STAT303 Sec 502 and 503
Spring 2015
Exam #1
Form C
Instructor: Julie Hagen Carroll
February 18, 2015
Some questions have been adapted from Statistics: The Art and Science of Learning from Data, 3rd ed, by Agresti &
Franklin.
1. Don’t even open this until you are told to do so.
2. Remember to turn your phone off now.
3. Please turn your hats around backwards or take them off.
4. Please put your backpack and other things along the walls or at the front of the room.
5. You need a gray, 81/2 × 11” scantron, pencil, calculator and you may have 5 sheets of notes.
6. There are 20 multiple-choice questions on this exam, each worth 5 points. There is partial credit. Please mark your
answers clearly. Multiple marks will be counted wrong.
7. You will have 60 minutes to finish this exam.
8. If you have questions, please write out what you are thinking on the back of the page so that we can discuss it after
I return it to you.
9. If you are caught cheating or helping someone to cheat on this exam, you both will receive a grade of zero on the
exam. You must work alone.
10. When you are finished please make sure you have filled in your name and marked your FORM (A, B, C or D) and
20 answers, then turn in JUST your scantron.
11. Good luck!
1
STAT303 sec 502 and 503
Exam #1, Form C
Spring 2015
1. Joe is in STAT303501. He made an 82 on his exam
4. When are bar charts preferred over pie charts?
that had an average of 75 with a standard deviation of
A. when there are a lot of categories, so there would
6. Jane is in STAT303502 and made a 79. Her class
be too many ’slices’ in the pie
average was 68 with a standard deviation of 8. Who
B.
when we want to compare variables, like voting by
did better relative to the rest of their class?
gender
A. Joe because 82 is better than 79.
C. when the proportions don’t add up to 100%
B. Joe because his class had a better average.
D. all of the above
C. Jane because her z-score is higher.
E. Never, either graph can be used for plotting cateD. Jane because she has more points above her class
gorical data?
mean (11 vs. 7).
5. When a polling organization asked, “Should laws be
E. We can’t compare them because we don’t know
passed to eliminate all possibilities of special interests
the shape of either distribution of grades.
giving huge sums of money to candidates?” 80% of the
sample answered yes. When they posed the question,
“Should laws be passed to prohibit interest groups from
contributing to campaigns, or do groups have a right to
contribute to the candidate they support?” only 40%
said yes. This is an illustration of problems caused by:
A.
B.
C.
D.
E.
lurking variables
undercoverage
voluntary response bias
response bias
nonresponse
6. If the 5-Number-Summary for a dataset is: 10, 39, 46,
57, 78, what is the shape?
A.
B.
C.
D.
E.
2. Which of the following is true about the graph above
showing the number of siblings? The numbers on x-axis
are the values of the bin to the right.
A. The median number of siblings is 1 since it’s the
most common value.
B. If the child with 8 siblings really had 10, the mean
and standard deviation of this dataset would increase.
C. The distribution is skewed to the left.
D. All of the above are true.
E. Two of the above are true.
uniform
normal
normal with at least one outlier
skewed left
skewed right with at least one outlier
7. Which of the following is true?
A. Sample surveys provided no useful information
since they are not random.
B. No inferences can be made from sample surveys
since they are not random.
C. Randomized experiments can never be done in
medical settings since it’s unethical to ‘make someone sick’.
D. Observational studies can never prove causation.
E. All of the above statements are false.
3. Suppose I multiply all data above by 5, which of the
following would be true?
A. Each value in the 5-Number-Summary would be 5
times larger.
B. All measures of spread would be 5 times larger.
C. The mean and standard deviation of the set of
z-scores (the distribution of z-scores) would be 5
times larger.
D. All of the above would be true.
E. Only two of the above would be true.
2
STAT303 sec 502 and 503
Exam #1, Form C
Spring 2015
8. The stacked bar chart above represents the number of
R
different colored Skittles
in five bags. Which of the
following is true?
11. What can be said based on the graph above which
shows protein and fiber content for some breakfast
foods?
A. There is a definite lack of consistancy of the proportion of different colors within the bags.
B. There is no association between the bags and the
number of Skittles of each color per bag. They all
contain the same colors.
C. The proportions of colors are different so there is
no association.
D. Two of the above are true.
E. None of the above is true.
A. To get more protein, you need more fiber.
B. Any of the food with 6g of fiber has more protein
than any of the lesser (fibers).
