- A) We can calculate the number of p-values larger than 0.5, and divide them by the number of simulations.
- B) We can calculate the number of p-values in the first bar (which contains all 'significant' p-values from 0.00 to 0.05) and divide the p-values in this bar by the total number of simulations.
- C) We can calculate the difference between p-values above 0.5 minus the p-values below 0.5, and divide this number by the total number of simulations.
- D) We can calculate the difference between p-values above 0.5 minus the p-values below 0.05, and divide this number by the number of simulations.

- A) 55%
- B) 60%
- C) 80%
- D) 95%

- A) The p-value distribution is exactly the same as with 50% power
- B) The p-value distribution is much steeper than with 50% power
- C) The p-value distribution is much flatter than with 50% power
- D) The p-value distribution is much more normally distributed than with 50% power

- A) The p-value distribution is exactly the same as with 50% power
- B) The p-value distribution is much steeper than with 50% power
- C) The p-value distribution is basically completely flat (ignoring some minor variation due to random noise in the simulation)
- D) The p-value distribution is normally distributed

- A) The power (or true positives)
- B) The true negatives
- C) The Type 1 error (or false positives)
- D) The Type 2 error (or false negatives)

- A) ~90%
- B) ~75%
- C) ~50%
- D) ~5%

- A) The effect is significant, and provides strong support for the alternative hypothesis.
- B) The effect is significant, but it is without any doubt a Type 1 error.
- C) With high power, you should use an alpha level that is smaller than 0.05, and therefore, this effect can not be considered significant.
- D) The effect is significant, but the data are more likely under the null hypothesis than under the alternative hypothesis.

- A) At best, p-values between 0.04 and 0.05 are equally likely under the alternative hypothesis, and under the null hypothesis.
- B) At best, p-values between 0.04 and 0.05 are approximately 4 times more likely under the alternative hypothesis, than under the null hypothesis.
- C) At best, p-values between 0.04 and 0.05 are ~10 times more likely under the alternative hypothesis, than under the null hypothesis.
- D) At best, p-values between 0.04 and 0.05 are ~30 times more likely under the alternative hypothesis, than under the null hypothesis.