STAT 35000 Name (Print):

Fall 2020
Sample Final
11/21/2020
Time Limit: 120 Minutes Instructor: Honglang Wang

This exam contains 10 pages (including this cover page) and 7 problems. Check to see if any pages
are missing. Enter all requested information on the top of this page, and put your initials on the
top of every page, in case the pages become separated.

You are required to show your work on each problem on this exam. The following rules apply:

• It’s open-book exam and you can refer to all course

Problem Points Score
materials.

• Report decimal answers to three decimal places. 1 35

• You are responsible for upholding IUPUI’s stan- 2 35

dard for academic integrity. This includes pro-
tecting your work from the eyes of other students. 3 30
• Organize your work, in a reasonably neat and
coherent way, in the space provided or your own 4 35
plain papers. Work scattered all over the page
without a clear ordering will receive very little 5 50
credit.
6 20
• Mysterious or unsupported answers will not
receive full credit. A correct answer, unsup- 7 45
ported by calculations, explanation, or algebraic
work will receive no credit; an incorrect answer Total: 250
supported by substantially correct calculations and
explanations might still receive partial credit.

• When you are done, please scan your solution to

a SINGLE PDF file and upload it on canvas
BOTH through ”Sample Final” and ”Sample Fi-
nal 2submission”.

Do not write in the table to the right.

STAT 35000 Sample Final - Page 2 of 10 11/21/2020

1. In 2001 a national vital statistics report indicated that about 3% of all births produced twins.
Is this rate of twin births the same among very young mothers? Data from a large city hospital
found that only 9 sets of twins were born to 500 teenage girls. Let’s perform a hypotheses
testing here with significance level 0.01.
(a) (10 points) Which of the following are the appropriate hypotheses?
A. H0 : p = 0.03 vs. H1 : p > 0.03;
B. H0 : p = 0.03 vs. H1 : p 6= 0.03;
C. H0 : p = 0.018 vs. H1 : p > 0.018;
D. H0 : p = 0.018 vs. H1 : p 6= 0.018;
(b) (10 points) What is the value of the test statistic?
A. -1.5730 B. 1.5730 C. -2.6419 D. 2.6419
(c) (10 points) Which of the following context describes the Type I error correctly?
A. The null assumption is wrong
B. Deciding there is a difference in the twin rate between very young mothers and
in general when there is difference indeed
C. Deciding there is a difference in the twin rate between very young mothers and
in general when there is no difference indeed
D. Deciding there is no difference in the twin rate between very young mothers and
in general when there is difference indeed
E. Deciding there is no difference in the twin rate between very young mothers and
in general when there is no difference indeed
(d) (5 points) What is the probability of making a Type I error?
A. 0.95 B. 0.90 C. 0.05 D. 0.10 E. 0.01
STAT 35000 Sample Final - Page 3 of 10 11/21/2020

2. A company’s old antacid formula provided relief for 70% of people who used it. The company
tests a new formula to see if it is better and gets a P-value of 0.07. Given the significance level
0.05.
(a) (5 points) Which of the following are the appropriate hypotheses?
A. H0 : p = 0.07 vs. H1 : p > 0.07;
B. H0 : p = 0.7 vs. H1 : p > 0.7;
C. H0 : p = 0.07 vs. H1 : p 6= 0.07;
D. H0 : p = 0.05 vs. H1 : p > 0.05;
(b) (10 points) Which conclusion will you make? And what type of error is possible to hap-
pen?
A. We reject H0 at this level; Type I Error
B. We fail to reject H0 at this level; Type II Error
C. We reject H0 at this level; Type II Error
D. We fail to reject H0 at this level; Type I Error
(c) (10 points) What of the following is the correct interpretation of the P-value?
A. There is a 7% chance that the new formula is better
B. There is a 7% chance that the natural sampling variation could produce the
sample result we observed
C. There is a 7% chance that the natural sampling variation could produce the
sample result we observed if the new formula is the same good as the old one
D. There is a 7% chance that the natural sampling variation could produce the
sample result we observed if the new formula is better than the old one
(d) (10 points) If they change the original set of hypotheses into the new set:
no change in percentage of relief vs. percentage changes
what is the P-value of the new test based on the same observed sample information?
A. 0.06 B. 0.08 C. 0.10 D. 0.12 E. 0.14

Fill your solution to the above multiple choice problems below:

(a) (b) (c) (d)
1

2
STAT 35000 Sample Final - Page 4 of 10 11/21/2020

3. (a) (15 points) When a customer uses a particular movie rental vending machine, there is a
20% chance they walk away without renting a movie, a 40% chance they rent one movie,
a 30% chance they rent 2 movies and a 10% chance they rent 3 movies. If 100 people use
the machine, what is the chance that they rent a total of more than 135 movies? (You
may assume that their choices are all independent and the machine does not run out of
any movies).

