Exercises

  1. Investigate the effect of skewness on the type I error rate by running simulations with at least 8 different parameter settings. Show your results in a table and write a summary statement of the effect of skewness.

  2. Investigate the effect of homogeneity of variance on the type I error rate by running simulations with at least 8 different parameter settings. In all cases, make the sample sizes for the two groups equal. Show your results in a table and write a summary statement of the effect of homogeneity of variance.

  3. Repeat question 2 except this time make the sample sizes different. Show your results in a table and write a summary statement. When does violating the assumption of homogeneity of variance lead to an inflated type I error rate? When does it lead to a type I error rate that is below the stated level?

  4. Compare the power of a design with sample sizes of 5 and 15 with the power of a design with sample sizes of 10 and 10.

  5. Describe the effects of the sample size and the standard deviation on the type I error rate and on power?