This video demonstrates how to perform a Bonferroni correction and the Holm method (post hoc tests) after a one-way ANOVA using Microsoft Excel. The Bonferroni correction is a conservative post hoc test (controls Type I Error rate) and increase the probability of a Type II error too much for some applications. The Holm method is an alternative to the Bonferroni correction. The Holm method increases the Type I error probability and decreases the Type II error probability. In this example, three groups are compared using a one-way AVOVA using the Data Analysis tools in Excel. In these tools, the one-way ANOVA is referred to as “ANOVA: Single Factor.” The results of this ANOVA are interpreted. The formulas to adjust the alpha levels for both the Bonferroni correction and the Holm method are demonstrated. An independent t test is performed for each pairwise comparison. For both methods, the alpha is adjusted from the level used for the ANOVA (in this example it is .05) based on the number of levels, and in the case of the Holm method, the rank of each significance level. The probability value (p value) from each t test is compared to the adjusted alpha values. If the p value is less than the alpha value, the null hypothesis is rejected, and it is assumed there is a difference between the groups. If the p value is greater than the alpha value, we fail to reject the null hypothesis, and no difference between the groups is assumed.
20 авг 2024