6 Apportionment and Voting

In this chapter, we discuss two of the most fundamental principles of democracy: the right to
vote and the value of that vote. The U.S. Constitution, in Article I, Section 2, states in part that

The House of Representatives shall be composed of members chosen every second

year by the people of the several states, and the electors in each state shall have the
qualifications requisite for electors of the most numerous branch of the state
legislature. . . . Representatives and direct taxes shall be apportioned among the
several states which may be included within this union, according to their respective
numbers. . . . The actual Enumeration shall be made within three years after the first
meeting of the Congress of the United States, and within every subsequent term of
ten years, in such manner as they shall by law direct. The number of Representatives
shall not exceed one for every thirty thousand, but each state shall have at least one
Representative; . .

This article of the Constitution requires that “Representatives . . . be apportioned among

the several states . . . according to their respective numbers. . . .” That is, the number of
representatives each state sends to Congress should be based on its population. Because
populations change over time, this article also requires that the number of people within a state
should be counted “within every subsequent term of ten years.”

The way representatives are apportioned has been a contentious issue since the founding
of the United States. The first presidential veto was issued by George Washington in 1792
because he did not approve of the way the House of Representatives decided to apportion the
number of representatives each state would have. Ever since that first veto, the issue of how to
apportion membership in the House of Representatives among the states has been revisited
many times.

6.1 Introduction to Apportionment

The mathematical investigation into apportionment, which is a method of dividing a whole into
various parts, has its roots in the U.S. Constitution. (See the chapter opener.) Since 1790, when
the House of Representatives first attempted to apportion itself, various methods have been
used to decide how many voters would be represented by each member of the House. The two
competing plans in 1790 were put forward by Alexander Hamilton and Thomas Jefferson.

To illustrate how the Hamilton and Jefferson plans were used to calculate the number of
representatives each state should have, we will consider the fictitious country of Andromeda,
with a population of 20,000 and five states. The population of each state is given in the table
below.
 Andromeda’s constitution calls for 25 representatives to be chosen from these states. The
number of representatives is to be apportioned according to the states’ respective populations.

6.1.1 The Hamilton Plan

Under the Hamilton plan, the total population of the country (20,000) is divided by the
number of representatives (25). This gives the number of citizens represented by each
representative. This number is called the standard divisor.

▼ Standard Divisor For

Standard divisor = 𝑁𝑢𝑚𝑏𝑒𝑟𝑇𝑜𝑡𝑎𝑙 𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛

𝑜𝑓 𝑝𝑒𝑜𝑝𝑙𝑒 𝑡𝑜 𝑎𝑝𝑝𝑜𝑟𝑡𝑖𝑜𝑛

Andromeda, we have

𝑇𝑜𝑡𝑎𝑙 𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 20,000

Standard divisor = 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑝𝑒𝑜𝑝𝑙𝑒 𝑡𝑜 𝑎𝑝𝑝𝑜𝑟𝑡𝑖𝑜𝑛 = = 800
25

Question: What is the meaning of the number 800 calculated above?

Now divide the population of each state by the standard divisor and round the quotient
down to a whole number. For example, both 15.1 and 15.9 would be rounded to 15. Each whole
number quotient is called a standard quota.
 ▼ Standard Quota
The standard quota is the whole number part of the quotient of a population
divided by the standard divisor.

Answer: It is the number of citizens represented by each representative.

From the calculations in the above table, the total number of representatives is 22, not
25 as required by Andromeda’s constitution. When this happens, the Hamilton plan calls for
revisiting the calculation of the quotients and assigning an additional representative to the state
with the largest decimal remainder. This process is continued until the number of representatives
equals the number required by the constitution. For Andromeda, we have
Take note:
Additional representatives are
assigned according to the largest
decimal remainders. Because the
sum of the standard quotas came
to only 22 representatives, we
must add three more
representatives. The states with
the three highest decimal
remainders are Cephus
(1.954), Apus (13.904), and Orion
(3.646). Thus each of these states
gets an additional
representative.

