Manual

of
Job-Related Thinking Skills

Office of Human Resources Management

Personnel Research and Assessment Division
Washington, DC 20536
Reviewed 2011

Acknowledgment
The Personnel Research and Assessment Division would like to acknowledge the assistance
of the following agents who reviewed this manual:
Larry Ball
Special Agent, Dallas District Office
Richard Barch
Assistant Patrol Agent in Charge, Uvalde, TX
James Clardy
Intelligence Agent, Havre, MT
John Esquivel
Supervisory Border Patrol Agent, El Paso, TX
Edward Gerber
Assistant Chief Patrol Agent, El Paso, TX
Eduardo Martinez
Assistant Patrol Agent in Charge, Comstock, TX

i Table of Contents

Table of Contents

Page

Unit I - Introduction to Job-Related Thinking Skills......................................................... 1

Pretest for Manual of Job-Related Thinking Skills.............................................................. 3
Unit II - Deductive Reasoning -- Empowering the Mind..................................................... 7
Introduction to Deductive Reasoning .................................................................................. 7
Part II.A. Reasoning With Two Sets ................................................................................... 9
Section II.A.1. What Are Sets?....................................................................................... 9
Section II.A.2. The Four Basic Statements About Sets................................................ 10
Section II.A.3. Conclusions From All Are Statements ............................................. 14
Section II.A.4. Conclusions From None Are Statements.......................................... 17
Section II.A.5. Conclusions From Some Are Statements ......................................... 19
Section II.A.6. Conclusions From Some Are Not Statements .................................. 21
Section II.A.7. Conclusions When the Basic Statements Are False............................. 24
Part II.B Reasoning with Three Sets.................................................................................. 29
Section II.B.1. Forms for Relating Three Sets.............................................................. 29
Section II.B.2. Advanced Topic -- Syllogistic Rules (reading this section is optional)35
Section II.B.3. Illogical Biases in Relating Three Sets................................................. 37
Part II.C. Reasoning with Connectives............................................................................. 39
Section II.C.1. Conjunction and Negation.................................................................... 40
Section II.C.2. The Conditional.................................................................................... 43
Section II.C.3. Other Basic Connectives ...................................................................... 48
Section II.C.4. Advanced Topic -- Complex Connectives ........................................... 52
Part II.D. Using Language Correctly in Deduction ......................................................... 55

Unit III. Induction 1 -- Reasoning About Real-World Events ....................................... 57

Part III.A. Introduction to Inductive Reasoning............................................................... 57
Part III.B. Inductive Reasoning About Sets ...................................................................... 61
Section III.B.1: Inductive Reasoning About Two Sets.................................................. 61
Section III.B.2. Advanced Topic -- Inductive Reasoning With More Than Two Sets.. 67

ii Table of Contents

Table of Contents

Page

Part III.C. Inductive Reasoning With Connectives........................................................... 70

Section III.C.1: Conditionals With Probabilities.......................................................... 70
Section III.C.2: Connectives With an Added Premise Containing a Probability ......... 73

Part III.D. Fallacies in Induction ............................................................................ 78

Unit IV - Induction 2: Statistical Reasoning and Estimating Probability..................... 80
Part IV.A. Populations and Samples ................................................................................ 80
Section IV.A.1. Law of Large Numbers....................................................................... 81
Section IV.A.2. Sample Representativeness................................................................. 83
Section IV.A.3. Regression to the Mean ...................................................................... 85
Section IV.A.4. Sample Validity .................................................................................. 87
Section IV.A.5. Summary............................................................................................. 88

Part IV.B. Estimating Probability ........................................................................... 89

Section IV.B.1.
Section IV.B.2.
Section IV.B.3.
Section IV.B.4.
Section IV.B.5.
Section IV.B.6.

Probability Defined............................................................................. 89
Base Rate Information and Probability Estimates .............................. 89
Occurrence of Two or More Events ................................................... 90
Multiplicative Rules of Probability..................................................... 90
Additive Rules of Probability ............................................................. 93
Summary............................................................................................. 95

Part IV.C. Biases in Statistical Reasoning and Estimations of Probability . 96

Section IV.C.1. Availability ......................................................................................... 96
Section IV.C.2. Representativeness.............................................................................. 96
Section IV.C.3. Dilution ............................................................................................... 97
Section IV.C.4. Anchoring and Adjustment................................................................. 97
Section IV.C.5. Order Effect ........................................................................................ 98
Section IV.C.6. Ignoring Relative Frequency .............................................................. 99
Section IV.C.7. Concrete Information.......................................................................... 99
Section IV.C.8. Gamblers Fallacy............................................................................. 100
Section IV.C.9. Misperceptions about Equiprobability.............................................. 100
Section IV.C.10. Overestimating Conjunctive Events ............................................... 101
Section IV.C.11. Summary......................................................................................... 102
Unit V - Posttest for Manual of Job-Related Thinking Skills........................................ 103
Unit II - Answers to Self Tests.......................................................................................... 111
Glossary of Key Terms...................................................................................................... 139

Unit I - Introduction to Job-Related Thinking Skills

A Border Patrol Agent deciding whether an impending confrontation involves dangerous
individuals, an Inspector determining if an alien should be admitted, a supervisory Special
Agent deciding whether to open a case based on certain evidence -- all of these are DHS
employees using their thinking skills. Every day, DHS employees use their thinking skills
in countless decisions, determinations, and investigations. Thinking skills are by no means
the only skills used by DHS employees, but they are the most important. They are used in
applying rules, making determinations, making predictions, and in problem solving, on-thespot decision making, and complex, deliberate decision making.
In spite of the importance of thinking skills, most people do not receive special training in
thinking. Employees receive training in job knowledge and supervisors usually receive
training in supervisory techniques. Those types of training help employees think about very
specific problems. By contrast, the purpose of this module is to teach general thinking
skills that can be applied to any subject or situation.
In college programs, thinking skills are taught in logic courses. In addition, courses in
research methods and statistics teach people how to collect and analyze data in order to draw
correct conclusions. This manual draws heavily on the parts of logic and statistical
reasoning that are useful to DHS employees.
Which Job-Related Thinking Skills Are Included in This Manual?
The first skills in this manual are the basic thinking skills, which can be called deduction
and induction. These are the two types of skills that are used in drawing conclusions from
given information. Deduction and induction are very similar. They differ in only two ways-the completeness of the available information and the degree of certainty of the conclusion,
as indicated below:

Deduction: The individual has all the information necessary to draw a conclusion.
The conclusion is certain; it is true if the evidence is true.

Induction: The individual does not have complete evidence. He or she draws a
conclusion based on the information available. The conclusion is uncertain, that is
to say, probabilistic; it may not be true even if the evidence is true.

In the work of DHS, deduction is typically used in applying laws and rules to specific
situations. Induction is used in situations in which officers need to make on-the-spot
probabilistic judgments, sometimes in life-threatening situations.

Statistical reasoning is a tool to help us draw conclusions when we do not have complete
information. We can think of statistical reasoning as a powerful aid to induction.
Throughout this manual, there will be tips to help you avoid common mistakes of reasoning
(these mistakes are called fallacies). In addition, some sections contain Logic Notes that
give you more detail about logical issues. You do not need to read these notes in order to
understand the rest of the manual, but you are encouraged to do so. In addition, there are a
number of sections that contain Advanced Topics, which are more difficult topics. If you
have time to study these topics, you are encouraged to do so. However, you do not need to
study these topics to understand the rest of the manual. At the back of the manual, there is a
Glossary of Key Terms that you should refer to whenever you want to know the meaning of
a term that is defined elsewhere in the manual.
How Should You Use This Training Manual?
This manual is designed for self-instruction. The information is presented in small sections
that you can study separately. At the end of each section, there is a Self-Test that you
should take in order to check your understanding of the material you have studied. It is very
important that you answer all of these questions and then study the answers to the questions.
The answers to these exercises are found at the end of the manual. On the side of each page
of answers, there are one, two, three, four, or five lines (like this
) which indicate that
the answers belong to Unit I, II, III, IV, or V. If you think you may want to work through
the manual a second time, you may wish to write your answers to the Self-Tests on a
separate piece of paper so that when you go back to review, the answers will not be already
filled in.
Self-Test: Unit I (answers are given on page 107)
For each of the statements below, decide which of the two words in parentheses makes the
statement a true statement.
1. Deduction and induction are (basic, advanced) types of thinking skills.
2. Deduction is the type of reasoning that leads to a (certain, probabilistic) conclusion.
3. Deductive conclusions are based on (incomplete, complete) evidence.
4. Induction is the type of reasoning that leads to a (certain, probabilistic) conclusion.
5. Inductive conclusions are based on (incomplete, complete) evidence.
6. Statistical reasoning is an aid to (deduction, induction).

3
Before studying the rest of the manual, take this 32-question pretest and check your
answers (answers begin on page 107). At the end of the manual, there is a posttest that
you can take to assess your improvement.

Pretest for Manual of Job-Related Thinking Skills

The following passage describes a set of facts. The passage is followed by eight
conclusions. Read the passage and then decide whether each conclusion is:
T) true, which means that you can infer the conclusion from the facts given; or
F) false, which means that the conclusion is contrary to the facts given; or whether there is
I) insufficient information to decide, which means that there is insufficient information for
you to determine whether the facts imply the conclusion or are contrary to the
conclusion.
All narcotics cases resulting from apprehensions by the Border Patrol are referred to the
Drug Enforcement Administration (DEA); narcotics cases are the only cases referred to
DEA by the Border Patrol. At Station X, last month was unusual in that none of the
criminal cases resulting from apprehensions by the Border Patrol were narcotics cases. A
few of the criminal cases resulting from apprehensions that month involved alien smuggling,
and these were not referred to another agency.
1.

Last month, there were no narcotics cases among the criminal cases that resulted from
apprehensions by the Border Patrol in Station X. (T / F / I)

2.

Last month, some of the criminal cases resulting from apprehensions by the Border
Patrol in Station X were referred to another agency. (T / F / I)

3.

Last month, none of the criminal cases resulting from apprehensions in Station X were
referred to the DEA. (T / F / I)

4.

Any case that is not referred to DEA is not a narcotics case. (T / F / I)

5.

A few of the criminal cases resulting from apprehensions by the Border Patrol in Station
X last month were narcotics cases. (T / F / I)

6.

There are no narcotics cases resulting from apprehensions by the Border Patrol that are
not referred to the DEA. (T / F / I)

7.

There were some cases that were referred to another agency last month that were not
criminal cases. (T / F / I)

8.

Last month, there were some alien smuggling cases resulting from apprehensions by the
Border Patrol in Station X that did not involve narcotics. (T / F / I)

4
The following passage describes a set of facts. The passage is followed by eight
conclusions. Read the passage and then decide whether each conclusion is:
T) true, which means that you can infer the conclusion from the facts given; or
F) false, which means that the conclusion is contrary to the facts given; or whether there is
I) insufficient information to decide, which means that there is insufficient information for
you to determine whether the facts imply the conclusion or are contrary to the
conclusion.
If aliens do not present themselves for inspection at a designated port of entry, they cannot
enter the U.S.A. legally. In order to enter, the alien must have either a border crossing card
or a resident alien card or a passport. During a specific one-week period, records at one
port of entry showed that if an alien had a border crossing card, he or she was either visiting
family members or entering to shop. One individual, J.T., entered at this port of entry
during this week with a border crossing card. J.T. was not entering to shop in the U.S.A.
9.

An alien who does not have a border crossing card would need to present a resident
alien card or a passport to enter the U.S.A. legally. (T / F / I)

10. If an alien entered the U.S.A. legally, then that alien must have presented himself or
herself at a designated port of entry. (T / F / I)
11. Another alien, M.N., presented herself at the designated port of entry; therefore, she was
allowed to enter. (T / F / I)
12. During the one-week period referred to in the paragraph, if an alien who entered
through this port of entry was shopping in the U.S.A., that person had presented a
border crossing card. (T / F / I)
13. J.T. entered at the port of entry to visit his family. (T / F / I)
14. J.T. can enter the U.S.A. legally without presenting himself at the port of entry.
(T / F / I)
15. During the one-week period referred to in the paragraph, any alien who entered at this
port of entry without a border crossing card was not visiting family. (T / F / I)
16. During this one week period, if an alien did not present himself or herself at this port of
entry, he or she did not have either a border crossing card or a resident alien card or a
passport. (T / F / I)

5
The following passage describes a set of facts. The passage is followed by eight
conclusions. Read the passage and then decide whether each conclusion is:
T) true, which means that you can infer the conclusion from the facts given; or
F) false, which means that the conclusion is contrary to the facts given; or whether there is
I) insufficient information to decide, which means that there is insufficient information for
you to determine whether the facts imply the conclusion or are contrary to the
conclusion.
An illegal alien can either take voluntary departure or be deported, but not both. In a onemonth period in Sector X, all apprehensions of illegal aliens were of people who
subsequently took voluntary departure. During this period, there were several apprehensions
that involved extensive pursuit by local authorities. Three of these cases involved pursuit to
the same crossroads. At this crossroads, the road straight ahead was visible for a mile. A
left turn at this crossroads led to a road that was hidden from view. There was no right turn
at the crossroads. Therefore, in all three cases, when local officers lost sight of the person
they were pursuing just before coming to the crossroads, they knew that the individual had
either turned left, or been picked up by a vehicle, or both. In summarizing the cases after all
three were apprehended, the local authorities concluded that only one of the individuals was
picked up by a vehicle.
Evaluate these conclusions with respect to Sector X in this one-month time period.
17. Some cases that required extensive pursuit by local authorities resulted in the voluntary
departure of aliens. (T / F / I)
18. Some apprehended aliens were deported. (T / F / I)
19. All apprehensions of illegal aliens occurred as the result of an extensive pursuit by local
authorities. (T / F / I)
20. In at least two of the cases that were pursued to the crossroads, the individuals being
pursued took the left turn. (T / F / I)
21. There were some pursuits by local authorities that led to apprehensions of illegal aliens.
(T / F / I)
22. Every person who took voluntary departure was apprehended after an extensive pursuit
by local authorities. (T / F / I)
23. Some apprehensions did not occur without an extensive pursuit by local authorities.
(T / F / I)
24. One of the individuals pursued to the crossroads did not turn left and was not picked up
by a vehicle. (T / F / I)

6
The following passage describes a set of facts. The passage is followed by eight
conclusions. Read the passage and then decide whether each conclusion is:
T) true, which means that you can infer the conclusion from the facts given; or
F) false, which means that the conclusion is contrary to the facts given, or whether there is
I) insufficient information to decide, which means that there is insufficient information for
you to determine whether the facts imply the conclusion or are contrary to the
conclusion.
A railroad bridge that crosses the U.S. border has a catwalk for workers that is used by
aliens to enter the U.S.A. illegally. A special sensor installed on the bridge alerts Border
Patrol Agents to the presence of people on the bridge when no train is present. If there is no
train, 70% of sensor signals (so-called hits) produce apprehensions of at least one new
illegal alien. When a train is present, it masks the sensor hits. Apprehensions are made
when a train is present, but the sensor is not a useful tool for signaling the presence of
people.
In an average week, there are 100 sensor hits in the absence of trains and 10 apprehensions
in the presence of trains. A group of illegal aliens, which we will call group A, was
apprehended while crossing the railroad bridge. Another group, group B, crossed the bridge
but was not apprehended.
25. If an illegal alien crosses the bridge and activates the sensor when a train is present, the
probability is .10 that the alien will be apprehended. (T / F / I)
26. If a Border Patrol Agent responds to a sensor hit when no train is present, there is a 70%
chance that an alien will be apprehended. (T / F / I)
27. More likely than not, Group B crossed the railroad bridge when the train was present.
(T / F / I)
28. It is very likely that Group A crossed the railroad bridge when the train was not present.
(T / F / I)
29. Trains are present at this bridge approximately 10% of the time. (T / F / I)
30. It is not true that no apprehensions occur when a train is present. (T / F / I)
31. When a train is not present, a sensor hit is certain to produce an apprehension of an
illegal alien. (T / F / I)
32. More illegal aliens attempted to cross the bridge when no train was present than when a
train was present. (T / F / I)

End of Pretest. Check your answers against those on pages 107-110.

Unit II - Deductive Reasoning -- Empowering the Mind

Introduction to Deductive Reasoning
Deductive reasoning empowers your mind because it allows you to draw conclusions that
are certain, that is, conclusions that must be true if they are based on true information. This
can help you on your job because much of the work of DHS employees involves deductive
activities such as applying rules and making determinations.
The term valid conclusion means a conclusion that is justified, given the evidence. A
conclusion that is not justified is called an invalid conclusion.
Training in deductive reasoning teaches you to recognize several types of simple sentences
or statements and to understand which conclusions can be drawn from them. This may
sound like an academic exercise, but it is not. These simple sentences correspond to real-life
situations.
The following example shows you a simple statement and a sample of the valid conclusions
that can be drawn from it. There is also a sample of invalid conclusions--conclusions that
cannot be drawn from the statement. Study these sentences and be sure that you understand
why statements 1a and 1b are valid conclusions from statement 1. Also be sure you
understand why statements 1c and 1d are invalid conclusions from statement 1.
Statement 1.
Valid conclusions
Statement 1a.
Statement 1b.
Invalid conclusions
Statement 1c.
Statement 1d.

All DHS employees are employees of the Federal Government.

Some employees of the Federal Government are DHS employees.
Anyone who is not an employee of the Federal Government is not
a DHS employee.
No employees of the Federal Government are DHS employees.
Some DHS employees are not employees of the Federal
Government.

Statement 1 is an example of a sentence that has the following basic form:

Statement 2.

All [members of category A] are [members of category B].

In statement 1 above, DHS employees are the members of category A and employees of
the Federal Government are the members of category B.

8
Any statement that has the same form as statement 2 will have the same pattern of valid
conclusions as statements 1a and 1b and the same pattern of invalid conclusions as
statements 1c and 1d. These patterns, illustrated in terms of category A and category B, are
as follows:
Statement 2.

All [members of category A] are [members of category B].

Valid conclusions
Statement 2a.
Statement 2b.

Some [members of category B] are [members of category A].

All [nonmembers of category B] are [nonmembers of category A].

Invalid conclusions
Statement 2c.
Statement 2d.

No [members of category B] are [members of category A].

Some [members of category A] are not [members of category B].

