A 5% significance level means that there is a 5% chance that your conclusion could be incorrect. At the same time though, you’ll reach a conclusion less often. It means that you are more confident that a 10% significance level, but less confident than a 1% significance level (Lind, Marchal, & Wathen, 2011). A lower significance level raises the type II error, because there are more values that would fail to reject the null hypothesis (Lind, Marchal, & Wathen, 2011). If the hypothesis test is performed before the researcher has decided on the significance level, it is possible that an error may occur. This error is known as a Type I error. This happens when a true null hypothesis is being rejected, in lieu of a false hypothesis. Before performing any part of a test, the researcher decides on a significance level. These levels are 0.5%, 1%, 2.5%, 5% and 10%
Lind, D. A., Marchal, W. G., & Wathen, S. A. (2011). Basic statistics for business and economics (7th ed.). New York, NY: McGraw-Hill/Irwin.