6.1.2 The Jefferson Plan

As we saw with the Hamilton plan, dividing by the standard divisor and then rounding
down does not always yield the correct number of representatives. In the previous example, we
were three representatives short. The Jefferson plan attempts to overcome this difficulty by using
a modified standard divisor. This number is chosen, by trial and error, so that the sum of the
standard quotas is equal to the total number of representatives. In a specific apportionment
calculation, there may be more than one number that can serve as the modified standard divisor.
For instance, in the following apportionment calculation shown in the first table on the following
page, we have used 740 as our modified standard divisor. However, 741 can also be used as the
modified standard divisor.
 The table below shows how the results of the Hamilton and Jefferson apportionment
methods differ. Note that each method assigns a different number of representatives to certain
states.

Although we have applied apportionment to allocating representatives to a congress, there are

many other applications of apportionment. For instance, nurses can be assigned to hospitals
according to the number of patients requiring care; police officers can be assigned to precincts
based on the number of reported crimes; math classes can be scheduled based on student
demand for those classes.

Example Problem # 1: Apportioning Board Members Using the Hamilton and Jefferson
Methods
Suppose the 18 members on the board of the Ruben County environmental agency are
selected according to the populations of the fi ve cities in the county, as shown in the table
below.

a. Use the Hamilton method to determine the number of board members each city should
have.
b. Use the Jefferson method to determine the number of board members each city should
have.
Solution
a. First find the total population of the cities.

7020 + 2430 + 1540 + 3720 + 5290 = 20,000

Now calculate the standard divisor.

𝑇𝑜𝑡𝑎𝑙 𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 20,000

Standard divisor = 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑝𝑒𝑜𝑝𝑙𝑒 𝑡𝑜 𝑎𝑝𝑝𝑜𝑟𝑡𝑖𝑜𝑛 = = 1111.11
18

Use the standard divisor to find the standard quota for each city.

The sum of the standard quotas is 16, so we must add 2 more members. The two cities with the
largest decimal remainders are Pacific and Vista. Each of these two cities gets one additional
board member. Thus the composition of the environmental board using the Hamilton method
is Cardiff: 6, Solana: 2, Vista: 2, Pauma: 3, and Pacific: 5.

b. To use the Jefferson method, we must find a modified standard divisor that is less than the
standard divisor we calculated in part a. We must do this by trial and error. For instance, if
we choose 925 as the modified standard divisor, we have the following result.
This result
yields too many
board

members. Thus we must increase the modified standard divisor. By experimenting with different
divisors, we find that 950 gives the correct number of board members, as shown in the table on
the next page.

Thus the composition of the environmental board using the Jefferson method is Cardiff:
7, Solana: 2, Vista: 1, Pauma: 3, and Pacific: 5.

Suppose that the environmental agency in Example 1 decides to add one more member
to the board even though the population of each city remains the same. The total number of
members is now 19, and we must determine how the members of the board will be apportioned.
 20000
The standard divisor is now 19 = 1052.63. Using Hamilton’s method, the calculations
necessary to apportion the board members are shown below.

The table below summarizes the number of board members each city would have if the
board consisted of 18 members (Example 1) or 19 members.

Note that although one more board member was added, Vista lost a board member, even
though the populations of the cities did not change. This is called the Alabama paradox and has
a negative effect on fairness. In the interest of fairness, an apportionment method should not
exhibit the Alabama paradox.

6.1.3 Huntington-Hill Apportionment Method

House of Representatives are apportioned among the states every 10 years. The present method
used by the House is based on the apportionment principle and is called the method of equal
proportions or the Huntington-Hill method. This method has been used since 1940.

The Huntington-Hill method is implemented by calculating what is called a Huntington-

Hill number.
 ▼ Huntington-Hill Number
The value of
(𝑃𝐴 )2
𝑎(𝑎 + 1)

, where PA is the population of state A and a is the current number of

representatives from state A, is called the Huntington-Hill number for state A.

Example Problem #2: Use the Huntington-Hill Apportionment Principle

The table below shows the numbers of lifeguards that are assigned to three different
beaches and the numbers of rescues made by lifeguards at those beaches. Use the Huntington-
Hill apportionment principle to determine to which beach a new lifeguard should be assigned.

Solution

Calculate the Huntington-Hill number for each of the beaches. In this case, the
population is the number of rescues and the number of representatives is the number of
lifeguards.

Ferris has the greatest Huntington-Hill number. Thus, according to the Huntington-Hill
Apportionment Principle, the new lifeguard should be assigned to Ferris.