In the rest of Unit II, you will learn about many other forms for statements that permit valid
deductive conclusions. You will also learn how to combine information from two or more
statements to reach valid conclusions.
Self-Test: Introduction to Deductive Reasoning (answers are given on page 111)
1. A deductive conclusion is true if it is based on true evidence. (True / false)
2. A conclusion that is not justified, given the evidence, is invalid. (True / false)
3. From the statement All guns are weapons, write one valid conclusion (like 2a or 2b):
4. From the statement All guns are weapons, write one invalid conclusion (like 2c or 2d):
Questions 5 through 8 use nonsense words in order to let you answer the questions strictly
on the basis of the logical form, without being influenced by your prior knowledge.
If the statement All gorms are lames is true, which of the following conclusions are valid?
5. Some gorms are not lames. (Valid / invalid)
6. Some lames are gorms. (Valid / invalid)
7. No lames are gorms. (Valid / invalid)
8. All non-lames are non-gorms. (Valid / invalid)

Part II.A. Reasoning With Two Sets

The best starting place for learning about deduction is with sentences that show the
relationship between two sets or two categories, like the sentences in the preceding section.
Before we talk more about sentences, though, we should be sure the term set is clear to you.

Section II.A.1. What Are Sets?

Sets are groupings of individuals or things that share one or more characteristics. The
following words can be considered to represent sets:
laws
agencies

vehicles
employees

Border Patrol Agents

cases

The following lists show words and phrases that are not sets (left column) and a way of
rewriting them as a phrase that expresses a set (right column). As you can see, words that
express a characteristic can be changed into sets by adding a noun such as things or
individuals.
Not Sets

Sets

Green

green things

Illegal

illegal aliens or illegal acts

Detained

detained individuals

crossed the border

people who crossed the border

Frequently sentences refer to sets implicitly rather than explicitly. These sentences can be
rewritten to show explicitly the relationship between sets. The following is an example:
Statement 3a.

All supervisors will attend training this year.

Statement 3b.

All supervisors are employees who will attend training this year.

In statement 3a, the italicized portion is not written in terms of a set. In statement 3b, the
italicized portion is rewritten to express a set. Notice that the content of the two sentences is
essentially the same.1
Statements 4a and 4b illustrate the same point:

Logic Note. The subject of the sentence is usually one of the sets and the predicate of the sentence usually
represents the other set. You will notice that the italicized portion of sentences 3a and 4a are the predicates of
their sentences. These portions are rewritten as sets in sentences 3b and 4b.

10
Statement 4a.

Some illegal immigrants live in Texas.

Statement 4b.

Some illegal immigrants are people who live in Texas.

Self-Test: Section II.A.1 (answers are given on page 111)

In 1 through 5, decide if the word or phrase expresses a set. If it does not, change the word
or phrase so that it expresses a set.
If not a set, rewrite:
1.

listening

(Set / not a set)

2.

District Directors

(Set / not a set)

3.

hazardous

(Set / not a set)

4.

completed and filed

(Set / not a set)

5.

regular working hours (Set / not a set)

Each sentence in 6 through 10 below refers to two sets. What are the two sets referred to in
each sentence? Underline the two sets. In each sentence, rewrite the second set so that it is
expressed in set language. Refer to statements 3a and 3b and 4a and 4b for examples.
6.

Some computer software is user-friendly.

7.

Some law enforcement personnel carry guns.

8.

Some Mexicans were deported.

9.

All Border Patrol Agents speak Spanish.

10. All service employees are responsible for immediately reporting any allegation of
misconduct.

Section II.A.2. The Four Basic Statements About Sets

There are four basic statements about the relationships between two sets. In this section we
will show you what these four statements are like. In subsequent sections, we will show you
what types of conclusions you can draw from them.
These statements represent basic types of information that you receive on the job. From this
information, you must draw conclusions, make decisions, and take actions. You want to be
sure that your conclusions, decisions, and actions are correct, based on the information you
have. We will show you what conclusions are justified from the four basic statements
below.

11
To illustrate the four statements, we will use an example of information that you might
receive about whether or not the new cars in your offices fleet were available for use in a
particular month. The four possible statements are as follows:

All of our new fleet cars are available for use this month.

None of our new fleet cars are available for use this month.

Some of our new fleet cars are available for use this month.

Some of our new fleet cars are not available for use this month.

To give you a picture of what these statements mean, we will use diagrams in which circles
represent complete sets. Figure 1a below is a possible representation of the statement All
of our new fleet cars are available for use this month. As you can see, all of the set our
new fleet cars is included in the set vehicles available for use this month. This captures
the essential information in the statement, namely, that every new fleet car is available for
use this month. Figure 1a, however, does not depict the only possible meaning of the
statement All of our new fleet cars are available for use this month. Figure 1b shows the
other possible meaning, namely, that the two sets -- our new fleet cars and vehicles that are
available for use this month -- overlap completely. If we only have the information All of
our new fleet cars are available for use this month, we do not know whether Figure 1a or 1b
is the correct representation of the relationship between the two sets.
In the rest of this manual, we will use a diagram like Figure 1a to represent this type of
statement. However, you should bear in mind that a diagram like Figure 1b might represent
the true relationship between the two sets.
Figure 1a. Diagram for All of our new
fleet cars are available for use this
month.

Figure 1b. Another Possible Meaning

of All of our new fleet cars are
available for use this month.

vehicles available
for use this month

vehicles available
for use this month
=
our new
fleet cars

our new
fleet cars

12
Figure 2 below represents the statement None of our new fleet cars are available for use
this month. As you can see, there is no overlap of the two sets. All of the set our new fleet
cars is outside all of the set vehicles available for use this month.
Figure 2. Diagram for None of our new fleet cars are available for use this month.

vehicles
available
for use this
month

our new
fleet cars

To illustrate the statement Some of our new fleet cars are available for use this month, we
will need a symbol to represent part of a set. We will use a semicircle to represent part of a
set. Just as the word some has a very indefinite meaning, so you should also consider the
semicircle to represent an undefined proportion of the set. In order to emphasize this, we
have drawn the semicircle with a dashed line on one side.
Figure 3 illustrates the statement Some of our new fleet cars are available for use this
month. The diagram shows that some of the new fleet cars are available for use this month,
but it does not give any information about the rest of the new fleet cars. That is to say, it
does not give any information about what proportion of the new fleet cars are available for
use this month.
Figure 3. Diagram for Some of our new fleet cars are available for use this month.

vehicles available
for use this month
our
new
fleet
cars

Figure 4 below illustrates the statement Some of our new fleet cars are not available for use
this month. As you can see, part of the set our new fleet cars is outside of the set
vehicles available for use this month. This diagram should be interpreted as not telling us

13
definitively if any of our new fleet cars are inside the circle for vehicles available for use
this month. We only know that some of them are outside the circle.
Figure 4. Diagram for Some of our new fleet cars are not available for use this month.

vehicles
available
for use this
month

our
new
fleet
cars

Self-Test: Section II.A.2 (answers are given on page 111)

Using circles to represent complete sets and semicircles to represent parts of sets, draw a
diagram to illustrate the relationship of the sets in each of the statements below.
1. Every supervisory training course is filled to capacity for the next three months.

2. Some of the cases being handled by this office are very old.

14
3. None of the old computers is equipped with Windows software.

4. A few computers on our system are not working.

Section II.A.3. Conclusions From All Are Statements

If you have information that one set is totally included in another, such as All of our new
fleet cars are available for use this month, there are several conclusions that you can draw
from that information and several conclusions that you cannot draw. We will discuss these
conclusions as true conclusions, false conclusions, and conclusions for which there is
insufficient information. True conclusions are the same as valid conclusions, but invalid
conclusions either can be false or can be conclusions for which there is insufficient
information.
True Conclusions
The following three important true conclusions can be drawn from the statement All of our
new fleet cars are available for use this month.

Some vehicles that are available for use this month are our new fleet cars.

Explanation: Given that all of our new fleet cars are available for use this month, it follows
that there are some vehicles that are available for use this month that are our new fleet cars.

15

Any car that is not available for use this month is not one of our new fleet cars.

Explanation: Given that all of our new fleet cars are available for use this month, it follows
that any car that is not available for use this month cannot be one of our new fleet cars. This
can be seen by referring back to Figure 1a, the diagram of All of our new fleet cars are
available for use this month. Clearly, anything that is outside of the set of vehicles that are
available for use this month is also outside the set of our new fleet cars.2

None of our new fleet cars is unavailable for use this month.

Explanation: This true conclusion represents a double negation of the original statement.
We have negated the verb of the sentence and we have negated the set vehicles available
for use this month. In other words, instead of saying All of our new fleet cars are vehicles
that are available for use this month, we are saying None of our new fleet cars are
unavailable for use this month. One negation cancels out the other. Therefore, the
sentence says the same thing as the original statement.3
False Conclusions
The following two conclusions are clearly false if the statement All of our new fleet cars
are available for use this month is true.

None of our new fleet cars are available for use this month.

Some of our new fleet cars are not available for use this month.

Explanation: These two conclusions are contrary to the meaning of the original statement.
A Conclusion for Which There Is Insufficient Information
The following conclusion is uncertain if the statement All of our new fleet cars are
available for use this month is true.

All of the vehicles that are available for use this month are our new fleet cars.

Explanation: This conclusion is uncertain because, although we know that the set of our
new fleet cars is completely included in the set of vehicles that are available for use this
month, we do not have enough information to know if the two sets are identical. The
conclusion above would only be true if the two sets were identical. You may recall that

Logic Note: This true conclusion represents a logical form called the contrapositive. The positions of the
two sets in the original statement are reversed and the two sets are negated. Another example is as follows:
For the statement All guns are weapons the true contrapositive would be Anything that is not a weapon is
not a gun.
3
Logic Note: This true conclusion represents a logical form known as the obverse.

16
Figures 1a and 1b in the previous section represented the two possible meanings of All are
statements. Without obtaining more information, we cannot tell which of these meanings is
correct, and thus we cannot draw the uncertain conclusion given above.
The relationship between the original statement and the uncertain conclusion is shown in the
following formulaic sentences. The original statement says:
All [members of set A] are [members of set B].
The uncertain conclusion says:
All [members of set B] are [members of set A].
You can see that the positions of the two sets are exchanged in the second sentence. In
logic, the second sentence is called the converse of the first sentence.
Later in this unit you will see that some basic statements do have valid converses -converses that mean the same thing as the original statement. It is important to know that
the converse of a statement such as All of our new fleet cars are available for use this
month cannot be concluded with certainty. However, it is very common for people to
assume that this conclusion is valid. This reasoning mistake is so common that it has been
called an illogical bias by authors who have studied the fallacies of human reasoning.
In the exercises that follow in the Self-Test, you will be asked to evaluate the correctness of
various conclusions. Be especially careful to try to avoid the illogical bias referred to above.
It is a bias you should always avoid, particularly when making inferences and decisions on
the job.
Self-Test: Section II.A.3 (answers are given on page 113)
In the questions below, you will be given a statement that is followed by four conclusions.
Given that the statement is true, you should indicate whether each conclusion is true (you
can infer the conclusion from the statement) or false (you can infer that the conclusion is
contrary to the statement), or if there is insufficient information to decide whether the
conclusion is true or false.
1. All members of our staff have mailboxes at the front desk.
a. Some members of our staff do not have mailboxes at the front desk. (T / F / I)
b. Some people who have mailboxes at the front desk are members of our staff.
(T / F / I)
c. All people who have mailboxes at the front desk are members of our staff. (T / F / I)

17
d. Anyone who does not have a mailbox at the front desk is not a member of our staff.
(T / F / I)
2. Every supervisory training course is filled to capacity for the next three months.
a. The only courses that are filled to capacity for the next three months are the
supervisory training courses. (T / F / I)
b. No supervisory training courses are filled to capacity for the next three months.
(T / F / I)
c. None of the supervisory training courses is filled to less than capacity for the next
three months. (T / F / I)
d. Any course in the next three months that is not filled to capacity is not a supervisory
training course. (T / F / I)

Section II.A.4. Conclusions From None Are Statements

A statement such as None of our new fleet cars are available for use this month indicates
that there is no overlap between the two sets in the statement. In other words, the first set,
our new fleet cars, is completely excluded from the set vehicles that are available for use
this month. Given this relationship between the two sets, there are two conclusions you can
draw as well as two conclusions you cannot draw. We will discuss these as true conclusions
and false conclusions.
True Conclusions
The following two true conclusions can be drawn from the statement None of our new fleet
cars are available for use this month.

No vehicle that is available for use this month is one of our new fleet cars.

Explanation: Since vehicles that are available for use this month are totally excluded from
the set of our new fleet cars, any car that is available for use this month cannot be one of our
new fleet cars. You may recognize this conclusion as having the form of the converse,
which is the form in which the positions of the two sets in the sentence are exchanged.
You may recall that the converse was not a valid conclusion from the statement All of our
new fleet cars are available for use this month. However, you can see that the converse is a
valid conclusion from the statement None of our new fleet cars are available for use this
month. It is valid because the two sets are totally separated from each other. Figure 2 on
page 12 illustrates the separation of the two sets.

All of our new fleet cars are unavailable for use this month.

18
Explanation: This true conclusion represents a double negation of the original statement,
None of our new fleet cars are available for use this month (All replaces None and
unavailable for use this month replaces available for use this month). It has the same
meaning as the original statement because the two negatives cancel each other out.4
When we negate a set, we can use the prefix non or another prefix, such as dis, im,
in, or un, or we can use other wording that indicates that the set is negated. For
example, if we were negating the set of proper things, we could call that set improper
things. In the example we have been using above, we used unavailable for use this
month to negate the set available for use this month.
False Conclusions
The following two conclusions are clearly false if the statement None of our new fleet cars
are available for use this month is true.

All of our new fleet cars are available for use this month.

Some of our new fleet cars are available for use this month.

Explanation: These two conclusions are contrary to the meaning of the original statement.
Self-Test: Section II.A.4 (answers are given on page 113)
In the questions below, you will be given a statement that is followed by four conclusions.
Given that the statement is true, you should indicate whether each conclusion is true (you
can infer the conclusion from the statement) or false (you can infer that the conclusion is
contrary to the statement), or if there is insufficient information to decide whether the
conclusion is true or false.
1. None of the new Jeeps are being deployed to Station 1.
a. Some of the new Jeeps are being deployed to Station 1. (T / F / I)
b. None of the vehicles being deployed to Station 1 is a new Jeep. (T / F / I)
c. All of the new Jeeps are among the vehicles that are not being deployed to Station 1.
(T / F / I)
d. All of the new Jeeps are being deployed to Station 1. (T / F / I)

Logic Note. You may recall from the previous section that this form is called the obverse.

19
2. None of the Assistant Chiefs will be able to attend the meeting.
a. All of the Assistant Chiefs will be able to attend the meeting. (T / F / I)
b. None of the people who will be able to attend the meeting are Assistant Chiefs.
(T / F / I)
c. Some of the people who will be able to attend the meeting are Assistant Chiefs.
(T / F / I)
d. All of the Assistant Chiefs will be unable to attend the meeting. (T / F / I)

Section II.A.5. Conclusions From Some Are Statements

A statement such as Some of our new fleet cars are available for use this month indicates
that there are at least some members of one set that are included in the other set. From this
information, there are two true conclusions that you can draw and one false conclusion that
you cannot draw. In addition, there are two conclusions that you cannot draw because you
have insufficient information.
True Conclusions
The following two true conclusions can be drawn from the statement Some of our new fleet
cars are available for use this month.

Some vehicles that are available for use this month are our new fleet cars.

Explanation: Since some of our new fleet cars are available for use this month, it is clear
that there are some vehicles that are available for use this month that are our new fleet cars.
Perhaps you recognize that this form is the converse, the form in which the positions of the
two sets in the sentence are exchanged. The converse is a valid conclusion from a statement
that some members of one set are members of another set.

Some of our new fleet cars are not unavailable for use this month.

Explanation: This true conclusion represents a double negation of the original statement
Some of our new fleet cars are available for use this month. It has the same meaning as
the original statement because the two negatives cancel each other out.5

Logic Note. You may recall that this form is called the obverse.

20
False Conclusion
The following conclusion is clearly false if the statement Some of our new fleet cars are
available for use this month is true.

None of our new fleet cars are available for use this month.

Explanation: This conclusion contradicts the meaning of the original statement.

Conclusions for Which There Is Insufficient Information
The two conclusions discussed below are uncertain because the statement Some of our new
fleet cars are available for use this month does not give us sufficient information to know if
they are true or not.

All of our new fleet cars are available for use this month.

Explanation: This conclusion is uncertain because the original statement Some of our new
fleet cars are available for use this month refers to just part of the set of new fleet cars.
Therefore, strictly speaking, we do not have enough information to draw a conclusion about
all of the set. Of course, sometimes we use the initial statement to mean that some -- but not
all -- members of one set are members of another set. Therefore, when you interpret a
statement about some parts or members of a set, you should be cautious about whether or
not the statement is based on knowledge about part of the set or about all of the set. The
next uncertain conclusion illustrates the same point.

Some of our new fleet cars are not available for use this month.

Explanation: This is an uncertain conclusion because saying that there are some new fleet
cars that are available for use this month does not itself imply that there are some that are
not. For example, someone might have seen several of the new fleet cars being used and
might report to you (happily), Some new cars are available for use! Jumping to the
conclusion that the rest of the cars in the fleet are not available would obviously be mistaken
-- in fact, this type of mistake is called an illogical bias because it is a common mistake in
human reasoning. On the job, these illogical biases can be very costly, at best in terms of
monetary cost or, at worst, in terms of human life.
Of course, if you made a statement such as, I have inspected the records on all the new fleet
cars and only some of them are available for use this month, then it would be clear that you
know that some of the new cars are unavailable for use.

21
Self-Test: Section II.A.5. (answers are given on page 114)
In the questions below, you will be given a statement that is followed by four conclusions.
Given that the statement is true, you should indicate whether each conclusion is true (you
can infer the conclusion from the statement) or false (you can infer that the conclusion is
contrary to the statement), or if there is insufficient information to decide whether the
conclusion is true or false.
1. Some DHS employees are trained in the use of firearms.
a. Some people who are trained in the use of firearms are DHS employees. (T / F / I)
b. All DHS employees are trained in the use of firearms. (T / F / I)
c. No DHS employees are trained in the use of firearms. (T / F / I)
d. Some DHS employees are not untrained in the use of firearms. (T / F / I)
2. Some Canadian citizens are eligible to work in the United States.
a. Some Canadian citizens are ineligible to work in the United States. (T / F / I)
b. Some Canadian citizens are not ineligible to work in the United States. (T / F / I)
c. All Canadian citizens are eligible to work in the United States. (T / F / I)
d. Some people who are eligible to work in the United States are Canadian citizens.
(T / F / I)
Section II.A.6.

