Share
Explore BrainMass

# Questions in Statistics

1. Setting the significance level at a very high cutoff (such as 15%) increases chances of what type of error?

2. Why do Type II errors concern scientists?

3. Why have confidence levels of significance been set at 5% and 1%?

4. Failing to reject the null hypothesis when the research hypothesis is true is referred to as what type of error?

5. What are three reasons that it is useful to understand statistical power?

6. By starting with the effect size and the desired level of power, a researcher can work backwards through the power formula to determine what?

7. In the situation where the null hypothesis is true, the distribution from which the sample was taken is the same as what distribution?

8. In statistical notation, p < .05 means that there is less than a 5% chance of making which type of error?

9. When the standard deviation of the original population is small, the experiment tends to have a __________ level of power.

10. Having a very small amount of overlap between the experimental and the comparison distributions __________ the power.

#### Solution Preview

1. Setting the significance level at a very high cutoff (such as 15%) increases chances of what type of error?

The confidence level is equal to your chance of a type I error (rejecting the null hypothesis when it's really true), so a high significance level increases your chance of a type I error.

2. Why do Type II errors concern scientists?

A type II error occurs when you fail to reject the null hypothesis when the alternative hypothesis is true. In many scientific studies, you are trying to see if a treatment has an effect on heath. If there is a type II error in your analysis, the alternative hypothesis is true and treatment does affect health, but you do not reject the null hypothesis.

3. Why have confidence levels of significance ...

#### Solution Summary

The 10 questions in this problem set are meant to test your knowledge and comprehension. These questions are on significance levels, type I and type II error, null and alternative hypotheses, and statistical power.

\$2.19