Sunday, February 8, 2009

practice data

The following data set gives the values for the following variables.

Column #1: Abbreviation of state name
Column #2: Number of electoral votes in the state
Column #3: Political party that won that state's electoral vote in 2008
Column #4: Political party that won that state's electoral vote in 2004
Column #5: Political party that won that state's electoral vote in 2000

CA 55 D D D
TX 34 R R R
NY 31 D D D
FL 27 D R R
IL 21 D D D
PA 21 D D D
OH 20 D R R
MI 17 D D D
NJ 15 D D D
NC 15 D R R
GA 15 R R R
VA 13 D R R
MA 12 D D D
WA 11 D D D
IN 11 D R R
MO 11 R R R
TN 11 R R R
MD 10 D D D
MN 10 D D D
WI 10 D D D
AZ 10 R R R
CO _9 D R R
AL _9 R R R
LA _9 R R R
KY _8 R R R
SC _8 R R R
CT _7 D D D
OR _7 D D D
IA _7 D R D
OK _7 R R R
AR _6 R R R
KS _6 R R R
MS _6 R R R
NM _5 D R D
NV _5 D R R
NE _5 R R R
UT _5 R R R
WV _5 R R R
HI _4 D D D
ME _4 D D D
RI _4 D D D
NH _4 D D R
ID _4 R R R
DC _3 D D D
DE _3 D D D
VT _3 D D D
AK _3 R R R
MT _3 R R R
ND _3 R R R
SD _3 R R R
WY _3 R R R

Answer the following questions.
1) Give the five number summary for the number of electoral votes. Also find the IQR, the high and low thresholds for outliers and note what states, if any, have an outlying number of electoral votes.

2) Find the relative frequencies for groups of states with the following values. Use the +/- .1% rule to decide how far to round the numbers.

a) 3 D's in the last three columns
b) 2 D's and 1 R in the last three columns
c) 1 D and 2 R's in the last three columns
d) 3 R's in the last three columns

3) Given the groups of states in part 2), instead of counting the number of states, count instead the number of electors in each group and give the frequency and relative frequency for the elector totals (Note: n = 538 for the total number of electors. To make the problem easier, we will ignore the fact that Maine and Nebraska do not use the winner take all system).

Answers in the comments. The box and whiskers for problem #1 is as follows.

1 comment:

Prof. Hubbard said...

1)
Five number summary

High: 55
Q_3: 12
Q_2: 8
Q_1: 4
Low: 3

IQR = 8

Q_1 - 1.5*IQR = -4

No low outliers.

Q_3 + 1.5*IQR = 24

This means, CA, TX, NY and FL are outliers. When drawing the box and whiskers, the high whisker would extend to 21, noting IL and PA as the states with the most number of electors that are not outlyingly high.

2)
3 D's, 0 R's: 19/51 = 37%
2 D's, 1 R: 3/51 = 6%
1 D, 2 R's: 7/51 = 14%
0 D's, 3 R's: 22/51 = 43%

(Note: If you round to one place after the decimal, the total is 100.2%, which means nearest percent passes the rule but nearest tenth of a percent rounding doesn't.)

3)
3 D's, 0 R's: 248/538 = 46%
2 D's, 1 R: 16/538 = 3%
1 D, 2 R's: 100/538 = 19%
0 D's, 3 R's: 174/538 = 32%