Conclusions From Some Are Not Statements

If you have information that at least some part of a set is excluded from another set, then
there is one true conclusion you can draw and one conclusion that you know is false. There
are three other conclusions that are uncertain because you have insufficient information.
True Conclusion
The one true conclusion you can draw from the statement Some of our new fleet cars are
not available for use this month is the following:

Some of our new fleet cars are unavailable for use this month.

Explanation: This is true because not being available for use this month is synonymous with
being unavailable for use this month. If you are interested in studying this form more

22
deeply, you can see that this form is actually a double negation of the original statement.6
The original statement says:
Some [members of set A] are not [members of set B]
The true conclusion says:
Some [members of set A] are [non-members of set B]
The original verb are not was negated and the second set was negated. As you can see,
and as you may recall from Section II.A.4, the double negative of the verb canceled itself
out and the meaning of the original statement was retained.
False Conclusion
The following statement is false if the initial statement Some of our new fleet cars are not
available for use this month is true.

All of our new fleet cars are available for use this month.

Explanation: This conclusion contradicts the information in the initial statement.

Conclusions for Which There Is Insufficient Information
The three conclusions discussed below are all uncertain when an initial statement such as
Some of our new fleet cars are not available for use this month is true. They are all
uncertain because the initial statement only gives us information about some of the new fleet
cars.

None of our new fleet cars is available for use this month.

Explanation: We are not warranted in drawing a conclusion about the entire set of new fleet
cars if we only have information about some of them.

Some vehicles that are available for use this month are not our new fleet cars.

Explanation: This conclusion cannot be legitimately inferred from the initial statement
because the initial statement does not give any information about the vehicles that are
available for use this month. In fact, the initial statement does not even imply that there are
vehicles available for use this month.
You might have noticed that the uncertain conclusion above represents the converse of the
initial statement. That is to say, the order of two sets was exchanged in drawing the

Logic Note. This form is called the obverse.

23
conclusion. This particular conclusion is another powerful illogical bias, another common
reasoning mistake that you should try to avoid.

Some of our new fleet cars are available for use this month.

Explanation: This conclusion is uncertain, given that Some of our new fleet cars are not
available for use this month, because the initial statement has only told us about some new
fleet cars. It has not said anything about whether any of the rest of the new fleet cars are
available for use this month. It does not presume to give information about all of the new
fleet cars. This type of conclusion is another illogical bias which you should try to avoid in
drawing conclusions from partial information.
If the initial statement had been something like, I have inspected the records on all of the
new fleet cars and only a few are not available for use this month, then you would be
justified in concluding that there are some fleet cars that are available for use this month.
Many mistakes based on illogical biases have been made that have ruined budgets or
endangered human life. The following example shows this type of mistake: Some areas
with rough terrain are not safe to land a helicopter; therefore, some other areas with rough
terrain are safe to land a helicopter.
Self-Test: Section II.A.6 (answers are given on page 115)
In the questions below, you will be given a statement that is followed by four conclusions.
Given that the statement is true, you should indicate whether each conclusion is true (you
can infer the conclusion from the statement) or false (you can infer that the conclusion is
contrary to the statement), or if there is insufficient information to decide whether the
conclusion is true or false.
1. This morning I received some documents that were not in acceptable form.
a. This morning, all of the documents I received were in acceptable form. (T / F / I)
b. This morning I received some documents that were in acceptable form. (T / F / I)
c. Some documents I received this morning were in unacceptable form. (T / F / I)
d. None of the documents I received this morning were in acceptable form. (T / F / I)

24
2. Some computers in our office cannot run this new software.
a. Some computers that can run this new software are not in our office. (T / F / I)
b. None of the computers in our office can run this new software. (T / F / I)
c. Some computers in our office can run this new software. (T / F / I)
d. All of the computers in our office can run this new software. (T / F / I)

Section II.A.7. Conclusions When the Basic Statements Are False

In the preceding four sections, you have learned about conclusions you can draw when one
of the four basic statements is true. However, there are also conclusions you can draw if one
of the statements is false. Recognizing these conclusions will further expand your deductive
reasoning powers. In this section, you will learn these conclusions for all four of the basic
statements.
At first you will find this section easy and you will think it is repetitive relative to the
previous section, but as you read the material you will realize inferences from statements
that are false can be tricky and that it is very easy to lapse into reasoning errors. Because
these inferences occur constantly in real-life situations, it is important that you acquire
mastery in handling them.
Conclusions When an All Are Statement Is False
A true conclusion. When a statement such as All of our new fleet cars are available for use
this month is false, the following conclusion must be true.

Some of our new fleet cars are not available for use this month.

Explanation: You can probably see that if it is not true that all of our new fleet cars are
available for use this month, then it must be true that at least some of the cars are not
available for use this month.
The following is another example:
Initial statement: It is not true that all aliens who have proper documents are
admissible.
True conclusion: Some aliens who have proper documents are inadmissible.

25
Two uncertain conclusions. There are two conclusions that are uncertain when a statement
such as All of our new fleet cars are available for use this month is known to be false.
These two conclusions are:

None of our new fleet cars are available for use this month.

Some of our new fleet cars are available for use this month.

Explanation: These conclusions are uncertain because knowing that it is not true that all of
our new fleet cars are available for use this month does not give you enough information to
conclude that none of them are or that some of them are.
Conclusions When a None Are Statement Is False
A true conclusion. When a statement such as None of our new fleet cars are available for
use this month is false, then the following conclusion is true:

Some of our new fleet cars are available for use this month.

Explanation: If it is not true that none of the fleet cars are available for use this month, it
must be true that some of them are available for use this month.
Two uncertain conclusions. The following two conclusions are uncertain if it is not true
that none of our new fleet cars are available for use this month:

All of our new fleet cars are available for use this month.

Some of our new fleet cars are not available for use this month.

Explanation: These conclusions are uncertain because simply knowing that it is not true that
none of the new fleet cars are available for use this month does not give you enough
information to know if all of the cars are available for use this month or if some of them are
not available for use this month.
Conclusions When a Some Are Statement Is False
A true conclusion. When a statement such as Some of our new fleet cars are available for
use this month is false, then the following conclusion must be true:

None of our new fleet cars are available for use this month.

Explanation: When we say it is not true that some of our new fleet cars are available for
use this month, we are saying that there are not some fleet cars that are available for use
this month; in other words there are no fleet cars that are available.

26
The diagrams in Figure 5a and 5b may help to illustrate what we mean when we say that it
is not true that some of our new fleet cars are available for use this month. Figure 5a shows
you the meaning of Some of our new fleet cars are available for use this month. Figure
5b shows that the same statement is false. To indicate this, we have crossed out the
semicircle that indicates fleet cars that are available for use this month. This indicates that
there are none of our fleet cars in the possible set of vehicles that are available for use this
month.
Figure 5a. Diagram for Some of our
new fleet cars are available for use this
month.

Figure 5b. Diagram for It is not true

that some of our new fleet cars are
available for use this month.

vehicles available
for use this month

vehicles available
for use this month

our
new
fleet
cars

our
new
fleet
cars

To help you understand the relationship between the falsity of a Some are statement and
the truth of a None are statement, let us look at two more examples.
Initial statement: It is not true that some of these documents are counterfeit.
True conclusion: None of these documents are counterfeit.
Initial statement: It is not true that some of the new computers are laptops.
True conclusion: None of the new computers are laptops.
Both of the initial statements say that it is not true that some members of the first set are
members of the second set. The true conclusion says that therefore no members of the first
set are members of the second set.
A false conclusion. The following conclusion is false, when the statement Some of our
new fleet cars are available for use this month is false:

All of our new fleet cars are available for use this month.

Explanation: If it is not true that some of the new fleet cars are available for use this month,
then it certainly cannot be true that all of the cars are available for use this month.

27
Conclusions When a Some Are Not Statement Is False
A true conclusion. When a statement such as Some of our new fleet cars are not available
for use this month is false, then the following conclusion must be true:

All of our new fleet cars are available for use this month.

Explanation: When we say it is not true that some of our new fleet cars are not available
for use this month, we mean that there are not any fleet cars that are not available for use
this month. In other words, we are saying that all of our fleet cars are available for use this
month. The diagrams in Figure 6a and Figure 6b may help to illustrate what we mean when
we say that it is not true that some of our new fleet cars are not available for use this
month. Figure 6a shows you the meaning of Some of our new fleet cars are not available
for use this month. Figure 6b shows that the same statement is false. To indicate this, we
have crossed out the semicircle that indicates fleet cars that are not available for use this
month. This indicates that there are no fleet cars outside the set of vehicles that are available
for use this month.
Figure 6a. Diagram for Some of our
new fleet cars are not available for use
this month.

vehicles
available
for use this
month

our
new
fleet
cars

Figure 6b. Diagram for It is not true

that some of our new fleet cars are not
available for use this month.

vehicles
available
for use this
month

our
new
fleet
cars

Two more examples are given below to help ensure that you understand the relationship
between the falsity of a Some are not statement and the truth of an All are statement.
Initial statement: It is not true that some of the offices do not have windows.
True conclusion: All of the offices have windows.
Initial statement: It is not true that some employees did not receive the notice.
True conclusion: All employees received the notice.
Both of the initial statements say that it is not true that some members of the first set are not
members of the second set. The true conclusion says that therefore all members of the first
set are members of the second set.

28
A false conclusion. The following conclusion is false:

None of our new fleet cars are available for use this month.

Explanation: If it is not true that some of the new fleet cars are not available for use this
month, then it certainly cannot be true that none of the cars are available for use this month.
Self-Test: Section II.A.7 (answers are given on page 116)
In the questions below, you will be given a statement that is followed by four conclusions.
Given that the statement is false, you should indicate whether each conclusion is true (you
can infer the conclusion from the statement) or false (you can infer that the conclusion is
contrary to the statement), or if there is insufficient information to decide whether the
conclusion is true or false.
1. This statement is false: None of the computers in our office can run the new software.
a. All of the computers in our office can run the new software. (T / F / I)
b. Some of the computers in our office can run the new software. (T / F / I)
c. Some of the computers in our office cannot run the new software. (T / F / I)
2. This statement is false: Some of the employees in our office received awards this year.
a. All of the employees in our office received awards this year. (T / F / I)
b. None of the employees in our office received an award this year. (T / F / I)
3. This statement is false: All of the passengers boarding the plane have proper
documentation of U.S. citizenship.
a. None of the passengers boarding the plane have proper documentation of U.S.
citizenship. (T / F / I)
b. Some of the passengers boarding the plane have proper documentation of U.S.
citizenship. (T / F / I)
c. Some of the passengers boarding the plane do not have proper documentation of U.S.
citizenship. (T / F / I)

29
4. This statement is false: Some DHS managers have not completed the new Management
Issues training course.
a. All DHS managers have completed the new Management Issues training course.
(T / F / I)
b. No DHS managers have completed the new Management Issues training course.
(T / F / I)

Part II.B.

Reasoning with Three Sets

In this part, you will learn about reasoning with three sets by building on what you have
already learned about statements that relate two sets. This part, as well as all of Units II and
III in this manual, is very important for successful performance in your job. In performing
your job, you are frequently called upon to carry out reasoning and decision making relative
to three (or more) sets. The study of this part will give you the necessary skills for doing so
successfully.

Section II.B.1. Forms for Relating Three Sets

The following example shows how two statements can be used together to draw a new
conclusion about the relationship between two sets. You will see in Example 1 that the two
statements contain a total of three sets, one of which is contained in both statements. In this
and subsequent examples, the statements will be called premises. A premise is any evidence
that is used to support a conclusion.
Example 1
Premise 1:
Premise 2:
Conclusion:

All recently hired employees are very well qualified.

All of our trainees are recently hired employees.
All of our trainees are very well qualified.

The set that is common to both premises is recently hired employees. That set can be
called the term of comparison, although in logic it has traditionally been called the middle
term.
This reasoning form, which is sometimes called a syllogism, allows you to describe the
relationship between the set that is found only in the second premise (trainees) and the set
that is found only in the first premise (very well qualified [individuals]).
In the example given above, the set trainees is completely included in the set recently
hired employees, which itself is completely included in the set very well qualified
[individuals]. Thus, the set trainees is completely included in the set very well qualified
[individuals].

30

Figure 7 below illustrates the relationships among the sets in Example 1.

Figure 7. Diagram for Example 1.
very well
qualified individuals
recently hired
employees
trainees

The next example shows you a similar reasoning form which contains a negative premise.
Example 2
Premise 1:
Premise 2:
Conclusion:

No recently hired employee is certified in CPR.

All trainees are recently hired employees.
No trainee is certified in CPR.

In this example, the middle term is once again recently hired employees. The first
statement says that this set is completely excluded from the set of people who are certified
in CPR. Since all trainees are recently hired employees, they are all excluded from the set
of people who are certified in CPR. Figure 8 below illustrates the relationships among the
three sets in Example 2.
Figure 8. Diagram for Example 2.

certified in
CPR
recently hired
employees
trainees

31
In Examples 1 and 2, the middle term was the subject of the first premise and the predicate
term of the second premise. There are three other possible positions for the middle term,
and these positions help determine whether or not a conclusion can be reached from the pair
of premises. In the next example, the middle term is the predicate term of both premises.
You will see that no conclusion can be reached about the relationship between the other two
terms.
Example 3
Premise 1:
All of our analysts are recently hired employees.
Premise 2:
All of our trainees are recently hired employees.
Conclusion: No conclusion is possible about the relationship between the sets
trainees and analysts.
The first premise says that the set of analysts is completely included in the set of recently
hired employees. The second premise says that the set of trainees is also completely
included in the set of recently hired employees. However, we have no other information
about the set of recently hired employees; it could be very large compared to the other two
sets. This leaves open the possibility that all, some, or none of the trainees are analysts.
Figure 9a and 9b below show two possible ways the premises could be represented. The
sets of trainees and psychologists might be completely separate, as shown in Figure 9a; or,
they might be overlapping, as shown in 9b. Another possibility, not illustrated, is that the
two sets overlap completely. We do not have enough information to know which is the
correct representation.
Figure 9a. A Possible Representation
of Example 3.

recently hired
employees

trainees

analysts

Figure 9b. Another Possible

Representation of Example 3

recently hired
employees

trainees
analysts

Of all the possible combinations of the basic statements (there are 64 of them), only 15
combinations permit a valid conclusion to be drawn. Therefore, when you come across an

32
example of reasoning in which two sets are compared to a third set, it is important to
examine the form of the reasoning very carefully to be sure that the conclusion is valid.
Examples 4, 5, and 6 give you more examples of valid syllogisms.
Example 4
Premise 1:
Premise 2:
Conclusion:

All agents from this station have been issued the new pistol.
No one at the firing range today has been issued the new pistol.
No one at the firing range today was an agent from this station.

From the information in Premise 2, we can conclude that no one who has been issued the
new pistol was at the firing range today.7 Since Premise 1 tells us that all agents from this
station have been issued the new pistol, we can conclude that no one at the firing range
today was an agent from this station. Figure 10 below represents the relationships among
the sets in Example 4.
Figure 10. Diagram for Example 4.

at firing range
issued
new pistol
agents
from this
station

Example 5
Premise 1:
Premise 2:
Conclusion:

All agents from this station have been issued the new pistol.
Some agents from this station are new agents.
Some new agents have been issued the new pistol.

From the information in Premise 2, we can conclude that some new agents are agents from
this station.8 Therefore, based on Premise 1, we can conclude that these new agents have

Logic Note. You may recall from Section II.A.4 that it is valid to exchange the position of the sets in a
statement of the form No [members of set A] are [members of set B]. This is called the converse.
8
Logic Note. You may recall from Section II.A.5 that it is valid to exchange the position of the sets in a
statement of the form Some [members of set A] are [members of set B]. This is again called the converse.

33
been issued new pistols. Figure 11 below represents the relationships among the sets in
Example 5.
Figure 11. Diagram for Example 5.
issued
new pistol
agents from
this station
new
agents

Example 6 is the last example of a syllogism in this section.

Example 6
Premise 1:
Premise 2:
Conclusion:

Some of our trainees are not college graduates.

All of our trainees are recently hired employees.
Some recently hired employees are not college graduates.

From the second premise we can conclude that anyone who is one of our trainees is also one
of our recently hired employees. Therefore, when the first premise says that some of our
trainees are not college graduates, it is also implying that some of our recently hired
employees are not college graduates.
It would be impossible to diagram this syllogism using the type of diagrams that we have
been using.9 If you try to do it, you will find that you run into trouble picturing the set our
trainees, because you have to picture all of the set as being included in a second set and
part of the set as being excluded from a third set.
Why is it useful to know about valid syllogisms?
Valid syllogisms represent useful reasoning forms that enable you to draw new conclusions
when you have certain information about three sets. Knowing about syllogisms will help
you to build your own arguments and to evaluate the arguments that other people make. As

Logic Note. There is another type of diagrams, called Venn diagrams, that could be used for this syllogism.
These can be found in any Logic book.

34
we said in the beginning of Part II.B, in real-life job performance you are frequently called
upon to carry out reasoning and decision making involving three (or more10) sets.
The next section is an optional section that gives you logical rules that you can use to decide
whether an argument about three sets is valid or not. In Section II.B.3 you will build on
your knowledge by learning to avoid illogical biases in relating three sets.
Self-Test: Section II.B.1 (answers are given on page 117)
For each pair of statements below, underline the middle term and then write a valid
conclusion relating the other two terms.
1.

Premise 1: All DHS employees are Federal Government employees.

Premise 2: All ICE employees are DHS employees.
Conclusion:

2.

Premise 1: No Canadian citizens are U.S. citizens.

Premise 2: All citizens of Quebec are Canadian citizens.
Conclusion:

3.

Premise 1: No District 1 staff has completed the new training.

Premise 2: All recently hired adjudicators have completed the new training.
Conclusion:

4.

Premise 1: Some agents are trained in Advanced CPR.

Premise 2: All employees trained in Advanced CPR passed the Basic CPR course.
Conclusion:

5.

Premise 1: All ICE vehicles are listed in the vehicle database.

Premise 2: Some DHS vehicles are not ICE vehicles.
Conclusion:

10

Logic Note. When you have more than three sets, you can make an inference by developing a chain of
syllogisms in which the conclusion of the first syllogism would become the first premise of the second.
Logicians call this form of reasoning sorites.

35
6.

Premise 1: All persons born in the U.S.A. are U.S. citizens.

Premise 2: All persons born in the U.S.A. can get social security cards.
Conclusion:

Section II.B.2. Advanced Topic -- Syllogistic Rules (reading this section is optional)
Applying the rules in this section will allow you to decide quickly whether or not an
argument relating three sets permits a valid conclusion. The first three rules are very easy to
apply. The fourth and fifth rules require additional explanation, because they introduce the
new concept of a distributed term.
1.

