Stat 2000, Section 001, Homework Assignment 2 (Due 9/17/2001 11:59pm)
- 0) Reading: Section 1.2
- 1) Please work on the following textbook exercises in Moore/McCabe:
- Exercise 1.25 (1 Point), 1.29 (1 Point), 1.43 (3 Points)
- 2) As promised, here are the colored versions of the micromaps
we discussed in class on 9/10:
the HAP data,
the CO2 data, and
the wheat data.
You should be able to detect additional information in these
colored versions that is hard to see in the grayscale handout
provided in class.
Many more designs and examples for micromaps exist. Several articles
on micromaps have been published in the Statistical Computing and
Statistical Graphics Newsletter by Carr et al., e.g., in
Vol. 7, No. 3, pp. 16-23, and Vol. 9, No. 1, pp. 24-32.
Actually, the micromaps with the CO2 data and the wheat data
are Figures 1 and 2 of the Carr et al. article in Vol. 9, No. 1.
The Statistical Computing and Statistical Graphics Newsletter
is accessible at
http://cm.bell-labs.com/cm/ms/who/cocteau/newsletter/index.html
and provides access to complete issues as pdf and postscript files.
It is informative and easy to read.
Here are three more micromaps for
Iowa,
Michigan, and
Pennsylvania.
Select one of these three maps and carefully
describe and interpret it (about one half page up to one page).
Use all your geographical knowledge (e.g., where are big cities,
sparsely populated rural areas, what is happening in
counties just outside the visible area) and technical
knowledge (where do benzene and lead originate from? -
note that the data for Michigan is an aggregation of
148 HAPs with lead and benzene as major components).
Also note that the data is from 1990. (6 Points)
- 3) This is an example from Mark Monmonier's book
"How to Lie with Maps". The 4 graphics (maps) show the
effect of selecting different class widths and starting
points when visualizing data in a geographic context. (8 Points)
- a) (i) If you were the governor of New York, which graphic would
you use to demonstrate how advanced your state is?
(ii) If you were the governor of Connecticut, which graphic would
you use to demonstrate how advanced your state is?
(iii) If you were the governor of New Jersey, which graphic would
you use to demonstrate how advanced your state is?
(iv) If you were the governor of Virginia, which graphic would
you use to demonstrate how far behind your state is and
desperately needs federal funding?
(v) If you were a historician who wants to show how few telephones
were around in 1960, which graphic would you use?
- b) Now find the "true" interval for the following 5 states.
Do this by calculating the intersection of the class intervals used within
the 4 graphics above:
Maine
New York
Connecticut
New Jersey
Virginia
- c) Describe your findings from b) in one or two sentences.
- d) Now think of micromaps. Can you manipulate these as easily
as these maps to express different political opinions? Explain.
Describe how a micromap of this data might look like.
- 4) What is wrong with these three graphics? (6 Points)
Unfortunately, the small print got lost when scanning in these
graphics.
In the upper graphic, the lines under "Fuel Economy..."
read "Set by Congress and supplemented by the Transportation Department.
In Miles per gallon".
In the middle graphic, the text at the 4 markers in the upper right part
reads: "Jan 1, 5% increase", "April 1, 3.809% increase",
"July 1, 2.294% increase", and "Oct. 1, 2.691% increase".
The Dollar amounts given underneath are "$13.34", "$13.84",
"$14.16", and "$14.54". The labels at the bottom under "1979"
read "Jan to March", "April to June", "July to Sept.", and
"Oct to Dec".
In the bottom graphic, it should read "1972 = 100".