|
Homework assignment |
Please complete the
following homework assignment from the textbook and deliver to me (via e-mail,
us mail, fax or by hand) by April 4, 2000. Note: You should attempt all the
problems BUT ONLY TURN IN THE PROBLEMS INDICATED IN RED TO ME. Show your
work when appropriate- lists of answers will not be graded.
|
HW #7: |
p. 421: 1,5,7,8,11,12 p. 432: 5-10 p. 440: 3,4,8,10,11,12 p.454: 7,8,11,12,14,15 p. 463: 3, 4, 6, 7, 8, 16 |
Writing Assignment #7
Choose ONE of the
following to do
1) Does Oreo "Double
Stuff" cookies really contain twice the filling as advertised on the box?
How could we test this hypothesis? First of all we have to decide what
"twice" the filling means. Do we mean the inner filling should be
twice as wide or twice as heavy or??. Well it is probably best to test weight
(what if the Oreo company just pumped in air to make it thicker - would you
feel cheated?) So to do this writing assignment you need an accurate food scale
- actually a metric scale is best (like ones they have in science labs). Here
is what to do
a) Buy some regular Oreo's
and some "Double Stuff" Oreo's. Choose 30 Oreo's randomly from each
pack
b) Scrape (using a knife or
wooden spoon), the filling from each Oreo and weigh it individually. You should
have 30 weights of the filling for your Oreo's. Do the same for the double
stuff Oreo's. After you have all the weights - you may eat the samples!
c) Do a hypothesis test of
two means (independent). Test the hypothesis that the stuffing in the double
stuff is more than (or equal to) twice the stuffing in the regular Oreo's.
Write up your conclusions. Use the 5% significance level
2) Do four cylinder cars
get better gas mileage than six cylinder cars? Most people on the street would
say yes - but..now that you know statistics you can actually see if this is
true. First go to the following Internet site
http://www.amstat.org/publications/jse/archive.htm
Look at the file called
93cars.dat and choose a random sample of 4 cylinder cars and 6 cylinder cars
(you decide on sample sizes). Do a hypothesis test on these two means to
determine your answer. Use the 5% significance level