(b) (15 points) The University of Utah admits 83% of all first year applicants. If I take a
simple random sample of 400 first year applicants to the University of Utah, what is the
chance that less than 85% of the applicants in my sample are accepted?
STAT 35000 Sample Final - Page 5 of 10 11/21/2020

4. An automated assembly line producing ball bearings should produce bearings with a diameter
of 10mm. Quality control personnel run the line and get a sample of ten bearings. The bearings
are known to follow normal distribution with standard deviation of σ = 0.5mm. The measured
ball bearing diameters are listed below:

bearings <- c(9.396, 9.497, 9.655, 9.578, 9.543,

9.575, 9.563, 9.549, 9.546, 9.489)
mean(bearings)

## [1] 9.5391

(a) (10 points) Construct a 95% CI for the mean diameter of the ball bearings and interpret
what this interval means precisely.

(b) (5 points) Manager is not satisfied with the margin of error, and wants an estimate accu-
rate up to the margin of error at 0.1 mm. Find a sample size n that attains this (while
using a 95% CI).

(c) (15 points) With 5% level of significance, using p value method to test that the mean
diameter of ball bearings is different from the specified 10mm. Please write down the null
and alternative hypotheses and state your conclusion in the context of the problem.

(d) (5 points) Interpret what this p value means precisely.

STAT 35000 Sample Final - Page 6 of 10 11/21/2020

5. Another factory that produces ball bearings is testing its assembly line to see whether the
line produces ball bearings with the specified diameter of 10 mm or whether the line is not
properly calibrated. The managers believe that the diameter of the bearings produced by this
line follows normal distribution with the standard deviation σ = 0.1 mm. They want to use
Z-test that is significant at the α = 0.05 significance level.
(a) (10 points) State the null and alternative hypothesis. And what is the probability of a
Type I error?

(b) (10 points) What is the probability of a Type II error when the ball bearings’ mean
diameter differ from the specified diameter by 0.1 mm and the sample size is n = 10?

(c) (10 points) The managers want to be able to detect a difference of 0.1 mm from the
specified diameter with probability 0.9995. Find a sample size that guarantees this under
our assumptions.
STAT 35000 Sample Final - Page 7 of 10 11/21/2020

(d) (20 points) Using the sample size n = 30, experimenters run the line and produce some
ball bearings. The following sample was observed:
bearings <- c(10.11, 9.858, 10.072, 10.007, 10.158, 9.878, 9.935, 9.787,
9.993, 10.008, 9.927, 9.959, 10.086, 10.001, 9.881, 10.057, 9.913,
9.744, 10.136, 10, 9.988, 10.022, 10.112, 10.013, 9.809, 10.014,
10.036, 9.977, 9.952, 9.963)
mean(bearings)
## [1] 9.979867
sd(bearings)
## [1] 0.1006574
Using the sample standard deviation rather than σ, perform the appropriate statistical
test to decide between H0 and H1 , computing p-value. And state your conclusion in the
context of the problem.
STAT 35000 Sample Final - Page 8 of 10 11/21/2020

6. Recently there have been campaigns encouraging people to save energy by carpooling to work.
Some cities have created ”carpool only” traffic lanes (i. e. only cars with 2 or more passengers
can use these lanes). In order to evaluate the effectiveness of carpool only lanes, toll booth
personnel in one city monitor 2,000 randomly selected cars in 2005 before the carpool lanes
were established and 1500 cars in 2008 after the lanes were established. The results are shown
below, where x1 (x2 ) is the number of cars with 2 or more passengers in the data for 2008
(2005).

• year 2008: n1 = 1500, x1 = 576

• year 2005: n2 = 2000, x2 = 652
(a) (10 points) Calculate a 95% confidence interval for the difference of population proportions
of people carpooling to work in 2008 and 2005.

(b) (10 points) Inteprete the above confidence interval to determine whether the data indicate
that the proportion of cars with carpool riders has increased over this period.
STAT 35000 Sample Final - Page 9 of 10 11/21/2020

7. Stevioside, a natural substance extracted from the leaves of a plant called Stevia rebaudiana
Bertoni has a long history of use as a sweetener for drinks and foods and has recently been
showing up in grocery stores as a low-calorie sweetener. Previous animal and human studies
have indicated that stevioside can reduce blood pressure. In a 2-year, randomized, double-
blind, placebo-controlled study of 168 people with mild hypertension, patients in the treatment
group were given 500 mg of stevioside 3 times daily, while those in the control group were given
a placebo. At the end of the study, the average systolic blood pressure for the 82 people in the
stevioside group was 140 mm Hg with an SD of 6.8 mm Hg, while the average systolic blood
pressure for the 86 people in the placebo group was 150 mm Hg with an SD of 7.0 mm Hg. In
this question, the goal is to test to see whether there is evidence that stevioside reduces blood
pressure.
(a) (5 points) Clearly state the null and alternative hypotheses.

(b) (10 points) Calculate the p-value for the above hypothesis testing.

(c) (5 points) Do you reject the null hypothesis with significance level 5%? Explain why or
why not.

(d) (5 points) Clearly state your conclusion in the context.

STAT 35000 Sample Final - Page 10 of 10 11/21/2020

(e) (20 points) Use confidence interval approach with confidence level 95% to investigate the
goal of the problem.