Exercise Set 6.1

1. Explain how to calculate the standard divisor of an apportionment for a total population p
with n items to apportion.

2. Teacher Aides A total of 25 teacher aides are to be apportioned among seven classes at a
new elementary school. The enrolments in the seven classes are shown in the following
table.

a. Determine the standard divisor. What is the meaning of the standard divisor in the
context of this exercise?
b. Use the Hamilton method to determine the number of teacher aides to be
apportioned to each class.
c. Use the Jefferson method to determine the number of teacher aides to be
apportioned to each class. Is this apportionment in violation of the quota rule?
d. How do the apportionment results produced using the Jefferson method compare
with the results produced using the Hamilton method?

3. College Enrolment. The following table shows the enrolment for each of the four divisions
of a college. The four divisions are liberal arts, business, humanities, and science. There are
180 new computers that are to be apportioned among the divisions based on the
enrolments.

a. What is the standard divisor for an apportionment of the computers? What is the
meaning of the standard divisor in the context of this exercise?
 b. Use the Hamilton method to determine the number of computers to be
apportioned to each division.
c. If the computers are to be apportioned using the Jefferson method, explain why
neither 86 nor 87 can be used as a modified standard divisor. Explain why 86.5
can be used as a modified standard divisor.
d. Explain why the modified standard divisor used in the Jefferson method cannot
be larger than the standard divisor.
e. Use the Jefferson method to determine the number of computers to be
apportioned to each division. Is this apportionment in violation of the quota rule?
f. How do the apportionment results produced using the Jefferson method
compare with the results produced using the Hamilton method?

4. Medical Care. A hospital district consists of six hospitals. The district administrators have
decided that 48 new nurses should be apportioned based on the number of beds in each
of the hospitals. The following table shows the number of beds in each hospital.

a. Determine the standard divisor. What is the meaning of the standard divisor in the
context of this exercise?
b. Use the Hamilton method to determine the number of nurses to be apportioned to
each hospital.
c. Use the Jefferson method to determine the number of nurses to be apportioned to
each hospital.
d. How do the apportionment results produced using the Jefferson method compare
with the results produced using the Hamilton method?

5. Hotel Management. A company operates four resorts. The CEO of the company decides to
use the Hamilton method to apportion 115 new LCD television sets to the resorts based on
the number of guest rooms at each resort.
 a. If the number of television sets to be apportioned by the Hamilton method increases
from 115 to 116, will the Alabama paradox occur?
b. If the number of television sets to be apportioned by the Hamilton method increases
from 116 to 117, will the Alabama paradox occur?
c. If the number of television sets to be apportioned by the Hamilton method increases
from 117 to 118, will the Alabama paradox occur?

6.2 Introduction to Voting

6.2.1 Plurality Method of Voting

One of the most revered privileges that those of us who live in a democracy enjoy is the right to
vote for our representatives. Sometimes, however, we are puzzled by the fact that the best
candidate did not get elected. Unfortunately, because of the way our plurality voting system
works, it is possible to elect someone or pass a proposition that has less than majority support.
As we proceed through this section, we will look at the problems with plurality voting and
alternatives to this system. We start with a definition.

TAKE
▼ The Plurality Method of Voting NOTE:
Each voter votes for one candidate, and the candidate with the most votes wins. When an
issue
The winning candidate does not have to have a majority of the votes. requires a
majority
vote, it
means that more than 50% of the people voting must vote for the issue. This is not the same as a plurality, in which the person
or issue with the most votes wins.

Example Problem #1: Determine the Winner Using Plurality Voting

Fifty people were asked to rank their preferences of five varieties of chocolate candy,
using 1 for their favorite and 5 for their least favorite. This type of ranking of choices is called a
preference schedule. The results are shown in the table below.
TAKE NOTE
A preference schedule lists the
number of people who gave a
particular ranking. For
example, the column shaded blue
means that 3 voters ranked solid
chocolate first, caramel centers
second, almond centers third, toffee
centers fourth, and vanilla centers
fifth.

According to this table, which variety of candy would win the taste test using the plurality
voting system?

Solution

To answer the question, we will make a table showing the number of first-place votes for each
candy.

Because toffee centers received 20 first-place votes, this type of candy would win the plurality
taste test.

Example Problem 2: Determine the 2nd Winner Using Plurality Voting

According to the table in Example 1, which variety of candy would win second place
using the plurality voting system?
Solution

Make a table showing the number of second-place votes for each candy.