If both premises are negative, there is no conclusion. (If both premises are negative, the
three sets are completely separate and no further conclusion can be drawn.)

2.

If one premise is negative, the conclusion is negative. (If one premise is negative, the
essence of the syllogism is to disconnect one set from another through its disconnection
from the middle term.)

3.

If neither premise is negative, the conclusion is affirmative. (If both premises are
affirmative, there is no reason to conclude anything negative.)

4.

The middle term must be distributed at least once in order for there to be a valid
conclusion. (See discussion immediately below.)

5.

No term may be distributed in the conclusion if it is not distributed in the premises. (See
discussion immediately below.)

When we say a term is distributed in a premise, we mean that the premise says something
about the entire set represented by the term.

In the statement All [members of set A] are [members of set B], set A is distributed
because the statement says that the entire set A is included in set B. That is to say,
the statement refers to set A in its entirety.

In the statement No [members of set A] are [members of set B], set A and set B are
both distributed because all of set A is excluded from all of set B.

In the statement Some [members of set A] are [members of set B], neither term is
distributed because we only know about part of set A and part of set B.

In the statement Some [members of set A] are not [members of set B], the set B is
distributed because the statement says that there are some members of set A that are
excluded from all of set B. (This explanation is a little hard to understand, but it
becomes clear when pictured in a diagram. There, the portion of set A is drawn
outside the entire set B; hence, it is clearer that we are referring to the whole of set
B.)

36
From the five basic rules, two more very useful rules can be derived:
6.

If both premises refer to only part of a set, there is no conclusion. (It is impossible to
relate the parts of the sets that are referred to by the vague quantifier some.)

7.

If one premise refers to only part of a set, the conclusion refers to only part of a set. (If
one premise refers to part of a set, the conclusion cannot generalize legitimately beyond
that.)

To gain further appreciation of these rules, we can apply them to some of the syllogisms that
have already been presented in Section II.B.1. The syllogism in Example 1 has the
following form (with M representing the term that is common to the two premises):
Premise 1: All [members of set M] are [members of set B]
Premise 2: All [members of set A] are [members of set M]
Conclusion: All [members of set A] are [members of set B]
This syllogism has two affirmative premises, so the conclusion must be affirmative (rule 3).
The middle term M is distributed in the first premise (rule 4). Set A is distributed in the
conclusion and in the second premise (rule 5).
Now let us apply the rules to Example 3 in Section II.B.1, which does not have a valid
conclusion. This syllogism has the following form:
Premise 1: All [members of set B] are [members of set M]
Premise 2: All [members of set A] are [members of set M]
Invalid Conclusions: All [members of set A] are [members of set B]
Some [members of set A] are [members of set B]
Since both premises are affirmative, the syllogism would have to have an affirmative
conclusion (rule 3). All [members of set A] are [members of set B] and Some [members of
set A] are [members of set B] are therefore possible conclusions. However, the middle term
M is not distributed in either premise (violation of rule 4). Therefore, the syllogism does not
have a valid conclusion.
The Self-Test will give you practice in applying these rules.
Self-Test: Section II.B.2 (answers are given on page 118)
In exercises 1 through 4, indicate which syllogisms are valid. For valid syllogisms, decide
which syllogistic rules apply. For invalid syllogisms, decide which syllogistic rules are
violated.
1.
Premise 1: All ICE Agents are Department of Homeland Security employees.
Premise 2: All ICE Agents are Federal Government employees.
Conclusion: All Federal Government employees are Department of Homeland
Security employees.

37
(valid / invalid)
If valid, which rules apply? If invalid, which rules are violated?
2.

Premise 1: All drug smugglers are violators of anti-drug laws.

Premise 2: No Station 2 detainees are violators of anti-drug laws.
Conclusion: No Station 2 detainees are drug smugglers.
(valid / invalid)
If valid, which rules apply? If invalid, which rules are violated?

3.

Premise 1: Some contraband seized was seized by DHS.

Premise 2: All contraband seized becomes property of the U.S. Government.
Conclusion: Some property of the U.S. Government is contraband seized by the
DHS.
(valid / invalid)
If valid, which rules apply? If invalid, which rules are violated?

4.

Premise 1: All illegal immigrants are under the jurisdiction of the DHS.
Premise 2: Some deported persons are illegal immigrants.
Conclusion: Some deported persons are not under the jurisdiction of the DHS.
(valid / invalid)
If valid, which rules apply? If invalid, which rules are violated?
Section II.B.3. Illogical Biases in Relating Three Sets

It has been shown in numerous studies that people make many mistakes when reasoning
with three sets. We could say that people have many illogical biases in this type of
reasoning. The purpose of this section is to show you the most prevalent of these biases so
that you can avoid them yourself.
One of the common mistakes of reasoning is to think that a valid conclusion can be drawn
from premises that do not warrant a valid conclusion. For example, in the following two
cases no conclusion can be drawn, although people tend to think otherwise.
Premise 1: No biased persons make fair decisions.
Premise 2: No adjudicators are biased.
Invalid Conclusion: No adjudicators make fair decisions.
Premise 1: Some detainees were deported.
Premise 2: Some deported persons came from overseas.
Invalid Conclusion: Some persons who came from overseas were detainees.
If you read the Advanced Topic on syllogistic rules in the previous section, you will see that,
in the first example, no conclusion can be drawn because both premises are negative and, in
the second example, no conclusion can be drawn because both premises refer only to part of
a set. If you construct diagrams for these two forms, you will appreciate graphically the

38
invalidity of these forms. In the first form, the sets are totally disconnected from one
another, thus allowing no conclusion. In the second form, the parts of the set denoted by the
vague quantifier some are undefined and hence cannot be related to one another.
Another common error is to draw a conclusion about an entire set when a conclusion is
warranted about only part of a set. The next two examples show this type of error.
Premise 1: All primary inspections are conducted by one inspector.
Premise 2: All primary inspections are brief.
Invalid Conclusion: All brief inspections are conducted by one inspector.
Valid Conclusion: Some brief inspections are conducted by one inspector.
The first conclusion is invalid because Premise 2 does not give us information about all brief
inspections. (As you will recall from your study of the converse in Section II.A.3, you
cannot conclude that All brief inspections are primary inspections from the statement that
All primary inspections are brief.) We can conclude from that premise only that some
brief inspections are primary inspections. Therefore, we can only conclude that some
brief inspections are conducted by one inspector.
Premise 1: No Texans live in Florida.
Premise 2: All Floridians live in the U.S.A.
Invalid Conclusion: No persons who live in the U.S.A. are Texan.
Valid Conclusion: Some persons who live in the U.S.A. are not Texan.
The invalid conclusion to this syllogism is based on the same type of reasoning error that
characterized the previous invalid syllogism. From the statement that All Floridians live in
the U.S.A. (Premise 2), you can only conclude that Some people who live in the U.S.A.
are Floridians. Therefore, you can only conclude that Some people who live in the U.S.A.
(the Floridians) are not Texans.
What is the significance of these illogical biases?
These illogical biases serve as a caution to you (and to all of us) to be very careful when
drawing conclusions about the relationships among sets. An incorrect conclusion may
feel right at first and you may be swayed into faulty conclusions that will affect the
integrity of your decision making on the job. The antidote to illogical biases is good
reasoning: your goal is to become a top reasoner by fine-tuning your skills through the use
of this manual.
Self-Test: Section II.B.3 (answers are given on page 119)
For each syllogism below, indicate if the conclusion is valid (true). If the conclusion is not
valid (that is, if it is false or if there is insufficient information), determine if there is a
different, valid conclusion.

39
1.

Premise 1: No Special Agents failed to qualify with their firearm last week.
Premise 2. All employees who failed to qualify with their firearm last week must
qualify this week.
Conclusion: No employees who have to qualify with their firearm this week are
Special Agents.
(valid / invalid)
If applicable, valid conclusion:

2.

Premise 1: All requests for computer assistance must be made to the Help Desk.
Premise 2: All Help Desk requests are entered into the service system.
Conclusion: All entries into the service system are requests for computer assistance.
(valid / invalid)
If applicable, valid conclusion:

3.

Premise 1: All furniture in the ABC building is the property of DHS.

Premise 2: Some of this departments furniture is in the ABC building.
Conclusion: Some of this departments furniture is the property of DHS.
(valid / invalid)
If applicable, valid conclusion:

4.

Premise 1: Some space on this floor belongs to the Logistics office.

Premise 2: Some space of the Logistics office is assigned to Jim.
Conclusion: Some space assigned to Jim is on this floor.
(valid / invalid)
If applicable, valid conclusion:

Part II.C. Reasoning with Connectives

Reasoning with connectives is something that you do everyday. For example, when you use
a sentence such as If it rains tomorrow, then I will not wash the car, you are using a
connective sentence called the conditional. Similarly, the statement People who are
eligible for this job have a college degree or three years of responsible experience is a
connective sentence called an alternation.
Connectives are words such as and, either...or, and if...then that express relationships
between simple statements.
These simple statements express possible or actual
circumstances such as it rains tomorrow, I will not wash the car, and you have a
college degree.

40
The Federal Government workplace, including the law enforcement professions, requires
employees to reason with connectives constantly. Frequently, these connectives are used in
very complex combinations, such as those found in laws pertaining to immigration and
naturalization.
The purpose of this part of the training is to make you aware of the connectives that you use
everyday and of what their implications are. This training should ensure that you use
connectives correctly in your reasoning, especially when you are dealing with complex
relationships. It will help you appreciate when work situations express a relationship
embodied in one of the connectives, such as an if...then situation or an either...or
situation. A good knowledge of connectives will help you formulate problems and decisions
correctly so that you can anticipate the consequences of your choices and actions in the
performance of your job.
We will begin by talking about the simple connective called conjunction and then return to
the subject of negation, which we first discussed with reference to reasoning with two sets.

Section II.C.1. Conjunction and Negation

Conjunction
Conjunction conveys the meaning of the English word and; it says that two or more things
occur together. The following sentences contain conjunctions:

Everyday I open the safe in the morning and she locks it at close of business.

You will receive on-the-job training in addition to classroom training.

The Attorney General and the Commissioner will attend the ceremony.

Notice that in the second and third examples, the conjunction joins two nouns (for example,
on-the-job training and classroom training) rather than two simple statements. Implicitly,
however, both of these sentences contain two simple sentences. For example, the second
sentence could be rewritten (somewhat inefficiently) as:

You will receive on-the-job training and you will receive classroom training.

The third sentence could be rewritten as:

The Attorney General will attend the ceremony and the Commissioner will attend
the ceremony.

41
It should be mentioned here that a conjunction can be used in the same sentence with
another connective to make a more complex sentence, such as the following:

If you complete both classroom training and on-the-job training, then you will have
completed the basic requirements.

Notation for connectives. In order to help in the rest of our discussion, we will introduce
some notation from the logic of connectives. In formal logic, simple statements are
represented by lower-case letters, such as p, q, and r. These letters represent the statements
that are joined together by connectives. The following table shows how some of our earlier
examples of conjunctions can be represented.
Connective

Statement
I open
morning

Connective
the safe in the
and

p
The Attorney General will
attend the ceremony
p

Statement
she locks the safe at close of
business
Q

and

the Commissioner will attend

the ceremony
Q

then
you will have completed the
you complete classroom
basic requirements
training and* you complete
on-the-job training
If
p and* q
R
*Note: The word and is an embedded connective; that is to say, it is embedded within the
primary connective if...then.
If

Negation
The correct use of negation is an essential part of reasoning with connectives. You can
negate a simple statement or a complex statement that contains connectives. In all cases, the
negation means It is not true that or It is not the case that. Negating a sentence simply
denies what the sentence says.
The following three examples show you how negation is used with sentences of different
levels of complexity. The examples also show how the notation is used.

Negate a simple sentence: It is not true that I open the safe in the morning. not p

Negate a conjunction: It is not the case that both the Attorney General and the
Commissioner will attend the ceremony. not (p and q)

42

Negate a complex connective sentence: It is not true that if you complete classroom
training and on-the-job training, you will have finished the training program.
not (if p and q, then r)

Self-Test: Section II.C.1 (answers are given on page 119)

1. Which of the following sentences expresses a conjunction? For those that express a
conjunction, underline the phrases or statements that are joined.
a. The Department of Homeland Security both confers benefits and enforces laws.
(conjunction / not a conjunction)
b. Aliens can enter the U. S. as either immigrants or nonimmigrants.
(conjunction / not a conjunction)
c. To be successful, a teletraining instructor must maintain eye contact with the
viewers and must involve viewers in interactive exercises.
(conjunction / not a conjunction)
d. A complete computer set-up includes a processor, a display, a keyboard, and a
mouse.
(conjunction / not a conjunction)
e. If the annual fire drill is conducted tomorrow, our conference will be interrupted.
(conjunction / not a conjunction)
2. Write the negation of each of the following sentences in the space provided below.
a. The new vehicles have four-wheel drive.

b. Agents Lpez and Johnson are on annual leave today.

c. The agency will use both classroom training and computer-based training in the
future.

d. Either you or I will attend the meeting.

e. If it rains, the ceremony will be canceled.

43
Section II.C.2. The Conditional
The conditional is the most important type of connective sentence. It represents
circumstances that are found everywhere in life. In fact, if after studying this section you
take a look around your home or office, you will find many implicit and explicit examples of
the conditional.
A conditional sentence has the form If p then q. Two examples of conditional sentences
are:

If you are an employee of DHS, then you are a Federal employee.

If there is a credible bomb threat to an office building, then the building is

evacuated until a search is completed.

In the first sentence, the statement you are an employee of DHS would be represented by
the letter p, and the statement you are a Federal employee would be represented by the
letter q. In the second sentence, the statement there is a credible bomb threat to an office
building would be represented by p and the building is evacuated until a search is
completed would be represented by q.
In a conditional sentence, the first simple statement (the one represented by p) is known as
the antecedent. The second simple statement (represented by q) is known as the consequent.
(The table below shows the parts of a conditional sentence.) The conditional sentence says
that if the antecedent is true, then the consequent must also be true. Another way of
expressing this is to say that the antecedent cannot be true if the consequent is false.
The parts of a conditional sentence
Antecedent

Consequent

[If] you are an employee of DHS,

[then] you are a Federal employee.

[If] there is a credible bomb threat to

an office building,

[then] the building is evacuated until a search

is completed.

To understand the meaning of the conditional, it may help to consider some examples based
on the conditional sentence If you are an employee of DHS, then you are a Federal
employee.

Suppose we had the information that Pat is an employee of DHS. Taking that
information into account with the conditional, we could conclude that Pat is a
Federal employee. (Symbolic version: If p then q; p is true, therefore q is true.)
The entire formula is:

44

If you are an employee of DHS, then you are a Federal employee.

Pat is an employee of DHS.
Therefore, Pat is a Federal employee.

Suppose we had the information that Lee is not a Federal employee. Taking that
information into account with the conditional, we would be able to conclude that
Lee is not an DHS employee. (Symbolic version: If p then q; q is not true,
therefore p is not true.) The formula is:
If you are an employee of DHS, then you are a Federal employee.
Lee is not a Federal employee.
Therefore, Lee is not an employee of DHS.

There are two conclusions that cannot be drawn on the basis of the conditional. First, if the
antecedent is false, you cannot draw a conclusion about whether or not the consequent is
false. The reason is that the range covered by the consequent is potentially larger than the
range covered by the antecedent. The following example should show why this is the case.

Suppose we had the information that Adrian is not an employee of DHS. We

would not be able to conclude that Adrian is not a Federal employee, because
Adrian might work for another Federal agency. We really cannot tell whether or not
Adrian is a Federal employee based on the information we have. (Symbolic
version: If p then q; p is false, therefore, q may be either true or false.) The formula
is:
If you are an employee of DHS, then you are a Federal employee.
Adrian is not an employee of DHS.
Invalid conclusion: Adrian is not a Federal employee.

The second case in which you cannot draw a conclusion from the conditional occurs if you
know that the consequent is true. From that information, you cannot tell if the antecedent is
true or not.
Once again, the reason is that the range covered by the consequent is
potentially larger than the range covered by antecedent. The following example illustrates
this point.

Suppose we had information that Ren is a Federal employee. We would not be

able to conclude that Ren is an employee of DHS, because Ren might work for
another Federal agency. We really cannot tell whether or not Ren is a DHS
employee based on the information we have. (Symbolic version: If p then q; q is
true, therefore p may be either true or false.) The formula is:
If you are an employee of DHS, then you are a Federal employee.
Ren is a Federal employee.
Invalid Conclusion: Ren is an employee of DHS.

45
The following chart summarizes conclusions that can and cannot be made from the
conditional sentence If you are an employee of DHS, then you are a Federal employee (If
p then q).
Initial conditional sentence: If you are an employee of DHS, then you are a Federal
employee. (If p then q)
Conclusion

Validity

If you are not a Federal employee, then you are not an employee of
DHS. (If not q, then not p)11

valid

If you are a Federal employee, then you are an employee of DHS.

(If q, then p)

invalid

If you are not an employee of DHS, then you are not a Federal
employee. (If not p, then not q)

invalid

The second conclusion in the table is called the converse. The converse is created by
interchanging the two parts of the sentence, the antecedent and the consequent. As you may
recall from Section II.A.3, this form is not a valid conclusion from a statement such as All
[members of set A] are [members of set B]. It is also not a valid conclusion from a
conditional sentence.
The third conclusion in the table is called the inverse. The inverse is created by negating
both terms, but leaving them in their original positions. Like the converse, the inverse is not
valid because it is not equivalent in meaning to the original conditional sentence If you are
an employee of DHS, then you are a Federal employee.
The inverse and converse of the conditional are both powerful illogical biases. They
represent erroneous conclusions that people tend to make frequently. You should study
them carefully to be sure that you can avoid making these erroneous conclusions yourself.
In the examples that we provided on the previous page it was easy to see why the inverse
and the converse were wrong, because you are well aware of the relationship between being
a DHS employee and being a Federal employee. In order to study the biases further, let us
see how they work with the second conditional sentence: If there is a credible bomb threat
to a Federal office building, then the building is evacuated until a search is completed.
Assuming this sentence is true, let us look at what conclusions we can and cannot draw
when the four types of evidence are presented.

11

Logic Note. This logical form, in which the two parts of the sentence exchange position and are negated is
called the contrapositive. The contrapositive was first introduced in Section II.A.3.