C. Because there are two quantitative variables, we
need a scatterplot. This graph is inappropriate.
D. Two of the above are correct.
E. None of the above is correct.
12. Statistically significant means
A. there is DEFINITELY something going on in the
data (the significance is important).
B. there is a strong relationship between the two variables.
C. one variable actually caused the other variable to
vary.
D. the researcher conducted a randomized experiment.
E. the data was properly randomized.
9. Which of the following is/are true?
A. Uniform data can never have outliers.
B. The only type of data that can have outliers
on both ends (extremely large points AND small
points) is normal.
C. If the data is skewed left, the outliers are on the
left.
D. All of the above are true.
E. Exactly two of the above are true.
13. A nine-year-old genius appeared on The David Letterman Show. Dave asked him what his IQ was, and he
replied that it was “off the bell curve”. For his IQ to
be an outlier about how large must it be? NOTE: Assume IQ’s are normally distributed with mean = 100
and standard deviation = 15.
10. Which of the following is true?
A. We can only calculate z-scores for normal data.
B. A z-score of -1 for a test score means you only
missed 1 question.
C. A z-score of 10 is impossible for normal data.
D. Scale changes affect z-scores, but shift changes
don’t.
E. None of the above are true.
A.
B.
C.
D.
E.
3
over 130
over 115
over 100
over 150
more information is needed
STAT303 sec 502 and 503
Exam #1, Form C
Spring 2015
14. A random sample of Americans were asked how many 17. Suppose we have a bell-shaped distribution with mean,
hours they slept per night. The standard deviation was
x̄ = 14, and standard deviation, sx = 2. The approxi2.2. Interpret this standard deviation in context.
mate 95th percentile is
A.
B.
C.
D.
E.
A. The typical difference between the number of
hours a person slept and the average is about 2.2.
B. About 2.2% of Americans had a number of hours
of sleep that was more than 3 standard deviations
above or below average.
C. The difference between the third and first quartiles
of the number of hours a person slept is about 2.2.
D. The average number of hours Americans slept is
about 2.2.
E. The difference between the largest amount of sleep
in the sample and the smallest is 2.2.
95
16
17
18
19
Consistantly?
Almost Always
Always
Sometimes
Total
5
1
4
1
6
10
0
1
2
3
15
5
8
1
14
20
20
49
6
75
Total
26
62
10
98
18. “How long do you wash your hands?” “How often?” The
table above represents the results of a survey of hospital
staff who were asked these questions about their habits
after leaving a patient’s room. How likely is someone
who washes their hands for only 5 seconds to Always
wash them?
A.
B.
C.
D.
E.
5/98
4/98
4/62
4/6
4/6 ∗ 4/62
19. Using the same table, how likely is someone to only
wash their hands Sometimes?
A.
B.
C.
D.
E.
15. How strong is the correlation in the graph above?
A.
B.
C.
D.
E.
strongly negative
moderately negative
weak
moderately positive
strongly positive
1/10
10/98
6/98 ∗ 10/98
6/10
yuck!
20. In one newspaper article, the author discussed having met someone who didn’t actually enjoy the taste
of sugar in any form, candy or otherwise. Then the
author asked readers to write in to talk about whether
they liked sugar. A surprising 40% of 1200 readers who
responded said that they also didn’t like sugar. Why is
it not safe to infer that 40% of the population doesn’t
like sugar?
16. Which of the following is true about bivariate, (x, y)
data?
A. Changing the scale of y changes the value of the
slope but not the intercept.
B. If we move from x to x + 10, y moves 10 ∗ b1 .
C. Adding a point at (x̄, ȳ) doesn’t change the correlation, the slope or the intercept.
D. All of the above are true.
E. Two of the above are true.
A. This was only a single study. We should not use
the results of one survey to make inferences to the
general population.
B. Only 1200 people wrote in. In order to infer something about all Americans, they need a bigger sample size.
C. Perhaps those who also don’t like sugar were more
likely to write in than those who do.
D. People who read this article are apparently less
likely to like sugar than the general population.
E. People who read the newspaper might be wealthier
than the general population.
1C,2B,3E,4D,5D,6C,7E,8A,9E,10C,11E,
12B,13D,14A,15C,16E,17C,18D,19B,20C
4