It shows that the largest number of second-place votes (26) were for almond centers. Almond
centers would win second place using the plurality voting system.

Example 1 can be used to show the difference between plurality and majority. There were 20
first-place votes for toffee-centered chocolate, so it wins the taste test. However, toffee-
20
centered chocolate was the first choice of only 40% (50 = 40%) of the people voting. Thus less
than half of the people voted for toffee-centered chocolate as number one, so it did not receive
a majority vote.

6.2.2 Borda Count Method of Voting

The problem with plurality voting is that alternative choices are not considered. For instance, the
result of the Minnesota governor’s contest might have been quite different if voters had been
asked, “Choose the candidate you prefer, but if that candidate does not receive a majority of the
votes, which candidate would be your second choice?”

To see why this might be a reasonable alternative to plurality voting, consider the
following situation. Thirty-six senators are considering an educational funding measure. Because
the senate leadership wants an educational funding measure to pass, the leadership first
determines that the senators prefer measure A for $50 million over measure B for $30 million.
However, because of an unexpected dip in state revenues, measure A is removed from
consideration and a new measure, C, for $15 million, is proposed. The senate leadership
determines that senators favor measure B over measure C. In summary, we have

A majority of senators favor measure A over measure B.

A majority of senators favor measure B over measure C.

 From these results, it seems reasonable to think that a majority of senators would prefer
measure A over measure C. However, when the senators are asked about their preferences
between the two measures, measure C is preferred over measure A. To understand how this
could happen, consider the preference schedule for the senators shown in the following table.

Notice that 15 senators prefer measure A over measure C, but 12 + 9 = 21 senators, a

majority of the 36 senators, prefer measure C over measure A. According to the preference
schedule, if all three measures were on the ballot, A would come in first, B would come in second,
and C would come in third. However, if just A and C were on the ballot, C would win over A. This
paradoxical result was first discussed by Jean C. Borda in 1770.

In an attempt to remove such paradoxical results from voting, Borda proposed that voters
rank their choices by giving each choice a certain number of points.

▼ The Borda Count Method of Voting

If there are n candidates or issues in an election, each voter ranks the candidates or
issues by giving n points to the voter’s first choice, n - 1 points to the voter’s second
choice, and so on, with the voter’s least favorite choice receiving 1 point. The candidate
or issue that receives the most total points is the winner.
Applying the Borda count method to the education measures, a measure receiving a first-
place vote receives 3 points. (There are three different measures.) Each measure receiving a
second-place vote receives 2 points, and each measure receiving a third-place vote receives 1
point. The calculations are shown below.

Measure A: 15 first-place votes : 15 . 3 = 45

0 second-place votes : 0.2 =0
21 third-place votes : 1 . 1 = 21
Total : = 66

Measure B: 12 first-place votes : 12 . 3 = 36

9 second-place votes : 9 . 2 = 18
15 third-place votes : 15 . 1 = 15
 Total : = 69

Measure C: 9 first-place votes : 9 . 3 = 27

27 second-place votes: 27 . 2 = 54
0 third-place votes : 0.1 =0
Total : = 81

Using the Borda count method, measure C is the clear winner (even though it is not the
plurality winner).

Example Problem #3: Use the Borda Count Method

The members of a club are going to elect a president from four nominees using the
Borda count method. If the 100 members of the club mark their ballots as shown in the table
below, who will be elected president?

Solution
Using the Borda count method, each first-place vote receives 4 points, each second-
place vote receives 3 points, each third-place vote receives 2 points, and each last-place vote
receives 1 point. The summary for each candidate is shown below.
 Avalon: 0 First Place Votes : 0.4 =0
90 Second Place : 90 . 3 = 270
10 Third Place : 10 . 2 = 20
0 Fourth Place : 0.1 = 00
Total : = 290

Branson: 36 First Place Votes : 36 . 4 = 144

10 Second Place : 10 . 3 = 30
12 Third Place : 12 . 2 = 24
42 Fourth Place : 42 . 1 = 42
Total : = 240

Columbus: 28 First Place Votes : 28 . 4 = 112

0 Second Place : 0.3 =0
54 Third Place : 54 . 2 = 108
12 Fourth Place : 12 . 1 = 12
Total : = 232

Dunkirt: 36 First Place Votes : 36 . 4 = 144

0 Second Place : 0.3 =0
24 Third Place : 24 . 2 = 48
40 Fourth Place : 40 . 1 = 40
Total : = 232

Avalon has the largest total score. By the Borda count method, Avalon is elected president.