46
Second conditional sentence: If there is a credible bomb threat to a Federal office building,
then the building is evacuated until a search is completed.
Conclusion

Validity

If we know that there is a credible bomb threat to a particular Federal office

building, we can conclude that the building is evacuated until a search is
completed. (If p, then q)

valid

If we know that a particular Federal office building has not been evacuated
pending a search, then we can conclude that there has not been a credible
bomb threat to that building. (If not-q, then not-p)

valid

If we know that there has not been a credible bomb threat to a particular
Federal office building, we cannot conclude that the building has not been
evacuated pending a search, because it may have been evacuated for some
other reason (a fire alarm, for example). (If not-p, then not-q cannot be
concluded)

invalid

If we know that a particular building has been evacuated pending a search, we

cannot conclude that there has been a credible bomb threat to that building,
because, once again, there may have been some other reason for the
evacuation. (If q, then p cannot be concluded)

invalid

It is recommended that you study the above examples and the exercises in the Self-Test so
that you can detect illogical biases in the conditional and avoid making these erroneous
conclusions.
Other Ways of Expressing the Conditional
The conditional is sometimes expressed without using the connective words if...then.
Some of these alternative expressions are when, whenever, any time that. For
example, the sentence about bomb threats could have be written in the following way:

Whenever there is a credible bomb threat to a Federal office building, that building
is always evacuated until a search is completed. This sentence contains the basic
idea of the conditional: If p occurs, then q occurs.

Another way of expressing the conditional is to use the connective words only if.
However, the expression only if represents conditionality opposite to that which might
initially appear to be the case. For example, in the sentence You are an employee of DHS
only if you are a Federal employee, the antecedent of the conditional is the first part of the
sentence (You are an employee of DHS), rather than the second part. The sentence means
exactly the same thing as the sentence we used as our original example of the conditional:
If you are an employee of DHS, then you are a Federal employee.

47
The conditional is also often expressed by using the word if, but leaving the word then
as understood but not expressed. For example, you could say If you are an employee of
DHS, you are a Federal employee.
Self-Test: Section II.C.2 (answers are given on page 120)
1. Which of the following sentences expresses a conditional? For each conditional
sentence, underline the antecedent and the consequent.
a. If you contribute to the Combined Federal Campaign through payroll deduction, then
there is a record of a deduction on your biweekly earnings statement.
(conditional / not a conditional)
b. If an international flight arrives, Inspectors process the arriving passengers.
(conditional / not a conditional)
c. You can take the advanced supervisory course only if you have taken the basic
supervisory course. (conditional / not a conditional)
d. I will either attend computer training or use the computerized tutorial or both.
(conditional / not a conditional)
e. Whenever an entire office undergoes a move, there is an inevitable period of
disruption. (conditional / not a conditional)
2. Each of the following items contains a pair of statements that might be joined in a
conditional sentence. Write a conditional sentence based on each pair in the space
provided below.
Example: a person is French, a person is European
If you are French, then you are European.
a. a person is a CBP Inspector, a person works for DHS
b. a person is hired into the Border Patrol, a person attends training in Glynco
c. a person receives e-mail messages, a persons computer is connected to a network
d. a person studies a foreign language enthusiastically, a person develops a good
vocabulary in a foreign language

48
In questions 3 and 4 below, you will be given a conditional statement that is followed by four
conclusions. Given that the statement is true, you should indicate whether each conclusion is
true (you can infer the conclusion from the statement) or false (you can infer that the conclusion
is contrary to the statement), or if there is insufficient information to decide whether the
conclusion is true or false.
3. Conditional sentence: If the computer was purchased for someone on our team, the Ateam, then it has modem.
a. The computer used by Mary has a modem; therefore, it was purchased for the Ateam. (T / F / I)
b. John uses one of the computers that was purchased for the A-team; therefore, it has a
modem. (T / F / I)
c. The computer used by Tom does not have a modem; therefore, it was not originally
purchased for the A-team. (T / F / I)
d. The computer used by Anne was purchased for the B-team originally; therefore, it
does not have a modem. (T / F / I)
4. Conditional sentence: If we request the same spending level as last year, our budget will
be approved.
a. We requested a higher spending level than last year; therefore, our budget will not be
approved. (T / F / I)
b. Our budget was not approved; therefore, we did not request the same spending level
as last year. (T / F / I)
c. We requested the same spending level as last year; therefore, our budget will not be
approved. (T / F / I)
d. We requested the same spending level as last year; therefore, our budget will be
approved. (T / F / I)

Section II.C.3. Other Basic Connectives

This section presents three other basic connectives -- biconditionality, the alternation, and
the disjunction. You will find that these forms represent situations that you find in your
everyday experience.

49
Biconditionality
A biconditional sentence expresses conditionality that goes, in a manner of speaking, in two
directions. The following sentence expresses biconditionality:
If the Secretary of DHS is the director of your agency, then you are an employee of
DHS; and if you are an employee of DHS, then the Secretary of DHS is the director
of your agency.
A simpler way of expressing the idea in this sentence is to say: The Secretary of DHS is the
director of your agency if and only if you are an employee of DHS.
A symbolic expression of the biconditional sentence is: If p then q, and if q then p. Another
way of expressing the biconditional is to say p if and only if q.
The following chart summarizes all of the conclusions that can be made from this
biconditional sentence.
Initial biconditional sentence: The Secretary of DHS is the director of your agency if and
only if you are an employee of DHS (p if and only if q)
Conclusion

Validity

If you are not an employee of DHS, the Secretary of DHS is not the
director of your agency.12 ( If not q, then not p)

valid

If you are an employee of DHS, the Secretary of DHS is the director of

your agency. (If q, then p)

valid

If the Secretary of DHS is not the director of your agency, then you are
not an employee of DHS. (If not p, then not q)

valid

The second conclusion in the table is called the converse, because the parts of the sentence
are interchanged. The third conclusion is called the inverse, because both parts of the
sentence are negated.
You can see that all of the conclusions are valid, in contrast to the case with the conditional.
As a consequence, there are no illogical biases for reasoning with the biconditional. The
reason why this occurs is not because the inverse and the converse have all of a sudden

12

Logic Note. This logical form, in which the two parts of the sentence exchange positions and are negated is
called the contrapositive. The contrapositive was first introduced in Section II.A.3 and was then reintroduced
in the discussion of the conditional in Section II.C.2.

50
become logically valid. The reason is that both statements refer to exactly the same
individuals or things (in other words, p is the same as q). Accordingly, if one statement is
true, the other is also true, and if one statement is false the other is also false.
Alternation
A sentence that is an alternation says: Either p or q, or both p and q. An example of an
alternation is:

This malfunctioning computer has either a damaged hard drive or too little memory,
or both of these problems.

The alternants are the simple statements that may be true or false. If the alternation is true
and if it is discovered that one of the alternants is false, then the other alternant must be true.
For example, if the sentence about the malfunctioning computer is true and it is discovered
that the computer does not have too little memory, then we would have to conclude that the
computers hard drive is damaged.
Another example of an alternation is the following:

This report must have been written either by Robert or by Clara or by the two of
them together.

If you find out that Clara did not write the report, then you can conclude that Robert did.
Disjunction
A disjunction is a sentence that has the following form: Not both p and q. You can see that
this is the negation of the conjunction of p and q. It states that It is not the case that both p
and q are true. A real-life example of a disjunction would be:

An employee cannot take both annual leave and sick leave at the same time.

The disjuncts are the basic statements that cannot both be true together. If one of the
disjuncts is known to be true, the other one must be false. For example, if we knew that a
certain employee had taken eight hours of annual leave yesterday, we would have to
conclude that the same employee did not take sick leave yesterday.
Another example of a disjunction is:

An alien cannot be both deported and returned voluntarily to the country of origin.

If you learn that a certain alien returned voluntarily to his or her country of origin, then you
know that the alien was not deported.

51

Self-Test: Section II.C.3 (answers are given on page 122)

1. Decide if each of the following sentences is a conditional, biconditional, alternation, or
disjunction.
a. If the suspects file was filed in the right place, it is filed under S.
(conditional / biconditional / alternation / disjunction)
b. The Chiefs office will hold a sofa and chair or a small conference table and chairs,
but not both. (conditional / biconditional / alternation / disjunction)
c. Your computer can be linked to the printer in Room 243, the printer in Room 246, or
both. (conditional / biconditional / alternation / disjunction)
d. If and only if you work more than eight hours in one day are you eligible for
overtime. (conditional / biconditional / alternation / disjunction)
e. If the evidence is strong enough for probable cause then it is strong enough for
reasonable suspicion. (conditional / biconditional / alternation / disjunction)
In questions 2 and 3 below, you will be given a statement that is followed by four
conclusions. Given that the statement is true, you should indicate whether each conclusion
is true (you can infer the conclusion from the statement) or false (you can infer that the
conclusion is contrary to the statement), or if there is insufficient information to decide
whether the conclusion is true or false.
2. Statement: The Jeep either has a faulty starter, bad spark plugs, or both.
a. The Jeep has a faulty starter; therefore, the Jeep does not have bad spark plugs.
(T / F / I)
b. The Jeep has bad spark plugs; therefore, the Jeep has a faulty starter.
(T / F / I)
c. The Jeep does not have bad spark plugs; therefore, the Jeep has a faulty starter.
(T / F / I)
d. The Jeep does not have a faulty starter; therefore, the Jeep has bad spark plugs.
(T / F / I)

52
3. Statement: No branch in this office has both a color printer and a black-and-white
printer.
a. A branch has a color printer; therefore, the branch does not have a black-and-white
printer. (T / F / I)
b. A branch does not have a black-and-white printer; therefore, the branch does not
have a color printer. (T / F / I)
c. A branch does not have a black-and-white printer; therefore, the branch has a color
printer. (T / F / I)
d. A branch has a black-and-white printer; therefore, the branch has a color printer.
(T / F / I)
Section II.C.4. Advanced Topic -- Complex Connectives
Although this is an advanced topic, it is extremely important in real-life reasoning
performance. You are strongly advised to study it when you have time to concentrate on it
and to give it the attention it deserves.
The complex connectives we will present fall into two categories: extended forms and
compound forms. We will present these separately.
Extended Forms
The extended forms we will present are based on the conditional. The forms each have two
conditional premises. The first form is based on two straightforward conditionals as in the
following example:
Premise 1: If additional staff are assigned, special funding will be needed.
Premise 2: If it is a holiday weekend, additional staff are assigned.
Conclusion: Therefore, if it is a holiday weekend, special funding will be needed.
You can see that the same statement (additional staff are assigned) is the consequent of the
second premise and the antecedent of the first premise. It is similar to the middle term in the
logic of three sets (see Section II.B.1). This same formula can be represented in symbols in
the following way:
If q then r
If p then q
Therefore, If p then r
The next extended form involves negation. An example of this form is as follows:

53
Premise 1: If Group B had the surveillance equipment, then Group C could not have
watched the compound last night.
Premise 2: If Group C had the Team Truck, then Group C watched the compound
last night.
Conclusion: Therefore, if Group C had the Team Truck, then Group B did not have
the surveillance equipment.
To understand why this conclusion is valid, start by looking at the second premise. The
consequent of that premise, Group C watched the compound last night, is equivalent to the
negation (or the contradiction) of the consequent of the first premise, Group C could not
have watched the compound last night. When we negate the consequent, we must also
negate the antecedent, Group B had the surveillance equipment. (Refer to the discussion
and table on valid conclusions from the conditional in Section II.C.2.) Therefore, if we
know that Group C had the Team Truck, then we must conclude that Group B did not have
the surveillance equipment.
The symbolic version of this formula is as follows:
If r then not q
If p then q
Therefore, If p then not r
Compound Forms
This is the last set of deductive forms to be presented. These last forms are just a sample of
the many ways in which the various connectives can be combined to produce more complex
forms. Therefore, as you study these forms, you should be learning the basic principles for
combining forms rather than trying to memorize any specific ones.
You have already seen in Section II.C.1 that a conjunction can be used as the antecedent in a
conditional (If p and q, then r). This form demonstrates the principle that a basic connective
form can be used as one of the terms in another basic form. Some other examples would be
the following:

alternation as antecedent of conditional: If either John or Mary conducts the

meeting, the meeting will be a success.
disjunction as consequent of conditional: If we do not receive additional funds, we
cannot both hire more staff and buy more equipment.
alternation as one of two conjuncts: Both the Assistant Chief and either Agent
Jones or Agent Preston (or both) must attend.

These forms are presented in symbols as follows:

alternation as antecedent of conditional:

disjunction as consequent of conditional:
alternation as one of two conjuncts:

If either p or q, then r
If p, then not both q and r
Both p and either q or r (or both q and r)

54

The second principle for using these compound forms is that the basic form that is
embedded must be treated as a single unit. If the unit is negated, the entire embedded basic
form must be negated. The following examples show how this principle is applied to the
compound forms presented. We will show the valid conclusion in which the antecedent and
consequent are exchanged and negated.13
Statement 1: If either John or Mary conducts the meeting, the meeting will be a success.
Valid conclusion:
If the meeting is not a success, then neither John nor Mary conducted it.
Statement 2: If the new information system is implemented, electronic records cannot have
both restricted and unrestricted access.
Valid conclusion:
If electronic records have both restricted and unrestricted access, the new
information system has not been implemented.
The symbolic version of these forms is as follows:
Compound Form

Valid Conclusion

If either p or q, then r

If not r, then neither p nor q

If p, then not both q and r

If both q and r, then not p

Illogical Biases with Extended and Compound Forms

Any extended or compound form that is based on the conditional will be subject to the same illogical
biases as the conditional. These two biases are, you will recall, the inverse and the converse.
The illogical biases for the extended conditional are shown using our previous example:
Premise 1: If additional staff are assigned, special funding will be needed.
Premise 2: If it is a holiday weekend, additional staff are assigned.
Valid conclusion: Therefore, if it is a holiday weekend, special funding will be needed.

13

Inverse (invalid): Therefore, if it is not a holiday weekend, special funding will not be
needed.

Converse (invalid): Therefore, if special funding is needed, it is a holiday weekend.

You may recall that this form is called the contrapositive.

55

The illogical biases for the compound conditional are shown using our previous example:
Premise: If either John or Mary conducts the meeting, the meeting will be a success.

Inverse (invalid): If neither John nor Mary conducts the meeting, the meeting will not be
a success.

Converse (invalid): If the meeting is a success, then either John or Mary conducted it.

Guarding against illogical biases is probably more difficult in these more complex forms than it is
for the simple conditional form. Consequently, one should use even greater vigilance to avoid these
biases.
Self-Test: Section II.C.4 (answers are given on page 123)
You will be given a statement that is followed by five conclusions. Given that the statement is true,
you should indicate whether each conclusion is true (you can infer the conclusion from the
statement) or false (you can infer that the conclusion is contrary to the statement), or if there is
insufficient information to decide whether the conclusion is true or false.
Statement: If the consultants recommendations are implemented, we will increase efficiency by
15% and reduce costs by 10%.
1. Efficiency increased by 15% and costs were reduced by 10%; therefore, the
consultants recommendations were implemented. (T / F / I)
2. Efficiency increased by 15% but costs were unchanged; therefore, the consultants
recommendations were not implemented. (T / F / I)
3. The consultants recommendations were implemented; therefore, efficiency
increased by 15% and costs were reduced by 10%. (T / F / I)
4. Costs were reduced by 10%; therefore, the consultants recommendations were
implemented. (T / F / I)
5. The plan to be implemented will either increase efficiency by 15% or reduce costs
by 10% but not both; therefore, the consultants recommendations are not
implemented by this plan. (T / F / I)

Part II.D. Using Language Correctly in Deduction

You have finished studying the basic forms for correct deductive reasoning. Perhaps you
have noticed in studying these forms that if your language is not compatible with these
forms, then conclusions that appear to be deductively correct will actually be incorrect.

56
It is important to avoid ambiguity in language at all times, but especially when using
deductive forms. When you use a word to represent a set of things, you should be sure that
you do not change definitions from one use of the word to another. The following example
of reasoning with three sets shows the problem created by using an ambiguous term.
All newly hired employees should be briefed on Federal employee benefits.
Some newly hired employees were RIFed from other Federal agencies.
Therefore, some employees who were RIFed from other Federal agencies should be
briefed on Federal employee benefits.
In this syllogism, the middle term -- newly hired employees -- refers to different sets in the
two premises. In the first premise, it clearly refers to employees who are new to Federal
Government employment. In the second premise, the term also includes employees who
have worked for another Federal agency before, but who have recently been hired by this
agency. Since the middle term has two different meanings, it obviously cannot serve as the
term of comparison for the other two terms. Thus, the syllogism is not valid.
The above example is instructive because it illustrates a common cause of ambiguity -mistaking part of a set for the whole. The term newly hired employees in the first premise
refers to a subset of the set referred to in the second premise. In the second premise, newly
hired employees refers to all employees who are new to the agency, while in the first
premise the term refers to only those new employees who are new to the Federal service.
Another common problem encountered in using language to represent logical forms is the
use of incorrect negations. That is, two words may be used as the negation of each other
when in fact they are not. For example, the words hot and cold represent opposite ends
of a continuum, but they do not represent the negation of each other. Something could be
non-hot and yet not be cold. Thus, the negation of hot must be not hot rather than
cold.

57

Unit III. Induction 1 -- Reasoning About Real-World Events

Part III.A. Introduction to Inductive Reasoning

Inductive reasoning, as you learned in Unit I, is reasoning that occurs in the absence of
complete information. Induction leads to conclusions that are not necessarily true but only
have some probability of being true. In inductive reasoning, the evidence does not
guarantee the truth of the conclusion, but it does give us a good reason to believe in the truth
of the conclusion. We can only say that the premises confirm or support the conclusion.
This is in contrast to deductive reasoning, in which the truth of the evidence makes the truth
of the conclusion certain.
Why Learn About Induction?
Induction is important in all human activity, from the ordinary activities of daily life through
the highest levels of scientific inquiry. Induction is the type of reasoning we use to draw
conclusions about the real world. We use induction to form generalizations, to make
predictions, and to develop explanations. It is a key component in problem solving and
decision making. Thus, induction is an essential job-related thinking skill.
Comparison of Deductive and Inductive Forms
Let us look at and compare some deductive conclusions with apparently similar inductive
conclusions. The table below shows the premises and conclusions for deductive and
inductive forms. The explanation after the table will guide you in understanding the
similarities and differences.

Premise(s)

Deductive
Of all first-line supervisors in the
agency, 95% have taken the basic
supervisory course.

Inductive
Of all first-line supervisors in the
agency, 95% have taken the basic
supervisory course.
Pat is one of 20 first-line supervisors
in Sector X.

Conclusions

Nearly all first-line supervisors in It is very likely that Pat has taken the
the agency have taken the basic basic supervisory course.
supervisory course.
Probably some of the people who
Some people who have taken the have taken the basic supervisory
basic supervisory course are first- course are first-line supervisors from
line supervisors in the agency.
Sector X.

