STATISTICS (Answer the Following questions)
1. Researchers often choose to reject the null hypothesis when there is less than a 5% likelihood of observing a given pattern in the data if the null hypothesis is true. They select 5% because
a. It’s a rule handed down by Isaac Newton
b. It’s the established scientific norm
c. It’s the absolute best way of establishing the truth
2. What’s the trade-off between certainty and confidence intervals?
a. If you want to be more certain of your estimate, your confidence interval will be smaller
b. If you want to be more certain of your estimate, there must not be a confidence interval
c. If you want to be more certain of your estimate, your confidence interval will be larger
3. Which of these polls might suffer from selection bias?
a. A survey of Rutgers students that randomly selects students from the Dean’s registry of all students
b. A survey of Rutgers Sociology students that randomly selects students enrolled in a Sociology course this year
c. A survey of Rutgers students that randomly selects fans attending a Rutgers football game
4. Response rates allow us to assess:
a. A poll’s validity
b. A poll’s reliability
c. A poll’s importance
5. If you’re calculating the standard error of a proportion, it will be largest when that proportion is:
a. 0.04
b. 0.46
c. 0.23
6. The null hypothesis is false, but you fail to reject it. This is:
a. A Type 1 error
b. A Type 2 error
c. An innocent mistake
7. The null hypothesis is true, yet you reject it in favor of the alternative hypothesis. This is:
a. A Type 1 error
b. A Type 2 error
c. An innocent mistake
8. When constructing a confidence interval, you switch from a confidence level of 95% to a confidence level of 90%. Your confidence interval will:
a. Get smaller (more narrow)
b. Get larger (more wide)
c. Not change
9. When you switch from a 95% confidence level to a 90% confidence level, what else happens:
a. Your risk of committing a Type 1 error increases
b. Your risk of committing a Type 2 error increases
10. Let’s say that the null hypothesis appears to be incorrect, but you just don’t have enough evidence to reject it and don’t want to risk a Type 1 error, what’s one plausible (good) course of action:
a. Just fudge the numbers
b. Lower the confidence level
c. Collect more data
11. When your sample is below ____ observations, you should check closely for outliers.
a. 15
b. 30
c. 45
d. 100
12. The t-distribution, relative to the normal distribution, has tails that are:
a. fatter
b. narrower
c. cooler
13. The degrees of freedom in the t-distribution determines
a. the total probability under the curve
b. the size of your sample
c. the contours of the bell shape
14. With a sample of 35 observations, what degrees of freedom would we have?
15. As the degrees of freedom increase, the t-distribution…
a. becomes indistinguishable from the normal distribution
b. gets wider
c. stops working
16. If your sample size is 10 and you want to calculate a 90% confidence interval, what is your critical t-value?
17.If your sample size is 20 and you want to calculate a 95% confidence interval, what is your critical t-value?
18. If your sample size is 30 and you want to calculate a 99% confidence interval, what is your critical t-value?
19. The average height of 507 adults is 171.1 cm with a standard deviation of 9.4 cm. What is the standard error?
20. If you wanted to create a 95% confidence interval for the mean from the previous question, what would the critical t-value be?