Example Problem #4: Use the Borda Count Method

The preference schedule given in Example 1 for the 50 people who were asked to rank
their preferences of five varieties of chocolate candy is shown again below.

Determine the taste test favorite using the Borda count method.
Solution
Using the Borda count method, each first-place vote receives 5 points, each second-place
vote receives 4 points, each third-place vote receives 3 points, each fourth place receives 2 points
and each last-place vote receives 1 point. The summary for each chocolate candies is shown
below.

Caramel Center: 0 First Place Votes : 0x5 = 0

3 Second Place Votes : 3x4 = 12
0 Third Place Votes : 0x3 = 0
30 Fourth Place Votes : 30 x 2 = 60
17 Fifth Place Votes : 17 x 1 = 17
Total 89

Vanilla Center: 17 First Place Votes : 17 x 5 = 85

0 Second Place Votes : 0x4 = 0
0 Third Place Votes : 0x3 = 0
0 Fourth Place Votes : 0x2 = 0
33 Fifth Place Votes : 33 x 1 = 33
Total 118

Almond Center: 8 First Place Votes : 8x5 = 40

26 Second Place Votes : 26 x 4 = 104
16 Third Place Votes : 16 x 3 = 48
0 Fourth Place Votes : 0x2 = 0
0 Fifth Place Votes : 0x1 = 0
Total 192

Toffee Center: 20 First Place Votes : 20 x 5 = 100

2 Second Place Votes : 2x4 = 8
8 Third Place Votes : 8x3 = 24
20 Fourth Place Votes : 20 x 2 = 40
0 Fifth Place Votes : 0x1 = 0
Total 172

Solid Chocolate: 5 First Place Votes : 5x5 = 25

19 Second Place Votes : 19 x 4 = 76
26 Third Place Votes : 26 x 3 = 78
0 Fourth Place Votes : 0x2 = 0
 0 Fifth Place Votes : 0x1 = 0
Total 179

Using the Borda count method, Almond Center with highest score of 192 is the first choice
(even though it is not the plurality winner).

6.2.3 Plurality with Elimination

A variation of the plurality method of voting is called plurality with elimination. Like the Borda
count method, the method of plurality with elimination considers a voter’s alternate choices.

Suppose that 30 members of a regional planning board must decide where to build a new
airport. The airport consultants to the regional board have recommended four different sites.
The preference schedule for the board members is shown in the following table.

Using the plurality with elimination method, the board members first eliminate the site
with the fewest number of first-place votes. If two or more of these alternatives have the same
number of first-place votes, all are eliminated unless that would eliminate all alternatives. In that
case, a different method of voting must be used. From the table above, Bremerton is eliminated
because it received only two first-place votes. Now a vote is retaken using the following
important assumption: Voters do not change their preferences from round to round. This means
that after Bremerton is deleted, the 12 people in the first column would adjust their preferences
so that Apple Valley becomes their second choice, Cochella remains their first choice, and Del
Mar becomes their third choice. For the 11 voters in the second column, Apple Valley remains
their first choice, Cochella remains their second choice, and Del Mar becomes their third choice.
Similar adjustments are made by the remaining voters. The new preference schedule is
The board members now repeat the process and eliminate the site with the fewest first-place
votes. In this case it is Del Mar. The new adjusted preference schedule is

From this table, Apple Valley has 16 first-place votes and Cochella has 14 first-place votes.
Therefore, Apple Valley is the selected site for the new airport.

Example Problem #4: Use the Plurality with Elimination Voting Method
A university wants to add a new sport to its existing program. To help ensure that the new
sport will have student support, the students of the university are asked to rank the four sports
under consideration. The results are shown in the following table.

Use the plurality with elimination method to determine which of these sports should be added
to the university’s program.

Solution
 Because rowing received no first-place votes, it is eliminated from consideration. The new
preference schedule is shown below.

From this table, lacrosse has 464 first-place votes, squash has 494 first-place votes, and golf has
613 first-place votes. Because lacrosse has the fewest first-place votes, it is eliminated. The new
preference schedule is shown below.