58
The first premise is the same for both the deductive and inductive columns. It is a statement
that relates the two sets of first-line supervisors in the agency and people who have taken
the basic supervisory course. It states that 95% of first-line supervisors in the agency have
taken the course. You can see that this statement is not quite a universal statement, but it
has more information than a statement using the vague quantifier some.
We have drawn two deductive conclusions from this premise. In saying that Nearly all
first-line supervisors in the agency have taken the basic supervisory course, we are simply
replacing the quantity 95% with the expression nearly all. Assuming that 95% meets
our definition of nearly all, we are essentially restating the same information as the
original premise. The second conclusion, Some people who have taken the basic course in
supervisory skills are first-line supervisors in the agency, is a valid conclusion from the
statement Of all first-line supervisors in the agency, 95% have taken the basic supervisory
course. (See Section II.A.5.) Clearly, these conclusions are deductive; that is, they must be
true if the premise is true.
In the inductive column we have added a premise about an individual: Pat is one of 20 firstline supervisors in Sector X. Then, we have drawn two conclusions that are inductive. The
first says that it is very likely that Pat has taken the basic supervisory course. This
conclusion meets our criteria for inductive reasoning. First, it is based on incomplete
information. We do not have enough information to know for sure that Pat has taken the
course. Second, the conclusion is not true with certainty. Pat may be among the 5% of firstline supervisors in the agency who did not attend the training. However, given our evidence
that nearly all first-line supervisors did attend, it is very probable that Pat attended. Thus,
the conclusion has a high degree of probability (but not absolute certainty) associated with
it.
Criteria for Correct Induction
There are criteria for correct induction. First and foremost among these is the requirement
that the degree of probability claimed for the conclusion must be supported by the premises.
In our example involving the supervisor Pat, the conclusion that she is very likely to have
taken the basic supervisory course is clearly supported by the information that 95% of firstlevel supervisors in the agency have taken it. If we had concluded that she had certainly
taken the course or that she probably had not taken the course, our conclusion would not
have been supported by the evidence.
The second requirement is the requirement for use of the total available evidence. This
means that when forming an inductive conclusion, you should use all the evidence available
to you that is relevant to the conclusion. You may have a great deal of information that is
relevant or very little. Regardless of how much information you have, your inductive
reasoning is considered correct if you take all relevant information that you know into
account. You may need to revise your conclusion later if you obtain additional information
that casts doubt on the correctness of your original conclusion.

59
In the example given earlier about Pat, the first-level supervisor from Sector X, it is possible
that you might obtain the additional information that Supervisors in Sector X have attended
the basic supervisory course in lower-than-average proportions, because of other urgent
initiatives in the last year. Then, you would not be justified in concluding that there was a
95% chance that Pat had attended the training, because you would be ignoring a relevant
piece of information. You would be justified in concluding that there was less than a 95%
chance that Pat had attended the training.
The Meaning of Probability
We have been using the concept of probability to represent the likelihood that a conclusion
is true, given certain evidence. Another way of expressing this is to say that probability
represents the degree of confirmation for the conclusion provided by the premises. This is
the common interpretation of probability in inductive logic.
Probabilities vary from 0 to 1. The value of 1 represents absolute confirmation -- a
condition that applies only to deductive conclusions. A value of 0 represents no
confirmation. For the range between 0 and 1, numerical values may or may not be assigned.
Various words are used to refer to the degrees of certainty within this range. For example, a
typical meaning for the word probably is more likely than not or with a probability
greater than .5.
The word probability can also be used to represent relative frequency, in other words, how
often an event of interest occurs relative to other possible events. For example, in the case
of the first-level supervisors in the agency, our premise told us that 95 out of 100 of these
supervisors had taken the basic course in supervision. The relative frequency of 95/100
provided our best estimate of the probability that any one first-level supervisor had attended
the training.
We will use the concept of probability throughout this unit on induction and then later in the
unit on statistical reasoning.
In the remaining sections we will present some forms within which inductive reasoning can
occur. These forms are just a sample of the possible forms. They will be presented first for
reasoning about sets and then for reasoning with connectives. Although many of these
forms will be very similar to deductive forms, it should be remembered that they differ from
deductive forms in terms of the certainty of the conclusion.
Self-Test: Part III.A (answers are given on page 124)
In questions 1 through 7, decide if each of the sentences is true or false.
1. In induction, the premises provide a degree of confirmation for the conclusion.
(True / false)

60
2. An inductively correct conclusion is a necessary conclusion; that is, it cannot be false if
the premises are true. (True / false)
3. Induction must be based on all available evidence. (True / false)
4. Induction is based on complete evidence. (True / false)
5. A good inductive conclusion should not be altered if new evidence becomes available.
(True / false)
6. Probability can be viewed as the degree of confirmation that a set of premises provide
for a conclusion. (True / false)
7. Probability may have a numerical value ranging from 0 to 1. (True / false)
8. Given the following premises, decide if each of the conclusions represents an inductive
or a deductive conclusion. Then decide if the conclusion represents good reasoning or
poor reasoning.
Premises
Of all individuals who apply to take tests for Federal employment, only about 50%
actually report on the scheduled day to take the test.
Jane and John applied to take a test for a Federal law enforcement job, and they were
scheduled to be tested on October 1.
Conclusions
a. It is not true that everyone who applies to take a test for Federal employment reports
to take the test on the scheduled date.
(Inductive / deductive) (Good reasoning / poor reasoning)
b. Jane and John will both probably report to take the test on the scheduled date.
(Inductive / deductive) (Good reasoning / poor reasoning)
c. Some people who apply to take a test for Federal employment do not report to take
the test on the scheduled date.
(Inductive / deductive) (Good reasoning / poor reasoning)
d. There is a 50% chance that any one applicant will report to take the test on the
scheduled date.
(Inductive / deductive) (Good reasoning / poor reasoning)

61

Part III.B. Inductive Reasoning About Sets

Reasoning about two sets represents a good starting point for talking about induction. Part
II.A in the deductive reasoning unit gave you an extensive background that will give you a
head start in induction.
Section III.B.1: Inductive Reasoning About Two Sets
Inductive Reasoning About Two Sets When the Universe of Discourse Is Known
In reasoning deductively about the relationships between two sets, we start with one of the
basic premises, such as All [members of set A] are [members of set B]. In inductive
reasoning we can distinguish two different starting points that pertain to how much we know
about one of the sets we are interested in. In other words, we can have different outcomes
depending on how much we know about the universe of discourse for the set we wish to
relate to another set.
The universe of discourse in a deductive or inductive reasoning task is the range of
individuals, events, or things about which you wish to draw a conclusion. For an
illustration, let us refer to the following premise: Of all first-line supervisors, 95% have
taken the basic supervisory course. The universe of discourse for this statement needs to be
defined more explicitly. Clearly, the premise does not refer to all of the first-level
supervisors in the world, the nation, or even the Federal Government. In fact, it may apply
to first-level supervisors only in one occupation in DHS. Accordingly, let us say that we are
referring to the Border Patrol occupation.
If we know how many first-level supervisors exist in the Border Patrol and if we know how
many of them belong to the other set we are interested in (people who have taken the basic
supervisory course), then we can say that the universe of discourse is known. On the other
hand, if we do not know how many supervisors there are and if we have obtained
information on only some members of the set, then we say that the universe of discourse is
unknown.
We are in a much stronger position to draw conclusions if the universe of discourse is
known than if it is unknown. You will see that we can assign more exact probability values
to a prediction if we know the universe of discourse for one of the sets.
Now let us look at the types of conclusions that can be drawn when the universe of discourse
is known. First, let us return to the premise about the first-level supervisors. If we can
assume that the premise is based on reliable data about all of the first-level supervisors (in
other words, the universe of discourse is known), then we can confidently assign a .95
probability to the conclusion that any individual first-line supervisor had taken the basic
course. We could represent the reasoning form in the following manner:

62

Of all first-line supervisors in the Border Patrol, 95% have taken the basic
supervisory course.
Pat is a first-line supervisor in the Border Patrol.
Therefore, with a probability of .95, Pat has taken the basic supervisory course.
Now let us see what conclusion we can draw if we are trying to make an inference about an
individual who has taken the basic supervisory course. Our premise does not give us
complete information about the entire set of people who have taken the course. Therefore, if
the only information we have is that a particular person has taken the course, we can
conclude that there is some chance that the person is a first-level supervisor in the Border
Patrol, but we cannot assign any numerical value or any expression of strength to the
probability. We can express this in the following formula:
Of all first-line supervisors in the Border Patrol, 95% have taken the basic
supervisory course.
John has taken the basic supervisory course.
Therefore, John may be a first-line supervisor in the Border Patrol,
but the probability is unknown.
Now let us look at another type of scenario in which prediction can be made about an
individuals membership in one set based on information about membership in another set.
In this scenario, you will have one universal premise (a premise that tells about all the
members of a set) and an additional premise that gives a probability that an individual
belongs to one of the sets. The following example shows how this scenario would look:
All of the new trainees (in a specific occupation) are college graduates.
There is a 50% chance that your offices new employee this year will be one of the
new trainees.
Therefore, the probability that your office will be assigned a college graduate this
year is at least 50%.
The probability that your office will be assigned a college graduate might be greater than
50% because there is a chance that your office could be assigned an employee other than a
trainee who might also be a college graduate. This unknown probability must be added to
the 50% chance of obtaining a trainee (who would definitely be a college graduate).
Therefore, the chance of obtaining a college graduate is no lower than .50 and may be
higher.
This formula allows you to make an interesting conclusion if the second premise concerns
an individual who is in the second set in the original statement, the set of college graduates.
Let us assume that the second premise is There is a 50% chance that your offices new
employee will be a college graduate. This information combined with the universal
statement All of our new trainees are college graduates allows you to conclude that The
probability is no higher than .5 that our new employee will be one of the new trainees. This
formula is represented as follows:

63
All of the new trainees (in a specific occupation) are college graduates.
There is a 50% chance that your offices new employee this year will be a college
graduate.
Therefore, the probability is no higher than 50% that your offices new employee
will be one of the new trainees.
You are able to reach this conclusion because the universal statement puts an upper limit on
the probability that the new employee is a trainee. In other words, this person can only be a
new trainee if he or she is a college graduate.
This type of formula -- combining a universal statement about two sets with probabilistic
information about an individual -- is a very powerful formula that allows you to draw many
conclusions. By working through the exercises in the Self-Test, you will learn about more
of these conclusions.
A note about quantities. In discussing probabilities, we have been using various
expressions for the size of the probability. The following table shows you some of these
expressions and other expressions that are equivalent in meaning.
Degree of probability

Equivalent expressions

Probability greater than .7

greater than a 70% chance

Probability equal to or greater than .7

at least a 70% chance

no less than a 70% chance

Probability equal to .7

exactly a 70% chance

Probability equal to or less than .7

up to a 70% chance
no greater than a 70% chance
no more than a 70% chance

Probability less than .7

less than a 70% chance

Probably

greater than a 50% chance

more likely than not

Note: The probability of .7 was used as an example. Any other probability

value could be substituted in these expressions.

Inductive Reasoning About Two Sets When the Universe of Discourse Is Unknown
When the universe of discourse is unknown, we must rely on data from specific observations
to try to decide what the relationship between the two sets could be. This

64
use of specific observations as the basis for making a general rule is called inductive
generalization. A simple example about computer diskettes will illustrate the process:
I have tested every tenth diskette from this shipment and they are all defective.
Therefore, probably, all diskettes from this shipment are defective.
You can see that the conclusion is inductive. First, it is based on incomplete evidence
because I have not tested all of the diskettes from the shipment. Second, the conclusion is
only a probable conclusion; it could very well be wrong. If there are any nondefective
diskettes in the shipment, then the conclusion is wrong. Obviously this type of reasoning is
fraught with danger. No matter how regular events seem to be, there is still a real possibility
that such generalizations are wrong.
In spite of the fallibility of inductive generalization, it is a necessary tool for existence in the
real world. We must be able to generalize about phenomena rather than treat each event,
individual, or thing as if it were different from every other event, individual, or thing.
For example, suppose the law enforcement community refused to make the generalization
that every person has distinctive fingerprint patterns simply because the fingerprints of all
people (past, present, and future) had not been observed. Clearly, the community would
lose a valuable investigative tool. So, the generalization about fingerprints is a highly useful
one, even though there may be some (unobserved) exceptions to it.
Statistical reasoning is an aid to inductive generalization when the universe of discourse is
unknown. We can think of the universe of discourse as being the same as the statistical idea
of a population. Random sampling techniques are used to try to obtain a sample of
observations that represent the population. The statistics obtained from the sample are
evaluated by using statistical models, which estimate the range of likely population values.
There are also statistical techniques for making a prediction about an individual, given a
generalization about the relationship between two sets.
Statistical reasoning is explained in Unit IV.
Self-Test: Section III.B.1 (answers are on page 124)
1. For the following situations, decide if the universe of discourse is known or unknown.
a. A staff officer is writing a report on current and future trends in illegal immigration. In
the report the officer must draw some conclusions about the country of origin of all
illegal immigrants across the Southwest border. Is the universe of discourse known or
unknown for these illegal immigrants?

65
b. A personnel specialist in Headquarters, who has access to Servicewide personnel
records, needs to find out what percentage of the GS-9 employees at DHS are in each of
the Officer Corps occupations. Is the universe of discourse known or unknown for GS-9
employees with respect to what occupation they are in?
c. A budget analyst who has agency financial reports must report the proportion of the
agency budget that has been spent on travel in the past three fiscal years. Is the universe
of discourse known or unknown with respect to total expenditures and the percent spent
on travel?
d. A personnel recruitment specialist is interested in knowing how many college seniors
read advertisements about DHS jobs that have been placed in collegiate magazines. Her
only source of data is information from job applicants about where they learned about
DHS job opportunities. Is the universe of discourse known or unknown for the percent
of college seniors who read advertisements about DHS jobs?
2. Given the information in the following paragraph, decide which of conclusions a through d
can be correctly inferred.
Paragraph
In the last fiscal year, Sector X added 200 Border Patrol Agents, of which 180 were new
hires and 20 were acquired from other sectors. Sector Y acquired 400 new Border Patrol
Agents, of which 320 were new hires and 80 were acquired from other sectors.
Conclusions
a. Agent A. B. is one of the new agents in Sector X. Agent A. B. is very likely to be a new
hire. (Correct / incorrect)
b. Agent C. D. is one of the new agents in Sector Y. C. D. probably transferred from
another sector. (Correct / incorrect)
c. Agent E. F. is a new hire to the Border Patrol. E. F. probably works in Sector Y.
(Correct / incorrect)
d. Agent G. H. is a new agent in Sector X. There is only a 10% chance that G. H.
transferred from another sector. (Correct / incorrect)

66
3. Given the information in the following paragraph, decide which of conclusions a
through d can be correctly inferred.
Paragraph
According to records, on a certain vessel entering Long Beach seaport, all of the officers
in the crew are from Norway. However, three-fourths of the nonofficer members of the
crew are not from Norway. Inspector I. J. is one of the staff assigned to inspect this
vessel. Evaluate these conclusions for Inspector I. J.
Conclusions
a. A member of the crew who is a nonofficer has a one-fourth chance of being from
Norway. (Correct / incorrect)
b. A randomly selected member of the nonofficer members of the crew has a threefourths chance of not being from Norway. (Correct / incorrect)
c. A member of the vessels crew who is from Norway has a three-fourths chance of
being an officer. (Correct / incorrect)
d. A member of the vessels crew who is not from Norway has a three-fourths chance
of being an officer. (Correct / incorrect)
4. Given the information in the following paragraph, decide which of conclusions a
through d can be correctly inferred.
Paragraph
At a certain office of an agency, it was standard practice that on their first day in the
office all new employees would be given a tour of the office that was the same as the
tour given to official visitors. Due to organizational growth, there was a 50% chance on
any Monday of the year that a new employee would report for work in the office.
Conclusions
a. On any given Monday, there was at least a 50% chance that someone would be given
a tour of the office. (Correct / incorrect)
b. Anyone who was being given a tour of the office on a Monday was most likely a
new employee. (Correct / incorrect)
c. The probability that no one would have a tour on any given Monday was no greater
than .50. (Correct / incorrect)
d. On any given Monday, there was a 50% chance that there would not be a new
employee receiving a tour. (Correct / incorrect)

67
5. Given the information in the following paragraph, decide which of conclusions a
through d can be correctly inferred.
Paragraph
An administrative officer had to inventory 1,000 items of office equipment prior to a
move. He knew that there were exactly 300 pieces of computer equipment, none of
which had been declared surplus. The surplus status of the remaining equipment is
unknown prior to inventory. Evaluate these conclusions for the administrative officer.
Conclusions
a. If the administrative officer randomly chose where to begin the inventory, the first
piece of equipment to be inventoried would have a 30% or greater chance of not
being declared surplus. (Correct / incorrect)
b. The first piece of equipment inventoried had no more than a 70% chance of being
declared surplus. (Correct / incorrect)
c. If the first piece of equipment inventoried was one that had been declared surplus, it
had a 70% chance of not being computer equipment. (Correct / incorrect)
d. If the first piece of equipment inventoried had not been declared surplus, then the
probability that it was computer equipment is completely unknown.
(Correct / incorrect)
6. Which of the following statements represent inductive generalizations?
a. All squares have four sides.
b. Moss only grows on the north side of trees.
c. Radiation is harmful to the body.
d. In 1996, there were 50 states in the U. S.A.

Section III.B.2. Advanced Topic -- Inductive Reasoning With More Than Two Sets
Reasoning With More Than Two Sets When the Universe of Discourse Is Known
When the universe of discourse is known, valid forms for reasoning with three sets (these forms,
called syllogisms, were introduced in Section II.B.1) can be used as the basis for making predictions
about individuals. The following is an example of a form involving three sets:

68
Ninety percent of our recently hired employees are not certified in CPR.
All of our trainees are recently hired employees.
Sam is one of our trainees.
Therefore, the probability that Sam is not certified in CPR is 90%.
This syllogism has an expression of relative frequency (90%) in the first premise. The second
premise is a universal statement about our trainees. The third statement says that Sam is a trainee.
From that, we can infer that he is a recently hired employee, and since 90% of recently hired
employees are not certified in CPR, we can conclude that there is a 90% chance that Sam is not
certified.
Some syllogisms can be used in combination with a probabilistic statement to draw a probabilistic
conclusion about an individual. An example is as follows:
No recently hired employee is certified in CPR.
All customer service staff are certified in CPR.
There is a 50% chance that the employee detailed to the emergency site will be a member of
the customer service staff.
Therefore, the probability that the person detailed to the emergency site is not a recently
hired employee is at least 50%.
If the person detailed to the site has a 50% chance of being from the customer service staff, that
person is excluded from the set of recently hired employees with a probability of exactly .5; there is
also the possibility that that person is neither a recently hired employee nor a member of the
customer service staff. Thus, the probability that the person is not a recently hired employee is no
less than .5, and may be greater than .5.
Reasoning About More Than Two Sets When the Universe of Discourse Is Unknown
Logic books describe a type of informal reasoning called inductive analogy, which involves
reasoning with more than two sets when the universe of discourse is unknown. Stephen Barker, in
his text The Elements of Logic (1989) represents this type of reasoning as follows:
a, b, c ... each has been observed to be S and P.
k is an S.
Therefore, probably, k is P.
At first this seems to be simply inductive generalization with the added premise k is an S. But,
arguments by analogy depend more on the similarity between a, b, c, and k than they do on the
invariance of the relation between S and P. An example of this type of reasoning would be:
Six vans of a certain model, all of which had obstructed windows, were found to be
transporting illegal aliens. An agent observes another van of the same model which also has
obstructed windows. The agent concludes that the van is probably transporting illegal aliens.
Argument by analogy obviously can be a useful form of reasoning, but like inductive generalization,
it can lead to fallacious conclusions.

