Assignment:
Response in a True/False:
1 Based on a classical approach, the probability of an event is defined as the number of favorable outcomes divided by the total number of possible outcomes.
2 One characteristic of the F distribution is that F cannot be negative.
3 If the null hypothesis states that there is no difference between the mean net income of retail stores in Chicago and New York City, then the test is two-tailed.
4 Hypothesis testing is a procedure that uses sample evidence and probability theory to decide whether to reject or fail to reject a hypothesis.
5 The probability of rolling a 3 or 2 on a single die is an example of mutually exclusive events.
6 If the computed value of F is 0.99 and the critical value is 3.89, we would not reject the null hypothesis.
7 If we are testing for the difference between two population means, it is assumed that the sample observations from one population are independent of the sample observations from the other population.
8 As sample sizes decrease, we are more confident in sample estimates of population parameters.
9 A probability of 0.0 is assigned to an event that is certain not to occur.
10 To apply the special rule of addition, the events must be independent.
11 As sample sizes decrease, rejecting the null hypotheses is less likely.
12 An alternate hypothesis is a statement about a population parameter that is accepted when the null hypothesis is rejected.
13 Type II error is the probability or risk of rejecting a null hypothesis when it is actually true.
14 If the null hypothesis is true and the researchers do not reject it, then a correct decision has been made.
15 The fifth and final step in testing a hypothesis is taking a sample and, based on the decision rule, deciding if the null hypothesis should be rejected.
16 You can forecast benefit costs with 100% accuracy using time series regression analysis.
17 A standard distribution has more than one peak.
18 You can only forecast benefit costs within a certain amount of confidence interval using time series regression analysis.
19 Aren't you glad this class is over.
20 A standard distribution looks like a bell.
Multiple Choice -
1 Which of the following is NOT one of the five steps in the hypothesis testing procedure?
A. Formulate a decision rule
B. State the null and alternate hypotheses
C. Select a level for β
D. Identify the test statistic
2 What is a Type II error?
A. Accepting a false null hypothesis
B. Rejecting a false null hypothesis
C. Accepting a false alternate hypothesis
D. Rejecting a false alternate hypothesis
3 What is the level of significance?
A. Probability of a Type II error
B. Probability of a Type I error
C. z-value of 1.96
D. Beta error
4 Administering the same test to a group of 15 students and a second group of 15 students to see which group scores higher is an example of
A. a one sample test of means.
B. a two sample test of means.
C. a paired t-test.
D. a test of proportions.
5 Given the following regression model, which factor do you think is most important in predicting benefit costs?
Benefit costs = 0.9(employee population) +
0.8(claims experience) + 0.4(gender)
A. employee population
B. claims experience
C. gender
D. cannot be determined
6 Your assistant has determined that the regression model that best describes the relationship between performance (x) and salary (y) is:
y = 7.107 + 0.813x
How may this regression model be useful to you?
A. It shows that performance and salary move in opposite directions.
B. It can be used to determine the appropriate salary, given performance.
C. It can be used to determine performance given salary.
D. There is no relationship
7 What is the logarithm in the equation 109 = 1,000,000,000?
A. 10
B. 9
C. 1,000,000,000D. 1
8 In executive compensation, it is important to consider distributions because:
A. Most wage earners receive salaries on the lower end of the scale.
B. Low salaries throw the mean salary level off.
C. No company wants to pay above the market level.
D. None of the above.
9 The information in the data that is not explained by the regression model is called:
A. Hats
B. Estimates
C. Residuals
D. Curves
10 Collinearity occurs when:
A. independent variables are highly correlated with each other
B. all coefficients are large
C. independent variables are highly correlated with the dependent variable
D. None of the Above
11 Given the information below, what is the cost-of-living index of Oregon?
City Income ($)*
National 35,973
Michigan 35,862
Florida 36,985
Oregon 33,044
*The numbers are used for illustration purposes only.
A. 82
B. 92
C. 109
D. 33,044
12 Which statement is correct about the F distribution?
A. Cannot be negative
B. Cannot be positive
C. Is the same as the t distribution
D. Is the same as the z distribution
13 The result of a particular experiment is called a(n)
A. observation.
B. conditional probability.
C. event.
D. outcome.
14 What do we call the statement that determines if the null hypothesis is rejected?
A. Decision rule
B. Test statistic
C. Alternate hypothesis
D. Critical value
15 What is another name for the alternate hypothesis?
A. Null hypothesis
B. Hypothesis of no difference
C. Rejected hypothesis
D. Research hypothesis
16 Which approach to probability assumes that the events are equally likely?
A. Classical
B. Empirical
C. Subjective
D. Mutually exclusive
17 Using the median test only, is there a difference in compensation between the following 2 groups?
Protected Non-protected
1 $20,746 $20,972
2 $21,356 $21,864
3 $22,018 $22,808
4 $22,900 $22,900
5 $24,986 $25,087
6 $25,983 $26,076
7 $26,385 $27,476
A. There is a substantial difference in the medians of the 2 groups.
B. There is a small difference in the medians of the 2 groups.
C. There is no difference in the median value between the 2 groups.
D. Not enough information to make a determination.
18 To set executive pay using regression analysis, what is the first step you must take?
A. Identify the factors used in establishing compensation levels.
B. Find the regression equation.
C. Do a benchmark study.
D. Set you salary ranges.
19 The best way to determine employee benefit preferences is to use:
A. Benchmarking
B. Surveys
C. Chi-square test
D. Regression analysis
20 To conduct a test of hypothesis with a small sample, we need to be able to make an assumption that:
A. a larger computed value of t will be needed to reject the null hypothesis.
B. the region of acceptance will be wider than for large samples.
C. the confidence interval will be wider than for large samples.
D. the population is normally distributed.