From this table, squash received 744 first-place votes and golf received 827 first-place votes.
Therefore, golf is added to the sports program.

Example Problem #4: Use the Plurality with Elimination Voting Method
A service club is going to sponsor a dinner to raise money for a charity. The club has
decided to serve Italian, Mexican, Thai, Chinese, or Indian food. The members of the club were
surveyed to determine their preferences. The results are shown in the table below.

Solution

Because Indian received no first-place votes, it is eliminated from consideration. The new
preference foods is shown below.

Foods Ranking
  Italian 2 4 1 3 3
 Mexican 1 3 4 2 1
 Thai 3 1 3 4 2
 Chinese 4 2 2 1 4
Number of Ballots 33 30 25 20 18

From this table, Italian has 25 first-place votes, Mexican has 51 first-place votes, Thai has 30 first-
place votes and Chinese has 20. Because Chinese has the fewest first-place votes of 20, it is
eliminated. The new preference schedule is shown below.

Foods Ranking
 Italian 2 3 1 2 3
 Mexican 1 2 3 1 1
 Thai 3 1 2 3 2
Number of Ballots 33 30 25 20 18

The club now repeat the process and eliminate the food with the fewest first-place votes. In
this case it is Italian with 25 first-place votes as compared to Mexican with 71 first-place votes
and Thai with 30 first-place votes. The new adjusted preference schedule is

Foods Ranking
 Mexican 1 2 2 1 1
 Thai 2 1 1 2 2
Number of Ballots 33 30 25 20 18

In this ranking, Thai food received the fewest first-place votes, so it is eliminated. The
preference for the banquet food is Mexican.

Exercise 6.2
1. What is the difference between a majority and a plurality? Is it possible to have one
without the other?
2. Explain why the plurality voting system may not be the best system to use in some
situations.
3. Explain how the Borda count method of voting works.
4. Explain how the plurality with elimination voting method works.
5. Presidential Election The table below shows the popular vote and the Electoral College
vote for the major candidates in the 2000 presidential election.
 a. Which candidate received the plurality of the popular vote?
b. Did any candidate receive a majority of the popular vote?
c. Who won the election?

6. Cartoon Characters A kindergarten class was surveyed to determine the children’s

favorite cartoon characters among Dora the Explorer, SpongeBob Square Pants, and Buzz
Lightyear. The students ranked the characters in order of preference; the results are
shown in the preference schedule below.

a. How many students are in the class?

b. How many votes are required for a majority?
c. Using plurality voting, which character is the children’s favorite?

7. Catering A 15-person committee is having lunch catered for a meeting. Three caterers,
each specializing in a different cuisine, are available. In order to choose a caterer for the
group, each member is asked to rank the cuisine options in order of preference. The
results are given in the preference schedule below.

Using plurality voting, which caterer should be chosen?

 8. Movies. Fifty consumers were surveyed about their movie-watching habits. They were
asked to rank the likelihood that they would participate in each listed activity. The
results are summarized in the table below.

Using the Borda count method of voting, which activity is the most popular choice
among this group of consumers?

9. Cartoons. Use the Borda count method of voting to determine the children’s favorite
cartoon character in Exercise 6.

10. Catering. Use the Borda count method of voting to determine which caterer the
committee should hire in Exercise 7.

11. Class Election. A senior high school class held an election for class president. Instead of
just voting for one candidate, the students were asked to rank all four candidates in order
of preference. The results are shown below.

Using the Borda count method, which student should be class president?

12. Maritime and maritime-related Programs. A number of Senior High School students were
asked to rank five maritime and maritime-related programs in order of preference. The
responses are given in the table below.

Programs Ranking
Marine Transportation 3 1 5 2 5
 Marine Engineering 1 3 1 1 4
Customs Administration 4 2 4 5 2
Hotel and Restaurant Mgt 2 5 2 3 3
Tourism Management 5 4 3 4 1
Number of Ballots 57 72 38 61 15

Use plurality with elimination to determine the students’ preference among the five
programs.

