Hypothesis Testing

Question 1

Define p-value

Answer 1

the p-value is the probability that the null hypothesis is true. If the p-value is smaller than α, then that suggests that the null hypothesis can’t be true, leading us to reject the null. When not given α, we have levels of strength to talk about the p-value (very weak, weak, moderate, strong, very strong) based on the size of p-value.

Question 2

Type II Error

Answer 2

Occurs when we accept Ho, when it is actually false.
Example:
A guilty man goes free.

Question 3

Type I Error

Answer 3

Occurs when we reject Ho, when it is actually true.
Example:
Jail an innocent man.

Question 4

<p>What is the power of a test?</p>

Answer 4

<p>It is the likelihood that you will reject the null hypothesis</p>

Question 5

<p>How do you increase the power of a test?</p>

Answer 5

<ul>
<li>Increase the sample size (n)</li>
<li>Increase the siginificance level (n)</li>
<li>Reduce the amount of variability in the sample</li>
<li>Consider an alternate hypothesis further away from the null hypothesis</li>
</ul>

Question 7

<p>The higher the power of the test the more evidence you have to \_\_\_\_ the null hypothsis</p>

<p>A. Accept</p>

<p>B. Reject</p>

Answer 7

<p>B. Reject</p>

Question 8

<p>When you're writing a conclusion what should ithave?</p>

Answer 8

<ul>
<li>If you are rejecting or accepting the null hypothesis based on the amount of evidence you have</li>
<li>The relationship between the significance level and the p-value</li>
<li>Writing the conclusion in context</li>
</ul>

Question 9

How do you determine expected counts for a chi squared test?

Answer 9

For the variable that you are determining the expected count of, multiply the sum of it’s row by the sum of it’s column and then divide that by the population.

Question 10

Define α

Answer 10

α, or the statistical significance level, gives you a level at which you determine whether you reject the null hypothesis, or don’t have enough evidence to reject the null hypothesis.