value rejects at level 
when the p-value is
at most 
.
The significance level is chosen before examining the data and is not the probability
that the null hypothesis is true.
Significance Level
See also
Hypothesis Testing, P-Value, Significance TestExplore with Wolfram|Alpha
References
Casella, G. and Berger, R. L. Statistical Inference, 2nd ed. Pacific Grove, CA: Duxbury, 2002.Cite this as:
Weisstein, Eric W. "Significance Level." From MathWorld--A Wolfram Resource. https://mathworld.wolfram.com/SignificanceLevel.html