13. Class Election. Use plurality with elimination to choose the class president in Exercise 11.

14. Campus Club. A campus club has money left over in its budget and must spend it before
the school year ends. The members arrive at five different possibilities, and each member
ranks them in order of preference. The results are shown in the table below.

a. Using the plurality voting system, how should the club spend the money?
b. Use the plurality with elimination method to determine how the money should be
spent.
c. Using the Borda count method of voting, how should the money be spent?
d. In your opinion, which of the previous three methods seems most appropriate in this
situation? Why?

15. Recreation. A company is planning its annual summer retreat and has asked its
employees to rank five different choices of recreation in order of preference. The results
are given in the table below.
 a. Using the plurality voting system, what activity should be planned for the retreat?
b. Use the plurality with elimination method to determine which activity should be
chosen.
c. Using the Borda count method of voting, which activity should be planned?

4.3 Weighted Voting Systems

4.3.1 Biased Voting Systems

A weighted voting system is one in which some voters have more weight on the outcome of an
election. Examples of weighted voting systems are fairly common. A few examples are the
stockholders of a company, the Electoral College, the United Nations Security Council, and the
European Union.

The Electoral College. As mentioned in the Historical Note below, the Electoral College
elects the president of the United States. The number of electors representing each state is equal
to the sum of the number of senators (2) and the number of members in the House of
Representatives for that state. The original intent of the framers of the Constitution was to
protect the smaller states. We can verify this by computing the number of people represented
by each elector. In the 2010 election, each Vermont elector represented about 209,000 people;
each California elector represented about 677,000 people. To see how this gives a state with a
smaller population more power (a word we will discuss in more detail later in this section), note
that three electoral votes from Vermont represent approximately the same size population as
does one electoral vote from California. Not every vote represents the same number of people.
Another peculiarity related to the Electoral College system is that it is very sensitive to small vote
swings. For instance, in the 2000 election, if an additional 0.01% of the voters in Florida had cast
their votes for Al Gore instead of George Bush, Gore would have won the presidential election.

Historical Notes:
The U.S. Constitution, Article 2, Section 1 states that the members of the Electoral College elect the
president of the United States. The original article directed members of the College to vote for two
people. However, it did not stipulate that one name was for president and the other name was for
vice president. The article goes on to state that the person with the greatest number of votes becomes
president and the one with the next highest number of votes becomes vice president. In 1800, Thomas
Jefferson and Aaron Burr received exactly the same number of votes even though they were running
on a Jefferson for president, Burr for vice president ticket. Thus the House of Representatives was
asked to select the president. It took 36 different votes by the House before Jefferson was elected
president. In 1804, the Twelfth Amendment to the Constitution was ratified to prevent a recurrence
of the 1800 election problems.

HISTORICAL N OTE
Consider a small company with a total of 100 shares of stock and three stockholders,
A, B, and C. Suppose that A owns 45 shares of the stock (which means A has 45 votes),
B owns 45 shares, and C owns 10 shares. If a vote of 51 or greater is required to approve any
measure before the owners, then a measure cannot be passed without two of the three owners
voting for the measure. Even though C has only 10 shares, C has the same voting power as A and
B.

Now suppose that a new stockholder is brought into the company and the shares of the
company are redistributed so that A has 27 shares, B has 26 shares, C has 25 shares, and D has
22 shares. Note, in this case, that any two of A, B, or C can pass a measure, but D paired with any
of the other shareholders cannot pass a measure. D has virtually no power even though D has
only three shares less than C.

The number of votes that are required to pass a measure is called a quota. For the two
stockholder examples above, the quota was 51. The weight of a voter is the number of votes
controlled by the voter. In the case of the company whose stock was split A –27 shares, B –26
shares, C –25 shares, and D –22 shares, the weight of A is 27, the weight of B is 26, the weight of
C is 25, and the weight of D is 22. Rather than write out in sentence form the quota and weight
of each voter, we use the notation
 This notation is very convenient. We state its more general form in the following
definition.

Using
▼ Weighted Voting System
this
A weighted voting system of n voters is written {𝑞: 𝑤1 , 𝑤2 , … 𝑤𝑛 }, where q is
the quota and w1 through wn represent the weights of each of the n voters.
notation, we can describe various voting systems.

 One person, one vote: For instance, (5: 1, 1, 1, 1, 1, 1, 1, 1, 1). In this system, each person
has one vote and five votes, a majority, are required to pass a measure.

