Tiffani Priebe

9/8/16
3
Name_________________________________________Date____________________
Period_____________
Module 1 EDA Final Wrap-up and Review Due Friday 9 Sep 2016
You may use handheld calculations, graphing calculator or StatCrunch (through MyStatLab)
Regardless, however, all answers must be EXPLAINED and justified, including the multiple choice responses.
MULTIPLE CHOICE. Choose the one alternative that best completes the statement or answers the question.
1) The distribution below is the number of family members reported by 25 people in the 2010 Census.

D
The best description for the shape of this distribution is 1) _______
A) normal
B) approximately normal
C) bimodal
D) skewed right
E) skewed left

The graph is skewed to the right because the bulk of the data is towards the lowest values
and the tail is towards the higher values.

2) Which of these variables is most likely to follow a Normal model for U.S. adults?
2) _______
B
A) ACT scores
The normal U.S. adults are in the middle class so they will have about the same income maybe
B) income
a little over or above the average income.
C) monthly mortgage
D) eye color
E) commuting time

3) Which is true of the data whose distribution is shown?

I. The distribution is skewed to the right.

II. The mean is probably smaller than the median.
III. We should summarize with median and IQR.
A) I only
B) II only
C) I and II
D) II and III
E) I, II, and III

B
3) _______

The mean will be smaller because the outliers pull the graph the mean to that side of the graph, which
means that the mean would probably be around the 7 bar coming from the left side. Also the median
would be the 8 bar from the left.

SHORT ANSWER. Write the word or phrase that best completes each statement or answers the question.
4) Embryonic stem cells A Pew Research survey asked Americans their feelings on medical use of embryonic stem cells.
Say they surveyed 340 people and got the results summarized in the table.
a) 58 moderates said that it was morally acceptable about the medical use on embryonic cells. To get the percentage I did 58/149 got .3892
and multiplied it by 100*.3892 and got 38%.
b) To find the frequency i found the percentage of each answer of the conservatives. 36 conservatives said it was morally wrong and 36/99*100
equals 36%; 33 conservatives said it was not morally wrong and 33/99*100 equals 33%; 30 conservatives said it was morally acceptable and
30/99*100 equals 30%.
c) I would use a histogram because the histogram could show me the frequency in the data.
d) Yes their is evidence of political affiliation and feelings toward medical embryonic cells. Conservatives believe in traditional values and because
of this it probably made the percentages larger (33%) than the moderates (19%) who are strictly about political values and liberals (15%) who
are basically mixed feelings.

a. What percent of the moderates said it is morally acceptable?

b. What is the conditional relative frequency distribution of belief for conservatives?
c. If you wanted to show the association between political affiliation and feelings toward medical use of embryonic stem
cells, what kind of graph would you make? (Just name it.)
d. Is there evidence of an association between political affiliation and feelings toward medical use of embryonic stem
cells? Explain briefly. 4) _____________

5) Auto insurance The Insurance Institute for Highway Safety publishes ratings for all models of vehicles to compare the
relative risk of payouts. 100 is the mean rating for all vehicles. A rating of 122 means the vehicle is 22% worse than
average. The table shows the summary statistics for the collision ratings of 27 midsize cars.
a) To find the outliers I did Q3-Q1 to get IQR which equals 23; I then used the IQR rule to find the outliers, which
is Q1-(IQR*1.5) and Q3+(IQR*1.5) which is 99-(23*1.5) equals 64.5 and 122+(23*1.5) is 156.5. So anything outisde
(64.5, 156.5) would be an outlier. So according to my math their are outliers.
b) It would be more appropriate to use the median and the IQR because the outliers would pull the mean and
standard deivation into a larger or smaller number.

a. Were any of the ratings outliers? Show how you made your decision.
b. A histogram of the data is shown. Is it more appropriate to use the mean and standard deviation, or the median and
IQR to describe these data? Explain.

5) _____________

6) Soft drinks A restaurant owner wanted to improve the efficiency of his employees. One way he tried to do this was to
buy a machine that will automatically dispense 16 oz. of soda into a glass rather than have the employee hold the button
on the dispenser. The actual amount dispensed by the machine can be represented by the model X~N (16.2, 0.3)
a. Draw and clearly label the model.

15.3 15.6 15.9 16.2 16.5 16.8 17.1

-3 -2 -1
0
+1 +2 +3
b. The sales representative who sold him the machine said, 95% of the glasses you fill with soda will fall between
15.6
16.8
_________
and _________.
Fill in the blanks based on the normal model, then comment on this claim.
c. What is the 3rd quartile of amounts dispensed?
d. If a glass will actually hold 16.7 oz. of soda, what percent of the time would you expect the glass to overflow?
e. The manufacturer wants to reduce the overflow rate to only 1%. Assuming the mean amount dispensed will stay the
same, what standard deviation must they achieve?
f. Briefly explain what that change in standard deviation means in this context.
g. A competing manufacturer says that not only will 98% of their glasses be safe from overflowing, but 70% will have
more than 16 oz., reducing customer complaints. What Normal model parameters is that manufacturer claiming? Show
your work.
6) _____________
c) To find Q3 (quartile 3) I did 2nd vars, invNorm, and then (.75, 16.2, 0.3) and got 16.4 which means that the 75th percentile is around 16.4 oz.
d) To get the the percentage I did 2nd vars, normalcdf, and then (16.7, 1000, 16.2, 0.3) and got .0478 which means that approximately 4% of
the soda will overflow.
e) So the current overflow rate is 4% to bring this down I plugged in different numbers that was lower than the current standard dviation, and
my steps were 2nd vars, normalcdf (16.7, 1000, 16.2, ___); I plugged in .2, .28, .25, and .22; and .22 got to .011. So the company has drop the
standard deivation to .22 to get 1%.
f) To change the standard deviation you need to change the cluster around the mean. So to decrease the standard deviation have the cluster
closer to the mean, and to have a larger standard deviation have the cluster farther away.
g) The company is saying that the standard deivation is around .25 because I did 2nd vars, normalcdf, and then (-1000, 16.7, 16.2, .25) and
got approximately 98%. The mean would be the same as the other company. So the normal model parameters would be N(16.2, .25).