69
Self-Test: Section III.B.2 (answers are given on page 126)
1. Given the information in the following paragraph, decide which of conclusions a through d can
be correctly inferred.
Paragraph
In a work site that was about to be inspected by DHS agents, 30% of the employees were
undocumented aliens. None of the undocumented aliens was from a North, South, or Central
American country.
Conclusions
a. A randomly selected employee from that work site had at least a 30% chance of not being
from a North, South, or Central American country. (Correct / incorrect)
b. Any employee at that site who was from a North, South, or Central American country had a
70% chance of not being an undocumented alien. (Correct / incorrect)
c. A randomly selected employee from that work site had up to a 70% chance of being from a
North, South, or Central American country. (Correct / incorrect)
d. No one from a North, South, or Central American country worked at this site.
(Correct / incorrect)
2. Given the information in the following paragraph, decide which of conclusions a through c can
be correctly inferred.
Paragraph
In the newly decorated visitors area, all of the furniture has wood frames rather than plastic. Of
all the furniture that was in the visitors area prior to the redecoration, 80% did not have wood
frames.
Conclusions
a. An item of furniture in the redecorated visitors area has an 80% chance of not being a
holdover from before the redecorating. (Correct / incorrect)
b. An item of furniture that was used in the visitors area before redecoration had no more than a
20% chance of being used in the area after it was redecorated.
(Correct / incorrect)
c. An item of furniture that was used in the visitors area before redecoration has at least an 80%
chance of not being used in the area after it was redecorated. (Correct / incorrect)

70
3. Given the information in the following paragraph, decide which of conclusions a through d can
be correctly inferred.
Paragraph
In a one-year time period in a certain district, all cases of suspected document counterfeiting
were investigated vigorously. No cases that were investigated vigorously failed to yield a
successful prosecution. Agent Z worked in the district during the entire one-year period. That
year, a randomly selected case from Agent Zs caseload had a 50% chance of being a case of
suspected document counterfeiting.
Conclusions
a. This randomly selected case had at least a 50% chance of being investigated vigorously.
(Correct / incorrect)
b. This randomly selected case had at least a 50% chance of not failing to produce a successful
prosecution. (Correct / incorrect)
c. There was exactly a 50% chance that this randomly selected case was not investigated
vigorously. (Correct / incorrect)
d. There was a 50% chance that this case was prosecuted unsuccessfully.
(Correct / incorrect)

Part III.C. Inductive Reasoning With Connectives

In exploring inductive reasoning with connectives, we will take the approach, as we did with
sets, of exploring the types of inductive conclusions we can reach by adapting the deductive
forms.

Section III.C.1: Conditionals With Probabilities

The first form we will study is a conditional in which the consequent is true only with a
certain probability when the antecedent is true.14 Another way of saying this is that the
consequent is true for only a proportion of the time when the antecedent is true. This means
that part of the time the antecedent will be true and the consequent will be untrue. (This
concept will be studied again as the topic of conditional probability in the unit on statistical
reasoning.)

14

Logic Note. Remember that the antecedent of the conditional is the if part of the sentence and the
consequent is the then part of the sentence.

71
The following example illustrates this type of conditional:
If an agent applies for every vacancy in Sector X, he or she will have a 30% chance
of being promoted in Sector X this year.
Agent Jones will apply for every vacancy in Sector X this year.
Therefore, Agent Jones will have a 30% chance of being promoted in Sector X
this year.
Notice how different this formula is from the conditional used in deduction. The formula
yields only a probabilistic conclusion. Agent Jones may or may not be promoted in Sector
X if she applies for every vacancy.
Unlike the deductive conditional, this form does not permit a definite conclusion to be
drawn if the consequent is denied.15 In other words, from the information that an agent was
not promoted, we cannot draw a conclusion about the probability that the agent applied for
every vacancy.
Like the deductive conditional, this form does not permit a conclusion if the antecedent is
negated (the inverse) or the consequent is affirmed (the converse). In other words, from
information that an agent did not apply for all vacancies in Sector X, we would not be able
to estimate the likelihood that the agent was not promoted in Sector X (the inverse). From
information that the agent was promoted in Sector X, we would not be able to estimate the
probability that the agent applied for all vacancies in Sector X (the converse).
It is interesting to note here that the probability that an event will not occur is equal to 1
minus the probability that it will occur. Therefore, we can conclude that if a person does
apply for every vacancy, the person has a 70% chance of not being promoted. We can say
that these two probabilities (.3 and .7) are the complements of each other, because the two
probabilities add up to 1.
The Biconditional With Probabilities
We can define a biconditional premise that has two conditional probabilities associated with
it. It would retain the symmetry associated with the deductive biconditional. Such a
biconditional would look like this:

15

Logic Note. Another way of saying this is that the contrapositive is not a valid conclusion from this type of
conditional.

72
If a person has a certain type of infection, diagnostic test A has a 90% chance of
giving a positive diagnosis.
If a person (whose infection status is unknown) receives a positive diagnosis on test
A, there is a 90% chance that the person has this type of infection.16
Individual B is known to have this type of infection.
Therefore, individual B has a 90% chance of receiving a positive diagnosis on test
A.
The valid conclusion shown above is based on affirming the first part of this biconditional.
Another valid conclusion would be:
Individual C (whose infection status is unknown) has received a positive diagnosis
on test A.
Therefore, individual C has a 90% chance of having this type of infection.
Like the inductive conditional, this form still would not give a definite conclusion if either
the antecedent or the consequent were negated. In other words, from the information that a
person did not have this type of infection, we would not be able to draw any conclusion
about the probability that the person would have a positive diagnosis on test A. From
information that a person received a negative diagnosis on test A, we would not be able to
draw a conclusion about the probability that the person had the infection. Therefore, this
biconditional has some limitations on the types of conclusions that can be drawn, in contrast
to the deductive biconditional.
Self-Test: Section III.C.1 (answers are given on page 127)
1. The following paragraph contains a conditional statement with a probability. Given the
information in the paragraph, decide which of conclusions a through d are correct
inferences.
Paragraph
In a certain sector it was found that if a vehicle transporting illegal aliens was
apprehended within the sector by any law enforcement authority, there was a 20%
chance that it was stopped for speeding. Vehicle A was apprehended by a law
enforcement officer and was found to be transporting illegal aliens. Vehicle B was also
transporting illegal aliens, but it was not apprehended.

16

Logic Note. For purposes of simplicity, the same probability (.90) is used for the two parts of the
biconditional. However, different probabilities could be used for the two parts.

73
Conclusions
a. There is a 20% chance that vehicle A was stopped for speeding.
(Correct / incorrect)
b. There is an 80% chance that vehicle B was not stopped for speeding.
(Correct / incorrect)
c. There is an 80% chance that vehicle A was not stopped for speeding.
(Correct / incorrect)
d. Of all vehicles stopped for speeding in this sector, 20% were transporting illegal
aliens. (Correct / incorrect)
2. The following paragraph contains a biconditional with a probability. Given the
information in the paragraph, decide which of conclusions a through d are correct
inferences.
Paragraph
A librarian in a law enforcement agency, after studying circulation records and
surveying users, was able to draw two conclusions about library use. First, if a book was
checked out of the library for use on a specific work project, there was a 70% chance
that it would be returned on time. Also, of all the books that were returned on time, 70%
had been signed out for use on a specific work project.
Conclusions
a. If a book was signed out for some reason other than a specific work project, there
was a 70% chance that it would not be returned on time. (Correct / incorrect)
b. Of 1,000 books returned on time in a one-month period, probably about 700 of them
were signed out for use on a specific work project. (Correct / incorrect)
c. In one month, 500 books were signed out for specific work projects, and thus it can
be expected that approximately 350 of them will be returned on time.
(Correct / incorrect)
d. If a book was not returned on time, it was more likely than not signed out for a
reason other than a specific work project. (Correct / incorrect)

Section III.C.2: Connectives With an Added Premise Containing a Probability

Now let us look at another way of using the conditional in induction. As we did for sets, we
can combine information from a deductive conditional (if p then q) with probabilistic

74
information about the occurrence of the antecedent, p. Then we can draw an inductive
conclusion about the probability of the occurrence of the consequent, q.
In the example below, we use the conditional from the deductive reasoning section and add
a probabilistic premise:
If there is a credible bomb threat to a Federal office building, then the building is
always evacuated until a search is completed.
A Federal office building has a 1% chance of receiving a credible bomb threat
sometime within a year.
Therefore, a Federal office building has at least a 1% chance of being evacuated
during a years time.
The conclusion has the probability value of at least 1% because the building may have some
likelihood of being evacuated for reasons other than a bomb threat. That probability is not
contained in the information that our premises give us, but we must allow for it in our
conclusion.
Now let us vary this formula by combining the conditional with information about the
probability that the consequent is true. Suppose we have information that the chance that a
Federal office building will be evacuated for a search at some point during the year is 1%.
We can then conclude that the chance that a credible bomb threat is made to a Federal
building is no greater than 1%. This formula can be expressed as follows:
If there is a credible bomb threat to a Federal office building, then the building is
always evacuated until a search is completed.
The chance that a Federal office building will be evacuated for a search at some
point during the year is 1%.
Therefore, the chance that a credible bomb threat will be made to a Federal building
is no greater than 1% in a year.
This conclusion is true because the 1% chance of evacuation includes evacuations due to
credible bomb threats and any other causes. The conditional statement tells us that there
will not be a credible bomb threat without an evacuation of the building. So, the upper limit
on the chance of a bomb threat is 1%.
When a conditional is combined with an added probabilistic premise about either the
antecedent or the consequent, there is always a conclusion that has a range of probabilities.
We will illustrate the remaining conclusions through the use of formulas based on the
example above. (Further real-life examples of these formulas will be found in the Self-

75
Test.) The first is the formula that applies when you know the probability that the
antecedent is not true:
If there is a credible bomb threat to a Federal office building, then the building is
always evacuated until a search is completed.
A Federal office building has a 99% chance of not receiving a credible bomb threat
sometime within a year.
Therefore, a Federal office building has no more than a 99% chance of not being
evacuated during a years time.
Since the building has a 99% chance of not receiving a credible bomb threat, it has a 1%
chance of receiving such a threat and thus being evacuated. This relationship puts an upper
limit of 99% on the probability that the building will not be evacuated. In addition, there is
some (unknown) chance of its being evacuated for some other reason. Therefore, the
probability that the building will not be evacuated might be lower than 99%.
The next formula applies when you know the probability that the consequent is not true:
If there is a credible bomb threat to a Federal office building, then the building is
always evacuated until a search is completed.
The chance that a Federal office building will not be evacuated for a search at some
point during the year is 99%.
Therefore, the chance that a credible bomb threat will not be made to a Federal
building is equal to or greater than 99% in a year.
If the second premise is true, there is a 1% chance that the building will be evacuated for a
search at some point during the year. This means that there is no more than a 1% chance
that a credible bomb threat will be made and thus no less than a 99% chance that such a
bomb threat will not be made.
Biconditionals
When a biconditional is used with an added premise containing a probability, the resulting
conclusion has a specific probability rather than a range of probability. Let us illustrate this
with the following biconditional statement:
The Secretary of DHS is the director of your agency if and only if you are an
employee of DHS.
There is a 60% chance that a person attending this conference is an employee of
DHS.
Therefore, there is a 60% chance that the Secretary of DHS is the agency director of
a person attending this conference.

76
This formula is reversible; in other words, you can begin with information about the
probability that the Secretary of DHS is the agency director of a person attending this
conference and conclude that the person has the same probability of being an employee of
DHS.
You can also draw conclusions if you have information that the antecedent or the
consequent is not true. For example, suppose you have the information that There is a 30%
chance that a student in this computer training course is not an employee of DHS. From
that information, you could conclude that There is a 30% chance that the Secretary of DHS
is not the agency director of a student in this computer training course. This type of
formula will be used in exercise 3 in the Self-Test.
Advanced Topic: Other Connectives
Extended connectives based on the conditional can be used with added probabilistic premises. So
can compound connectives that are based on the conditional (see exercise 4 in the Self-Test).
However, the other connectives, such as alternations and conjunctions, cannot be adapted so easily
because they entail computations with the probabilities. This topic will be discussed in the unit on
statistical reasoning.

Self-Test: Section III.C.2 (answers are given on page 128)

1. The following paragraph contains a conditional premise and an added premise with a
probability. Given the information in the paragraph, decide which of conclusions a
through c are correct.
Paragraph
In one district, the Detention and Deportation staff had a busy caseload of criminal and
noncriminal alien cases in which deportation procedures were pending. If an alien in
one of these cases was a criminal, he or she was not detained at the DHS detention
facility. Instead, criminal aliens were placed in county jails. In this district, one-tenth of
all cases concerned criminal aliens.
Conclusions
a. A case in this district had at least a 10% chance of not involving someone who was
detained at the DHS detention facility. (Correct / incorrect)
b. A case in this district had no more than a 90% chance of involving someone who
was detained at the DHS detention facility. (Correct / incorrect)
c. A case in this district had less than a 90% chance of concerning a noncriminal alien.
(Correct / incorrect)

77
2. The following paragraph contains a conditional premise and an added premise with a
probability. Given the information in the paragraph, decide which of conclusions a
through c are correct.
Paragraph
If an employee contributes to the Combined Federal Campaign through payroll
deduction, there is a record of a special deduction on the employees biweekly earnings
statement. In a certain agency, there was a 90% chance that any randomly selected
employee had a record of some kind of special deduction on his or her earnings
statement.
Conclusions
a. There is at least a 10% chance that a randomly selected employee does not contribute
to the Combined Federal Campaign through payroll deduction.
(Correct / incorrect)
b. There is a 10% or greater chance that a randomly selected employee did contribute
to the Combined Federal Campaign through payroll deduction.
(Correct / incorrect)
c. There is a 10% chance that a randomly selected employee did not have a special
deduction on his or her biweekly earnings statement. (Correct / incorrect)
3. The following paragraph contains a biconditional premise and an added premise with a
probability. Given the information in the paragraph, decide which of conclusions a
through c are correct.
Paragraph
Employees in a certain category are eligible for overtime pay if and only if they work more
than eight hours in one day. Records showed that one employee in this category worked
more than eight hours on one-fourth of her regular work days in the last year.
Conclusions
a. There was more than a 25% chance that this employee was eligible for overtime pay
on any randomly selected day. (Correct / incorrect)
b. There was a 75% chance that this employee was not eligible for overtime pay on any
randomly selected day. (Correct / incorrect)
c. The chance that this employee was eligible for overtime pay on any randomly
selected day was 25%. (Correct / incorrect)

78
4. Advanced Exercise: The following paragraph contains a compound conditional
premise and an added premise with a probability. Given the information in the
paragraph, decide which of conclusions a through c are correct.
Paragraph
In a study of 500 work stations that had problems with LAN access, it was decided that
if the work stations got certain new hardware or upgraded software, the problems would
be corrected. After six months had elapsed, it was found that for 40% of the work
stations, the problems had not been corrected.
Conclusions
a. For any given work station, there was a 20% chance that it had not gotten new
hardware and a 20% chance that it had not gotten upgraded software.
(Correct / incorrect)
b. For any given work station, the probability was 60% that it had gotten both the new
hardware and the upgraded software. (Correct / incorrect)
c. For any given work station, the probability was at least .4 that it had gotten neither
the specified new hardware nor the upgraded software. (Correct / incorrect)

Part III.D. Fallacies in Induction

Just as there are prevalent fallacies in deduction, which we called illogical biases, so there
are fallacies in induction. First, there are what might be called schematic fallacies,
involving incorrect conclusions about converses, inverses, and other forms. The inductive
biases are related to the corresponding deductive biases, but they are not always the same.
Next, there are biases in interpreting probabilistic data from real-life observations. These
will be covered in the unit on statistical reasoning. These biases arise because, in trying to
interpret the mass of information that we take in everyday, we make various types of errors.
All of these errors violate one or more of the rules for statistical reasoning.
Finally, there are errors of inductive reasoning that the logician Stephen Barker (1989) calls
pure fallacies of induction. We will take a little time to review these because they provide
important cautions. There are three of these fallacies, as follows:

forgetful induction:
hasty induction:
slothful induction:

neglecting to use some of the information that you possess

basing a conclusion on very slight evidence
treating a conclusion as though it were less probable than
it really is

79
The following is an example of forgetful induction, because the reasoner neglects some
important information and draws a conclusion using only a small part of the available
information:
Ninety-five percent of the new vehicles we purchased last year performed very
well. Almost all of our officers expressed satisfaction with them. However, recently
I received a report that one officer drove three different new vehicles that had
mechanical problems. Therefore, I doubt that these vehicles are reliable enough for
our use.
The reasoner is neglecting the information about 95% of the new vehicles we purchased last
year and is only attending to the report about three vehicles.
The next paragraph shows an example of hasty induction, because the reasoner bases a
conclusion on slender evidence:
The only time that I attended training on computer applications the instructor was
poor and I didnt learn anything. Therefore, computer training seems valueless, and
I dont plan to attend it again.
Slothful induction is represented by the following reasoning:
The opinion polls by all the major polling organizations show that the incumbent
will win re-election to this office. However, I still dont think it is very likely that
the incumbent will win.
Slothful induction tends to occur when the reasoner is reluctant to believe the conclusion,
regardless of the evidence.
All of these fallacies violate the requirement to use all the available evidence in forming an
inductive conclusion. Both hasty induction and slothful induction also violate the cardinal
rule that the degree of the probability claimed for the conclusion must be supported by the
premises.
Reference for Unit III
Barker, Stephen F. (1989). The Elements of Logic, 5th edition. New York: McGraw-Hill.