 Dictatorship: For instance, (20: 21, 6, 5, 4, 3). In this system, the person with 21 votes can
pass any measure. Even if the remaining four people get together, their votes do not total
the quota of 20.

 Null system: For instance, (28: 6, 3, 5, 2). If all the members of this system vote for a
measure, the total number of votes is 16, which is less than the quota. Therefore, no
measure can be passed.

 Veto power system: For instance, (21: 6, 5, 4, 3, 2, 1). In this case, the sum of all the votes
is 21, the quota. Therefore, if any one voter does not vote for the measure, it will fail.
Each voter is said to have veto power. In this case, this means that even the voter with
one vote can veto a measure (cause the measure not to pass). A voter has veto power
whenever a measure cannot be passed without that voter's vote. If at least one voter in
a voting system has veto power, the system is a veto power system.

▼ Number of Possible Coalitions of n Voters

The number of possible coalitions of n voters is 2n - 1.

Example Problem #1: Determine Winning Coalitions in a Weighted Voting System

Suppose that the four owners of a company, Ang, Bonhomme, Carmel, and Diaz, own,
respectively, 500 shares, 375 shares, 225 shares, and 400 shares. There are a total of 1500 votes;
half of this is 750, so the quota is 751. The weighted voting system for this company is {751:
500, 375, 225, 400}.
 a. Determine the winning coalitions.
b. For each winning coalition, determine the critical voters.

Solution
a. A winning coalition must represent at least 751 votes. We will list these coalitions in the
table below, in which we use A for Ang, B for Bonhomme, C for Carmel, and D for Diaz.

TAKE NOTE: The coalition{𝐴, 𝐶} is not a winning coalition because the total number of votes for
that coalition is 725, which is less than 751.

b. A voter who leaves a winning coalition and thereby creates a losing coalition is a critical
voter. For instance, for the winning coalition{𝐴,𝐵, 𝐶}, if A leaves, the number of remaining
votes is 600, which is not enough to pass a resolution. If B leaves, the number of remaining
votes is 725—again, not enough to pass a resolution. If C leaves, the number of remaining
votes is 875, which is greater than the quota. Therefore, A and B are critical voters for the
coalition{𝐴,𝐵, 𝐶} and C is not a critical voter. The table below shows the critical voters for
each winning coalition.
Example Problem #2: Determine Winning Coalitions in a Weighted Voting System
Many countries must govern by forming coalitions from among many political parties.
Suppose a country has five political parties named A, B, C, D, and E. The numbers of votes,
respectively, for the five parties are

A=22 votes, B=18 votes, C=17 votes, D=10 votes, E=5 votes

a. Determine the winning coalitions if 37 votes are required to pass a resolution.

b. For each winning coalition, determine the critical voters.

Solution

a. A winning coalition must represent at least 37 votes. We will list these coalitions in the
table below.

Winning Coalition Number of Votes

{𝐴, 𝐵} 40
{𝐴, 𝐶} 39
A, B, C 57
A, B, D 50
A, B, E 45
A, C, D 40
A, C, E 44
A, D, E 37
B, C, D 45
B, C, E 40
A, B, C, D 67
A, B, C, E 62
A, B, D, E 55
A, C, D, E 54
B, C, D, E 50
A, B, C, D, E 72

TAKE NOTE: The following coalitions are not a winning coalition because the total number of
votes for that coalitions are less than 37.

Losing Coalition Number of Votes

{𝐴, 𝐷} 32
{𝐴, 𝐸} 27
{𝐵, 𝐶} 35
{𝐵, 𝐷} 28
{𝐵, 𝐸} 23
 {𝐶, 𝐷} 27
{𝐶, 𝐸} 22
{𝐷, 𝐸} 15
{𝐶, 𝐷, 𝐸} 32

b. Critical Voters:

Winning Coalition Number of Votes Critical Voters

{𝐴, 𝐵} 40 A, B
{𝐴, 𝐶} 39 A, C
A, B, C 57 A
A, B, D 50 A
A, B, E 45 A, B,
A, C, D 40 A, C, D
A, C, E 44 A, C, E
A, D, E 37 A, D, E
B, C, D 45 B, C, D
B, C, E 40 B, C, E
A, B, C, D 67 None
A, B, C, E 62 None
A, B, D, E 55 A
A, C, D, E 54 A
B, C, D, E 50 B, C,
A, B, C, D, E 72 None