Power and sample size

Type I error

This diagram shows the distribution of the test statistic (T) when the null hypothesis is true. Any value of T is plausible but the curve indicates that values close to zero are more likely to occur.

But even when the null hypothesis is true, some values of T will be further from zero than the cut off point corresponding to the significance level. When this occurs the researcher will reject the null hypothesis and make a type I error. The probability of making a type I error is equal to , corresponding to the significance level. The most common significance level is 5% but sometimes 1% is used.