80

Unit IV - Induction 2: Statistical Reasoning and Estimating

Probability
In your job, you need to follow the basic rules of statistical reasoning and probability every
day. In fact, it is the correct application of these rules that improves the accuracy with
which you can make generalizations based on information you receive while performing
your job. The purpose of this unit is to familiarize you with the basic principles of statistical
and probability theory. In the last part of the unit, heuristics or rules of thumb that often
lead to erroneous generalizations are discussed so that you can avoid these statistical
illogical biases while performing your job.

Part IV.A. Populations and Samples

Every day, you make inferences about the probability that an event will occur. The term
event covers just about anything you can think of. For example, an illegal entry into the
United States, a promotion, and the purchase of new computer equipment are all events.
The entire collection of the events of interest is referred to as the population17. In some
cases, the population is small and easily observed, such as the number of absences among
agents in a particular Border Patrol station. Other times, the population is large, making it
difficult to observe all events, such as the number of absences among all employees working
in the United States.
When the population is large, it is impractical to attempt to observe all possible events.
Instead, inferences about the probability that an event will occur are made on the basis of a
sample of observed events from the population. The term sample refers to a subset of all
events in the population. For example, news programs often conduct exit polls to project the
winner of an election before all the votes are counted. Since a large number of people vote
in an election, polls are conducted using a sample of all voters.
The extent to which a sample reflects the population has major implications for the accuracy
of the inferences you draw from the sample. For example, in 1948 the Chicago Tribune
conducted a poll to project the winner of the presidential election. Since the results of the
poll showed that Thomas Dewey had defeated incumbent Harry Truman, the Chicago
Tribune decided to use Dewey Defeats Truman as the headline for the morning edition of
the newspaper. Unfortunately for the Chicago Tribune, the sample of voters included in
their poll did not accurately reflect the population of all voters. The final results of the
election showed that Harry Truman had won 303 electoral votes to Thomas Deweys 189
(Strom Thurmond won 39 electoral votes running on the States Rights Democratic ticket).
As a result, the Chicago Tribune suffered an embarrassing blow when Harry Truman held a
copy of the newspaper over his head after the election. This part of the unit presents basic
17

can

Although we are used to thinking of the term population as referring to people, in statistics the term
refer to any events of interest.

81
rules of statistical reasoning which maximize the likelihood that a sample reflects the
population (and minimize the likelihood of making a mistake similar to that made by the
Chicago Tribune in 1948).

Section IV.A.1. Law of Large Numbers

The law of large numbers states that a sample is more likely to reflect the population when
the sample contains a large rather than small number of observations. For example, imagine
you are a supervisor and are asked to evaluate the quality of 50 graduates of a training class,
two of whom are assigned to your district. You are very impressed with the performance of
the two graduates, but you decide to obtain additional information before making an
evaluation. You call a supervisor at Alpha District who has been assigned ten graduates of
the training class. The supervisor tells you that he finds the graduates to be average
performers. Which sample is more likely to reflect the population of 50 graduates of the
training class?
While you are impressed with the quality of the two graduates assigned to your district, it
would be premature to make an evaluation of the entire training class on the basis of their
performance. According to the law of large numbers, the likelihood that a sample reflects
the population increases with the size of the sample. Therefore, the evaluation made by the
supervisor at Alpha District (based on the performance of ten graduates) is more likely to
reflect the quality of graduates of the training class than your evaluation (based on the
performance of two graduates).
The law of large numbers also states that estimates based on small samples are more likely
to be influenced by extreme observations (that is, observations that differ greatly from the
typical occurrence in the population) than are estimates based on large samples. For
example, imagine that your favorite baseball player had a batting average of .300 (an
average of 3 hits in every 10 official times at bat) during the past season. If you look at the
players batting average during individual games, you would find a number of games in
which his average was .000 (for example, no hits in four official times at bat), as well as a
number of games in which his average was 1.000 (for example, four hits in four official
times at bat). On the other hand, if you looked at the players batting average over the
course of ten games, it is unlikely that his average was .000 (for example, no hits in 40
official times at bat), or 1.000 (for example, 40 hits in 40 official times at bat). Therefore,
you are more likely to find extreme batting averages during individual games than you are
over the course of ten games.
While there are statistics that estimate the likelihood that a sample represents the population,
a presentation of these statistics is beyond the scope of this unit. However, it is important
that you understand the relationship between the size of the population and the sample size
necessary to maximize the likelihood that the sample reflects the population. For extremely
large populations, such as the population of all eligible voters in the United States, a sample
of less than 1% of the population may be sufficient. In fact, polls conducted to predict who

82
will win a presidential election are often based on a sample of 10,000 out of a population of
100,000,000 eligible voters. When the population is relatively small, such as the population
of 50 graduates of a training class described above, a sample of at least 30% of the
population may be necessary. Therefore, as the size of the population decreases, there is an
increase in the percentage of observed events in the population required to ensure that the
sample represents the population.
Self-Test: Section IV.A.1 (answers are given on page 130)
The problems below require you to apply the law of large numbers. Try to determine the
correct answer before reading the explanation given at the end of the manual.
1. In order to get information about a car he is interested in purchasing, John buys a copy
of Car and Driver magazine containing a survey on owner satisfaction. For each car,
1,000 owners participated in the survey. The results of the survey suggest that 80% of
the people who own the type of car John is interested in purchasing are satisfied.
While at work the next day, John is talking with three of his coworkers and tells them he
intends to purchase the car. It turns out that two of the coworkers have friends who own
the car and both coworkers claim that their friends are dissatisfied with its performance.
Should John consider buying the car? Justify your answer by applying the law of large
numbers.

2. The ACME Corporation has two claims processing centers. Center A processes 1,000
claims a day, while Center B processes 100 claims a day. At each center, it is estimated
that an average of 5% of the claims processed each day are incorrect. Beginning next
year, ACMEs Quality Control Division will be tracking the percentage of incorrect
claims processed daily at each center in order to get a more accurate estimate. Which
center is more likely to have days on which 7.5% or more of the claims processed are
incorrect? Justify your answer by applying the law of large numbers.

83
Section IV.A.2. Sample Representativeness
Another factor that impacts whether a sample reflects the population of events is sample
representativeness. According to the rule of sample representativeness, samples that are
similar to the population in terms of characteristics relevant to the event you are measuring
are more likely to be representative of the population. One way to ensure sample
representativeness is through random sampling. A sample is random if all possible events in
the population have an equal chance of being observed.
A type of random sampling, called stratified random sampling, involves sampling randomly
from sub-populations identified on the basis of key characteristics of the population. For
example, imagine you are interested in identifying the concerns of all employees in your
sector. Your sector has three stations, with 60% of the employees in Station 1, 20% of the
employees in Station 2, and 20% of the employees in Station 3. In the past, you have
noticed that employees in each station tend to have different concerns. One way to ensure
that each station is represented in the sample is to select employees randomly from each
station so that 60% of the sample is from Station 1, 20% of the sample is from Station 2, and
20% of the sample is from Station 3.
People who are not familiar with the importance of sample representativeness often base
their inferences on biased samples. A sample is biased when it does not reflect the
population on relevant characteristics. For example, because the media report routinely on
political scandals, some people assume that all politicians are crooked, even though very few
politicians are ever involved in a scandal.
The impact of sampling bias can also be demonstrated using the example of the 50 graduates
of a training class presented earlier. Imagine that the top 25 ranked graduates of the training
class are assigned to your district. After one month on the job, you find that most of the 25
graduates are superior performers. While it may be tempting to give the entire training class
an excellent evaluation, remember that the graduates assigned to your district represent the
top performers in the training class. Your evaluation might not be as favorable if the 25
graduates assigned to your district were ranked at the bottom of the training class.
In some cases, sampling bias can negate the benefits gained by satisfying the law of large
numbers and vice versa. In the training class example above, an evaluation made on the
basis of the 25 top ranked graduates of the training class satisfies the law of large numbers,
while violating the rule of sample representativeness. On the other hand, an evaluation
made on the basis of the performance of three graduates of the training class assigned
randomly to your district would satisfy the rule of sample representativeness, while violating
the law of large numbers. The best estimate of the quality of the graduates of the training
class would be obtained by observing a large number of high, medium, and low ranked
graduates selected randomly from the training class.

84
Self-Test: Section IV.A.2 (answers are given on page 130)
The problems below require you to apply the rule of sample representativeness. Try to
determine the correct answer before reading the explanation given at the end of the manual.
1. While reading the newspaper, Randy comes across an article that describes the results of
a poll asking people whom they intend to vote for in an upcoming local election.
According to the article, the sample was drawn by selecting telephone numbers
randomly from the telephone book. The results of the survey suggest that the incumbent
is in a close race with the challenger, with both candidates getting approximately 50% of
the votes.
The next day, Randy is watching a local television news program that conducted a
similar poll on the upcoming election. Viewers participated in the poll by calling a 1-900
number at a cost of $.50. The results of the poll suggest that the incumbent would get
60% of the votes, a comfortable margin of victory. Which poll should Randy consider to
be more accurate?
Justify your answer by applying the rule of sample
representativeness.

2. Lisa is a manager in a district with 500 employees. Due to an unusually high amount of
turnover, the district is almost always understaffed. While in the past managers could
fill open shifts by requiring employees to work overtime, the recent union contract
makes overtime voluntary.
In order to gauge available staff hours, the District Director asks Lisa to estimate the
percentage of employees in the district who would be willing to work overtime. Lisa
asks 10 employees working in her office if they would be willing to work overtime.
Much to her satisfaction, 8 of the employees say they would be willing to work
overtime.
Later that day, a coworker tells Lisa he feels it will be difficult to get enough volunteers
to fill the open shifts. He states that only 3 of the 10 employees he spoke with are
willing to work overtime. Lisas coworker surveyed employees working during
different shifts in key areas within the district. Which estimate should Lisa present to
the District Director?
Justify your answer by applying the rule of sample
representativeness.

85

Section IV.A.3. Regression to the Mean

The term regression to the mean can be defined two ways. First, regression to the mean can
refer to the fact that, on average, an extreme observation of an event will be followed by a
less extreme observation of the same event. That is, subsequent observations of the event
are likely to be more similar to the typical occurrence in the population. Regression to the
mean occurs when chance factors have an extremely negative (or extremely positive) impact
on the initial observation. For example, imagine you are the police commissioner in a major
U.S. city and would like to test the effectiveness of a new technique you have developed to
reduce crime. You decide to try the technique in the five precincts with the highest crime
rate during the past month. A month after introducing the new technique, you find that the
crime rate has gone down in four of the five precincts where the new technique has been
implemented.
Making an assessment at this point would be unwise because a number of chance factors
could have accounted for the reduced crime rate in the four precincts (for example, poor
weather keeping people off the streets for most of the month). The best way to evaluate the
effectiveness of the crime reduction technique would be to implement it in a large number of
randomly selected precincts. This satisfies both the law of large numbers and the rule of
sample representativeness and helps prevent you from making an erroneous evaluation
based on the regression to the mean phenomenon. If, over time, the crime rate goes down in
all precincts, you can conclude that the technique is effective. However, if the only
precincts to show improvement are those with a high crime rate during the month prior to
implementing the technique, the reduction in crime can be attributed to the phenomenon of
regression to the mean.
The second definition of regression to the mean refers to the fact that, on average, an
extreme observation of one event will be associated with a less extreme observation of a
different, but related event. For example, imagine you are assigned the 10 highest ranked
graduates of a training class. You are told that graduates with a high class rank during
training tend to be better performers on the job than graduates with a low class rank. On the
basis of this information, you estimate that the graduates will be among the best performers
in your district. While this estimate would be justified, it is important to keep in mind that
the relationship between training class rank and on-the-job performance is not perfect. As a
result, there will be instances where graduates with a low class rank turn out to be good
performers, as well as instances where graduates with a high class rank turn out to be poor
performers.

86
This phenomenon often occurs in colleges where class rank in high school is used to select
applicants for admission each year. Since high school rank and college performance are not
perfectly related, there will always be exceptions to the rule that applicants with a high class
rank will be good students, and applicants with a low class rank will be poor students.
Self-Test: Section IV.A.3 (answers are given on page 131)
The problems below require you to identify the regression to the mean phenomenon. Try to
determine the correct answer before reading the explanation given at the end of the manual.
1. Laura has developed a course outline that details an approach to teaching that she claims
will increase students scores on a standardized test. Eager to see if the teaching
approach is successful, Lauras supervisor assigns her to teach the course to 20 students
who scored in the bottom 10% of all test takers on a recent administration of the
standardized test. After completing the course, the students test scores increased an
average of 20 points, with most students now scoring in the bottom 30% of all test
takers. Despite the positive results, Laura warns her supervisor that it is premature to
call the course a success. Do you agree with Laura that it is premature to claim that the
course is effective in increasing scores on the standardized test? Justify your answer
considering the regression to the mean phenomenon.

2. Greg is interested in seeing a movie with his friend Jane. While looking through the
newspaper, Greg comes across a new movie he would really like to see. Unfortunately,
Jane is skeptical about going to see the movie because it stars a stand-up comedian.
Eager to see the movie, Greg points out that many stand-up comedians have gone on to
become Janes favorite actors. Convinced that a good stand-up comedian will be a good
actor, Jane agrees to go to the movie with Greg.
After the movie, both Greg and Jane comment on how disappointed they were with the
acting ability of the stand-up comedian. Should Greg and Jane be surprised that the
stand-up comedians acting ability was not as good as his ability to deliver a funny
comedy routine? Justify your answer considering the regression to the mean
phenomenon.

87

Section IV.A.4. Sample Validity

Sample validity refers to the extent to which a sample of observed events reflects what
interests you about the population. For example, imagine you are asked to determine the
best way to evaluate the performance of graduates of a training class. A large number of
graduates of the training class (law of large numbers) have been assigned randomly to work
in your district (sample representativeness). Aware of the rules of statistical reasoning, you
feel comfortable that the sample reflects the population of graduates of the training class.
You then set out to determine the best way to evaluate performance. You decide to use
specific aspects of the job performed by experienced employees as your measure of
performance.
While it is likely that the sample reflects the population of graduates of the training class,
your definition of performance is questionable. By focusing on activities performed by
experienced employees, you may make an invalid evaluation of the performance of the
graduates. A more valid way to measure the performance of graduates of the training class
would be to use job activities that are not impacted by experience.
Self-Test: Section IV.A.4 (answers are given on page 131)
Below are two problems that require you to apply the rule of sample validity. Try to
determine the correct answer before reading the explanation given at the end of the manual.
1. Julia is interested in buying a new stereo system. She listens to a number of stereo
systems and has a difficult time determining which is best. Julia decides that the most
accurate way to assess the quality of a stereo system is cost. She reasons that the most
expensive stereo system must be the best, or it would not cost more money than the other
stereo systems. Do you agree with Julias belief that the best stereo system is the one
that costs the most money? Justify your answer by applying the rule of sample validity.

88
2. David is asked to assess whether employees in his district are satisfied with their pay. In
order to make this assessment, David uses the results of a recently conducted survey on
overall job satisfaction of employees in his district. The survey sample was large and
representative of all employees in his district. The results of the survey showed that
85% of the employees in Davids district are satisfied overall with their job. David
concludes that if employees are satisfied with their job overall, they must be satisfied
with their pay. He tells his supervisor that most employees are satisfied with their pay.
Do you agree with Davids conclusion that employees who are satisfied overall with
their job are satisfied with their pay? Justify your answer by applying the rule of sample
validity.

Section IV.A.5. Summary

In this part of the unit, a number of rules of statistical reasoning were presented. Since you
are often faced with situations that require you to consider a sample of all possible events,
understanding these rules can increase the likelihood that your sample reflects the
population of interest. Below is a list of tips that can help you when applying these rules
while performing your job.

Clearly define the population of interest. It is in your best interest to define the
population as narrowly as possible. This helps to ensure that you are obtaining relevant
information, and makes drawing a sample more manageable.

Determine the number of observations you need to make in order to get an accurate
reflection of the population. Remember, when a population is small, you may have to
observe a larger portion of the population in order to draw an accurate conclusion.

Define the key characteristics of the population you are interested in. Determine if your
sample of observations reflects the population in terms of these characteristics.

Ensure that what you are observing reflects what you are interested in.

89

Part IV.B. Estimating Probability

We live in a probabilistic world in which very few things happen with complete certainty.
For example, most people expect to be able to take a particular route to work each day.
However, sometimes people are forced to take an alternate route due to accidents or road
construction. Thus, we expect to able to take our usual route to work, but we know that
there may be occasions when we will have to take an alternate route. Understanding basic
rules of probability will help you make better estimates about the likelihood that events will
occur. This part of the unit outlines some of the basic concepts and rules of probability and
shows how they are related to making inferences in your job.
Section IV.B.1. Probability Defined
As mentioned in Unit III, a probability ranges from 0 to 1, with 0 representing an event
that will definitely not occur, .5 representing an event that is as likely to occur as it is not
to occur, and 1 representing an event that will definitely occur. In everyday life, the term
chance is often used as an expression of probability. For example, when a weather
reporter talks about the likelihood of rain, he or she typically states there is a 50% chance
of rain rather than the probability that it will rain is .5. Therefore, probability will be
expressed as chance in this unit to keep consistent with the expression of probability in
everyday life.
Section IV.B.2. Base Rate Information and Probability Estimates
An estimate about the probability that an event will occur is often based on information from
a variety of sources. An important piece of information, which is often overlooked, is the
base rate. The base rate is also called the prior probability because it refers to the number
of times an event has occurred in the past. For example, imagine you play tennis with a
friend once a week. On average, you win two of the four games played. Therefore, the base
rate of games you win while playing tennis with your friend is 50%.
The base rate is especially useful in two situations. First, the base rate is used when
additional information is not available. For example, imagine you are in charge of security
for a facility near a seaside resort. The manager of the facility asks you to estimate how
many crimes will be committed during the summer season. If no other information is
available, your estimate should be based on the crime rate during previous summer seasons
(note that the term crime rate is used instead of base rate).
The base rate is also used when additional information provided is unreliable. Using the
example above, imagine you are provided with a schedule for the summer season. You
notice that an unusually large number of conventions are scheduled throughout the summer.
In the past, crime rates have tended to increase during conventions. However, the increase
in crime during conventions has varied from year to year. Information about previous crime

90
rates when conventions are held can be used to adjust your base rate estimate of the crime
rate during the summer season.

Section IV.B.3. Occurrence of Two or More Events

When considering the probability that two events will occur, it is important to identify their
relationship to each other. Two events are considered independent when the occurrence of
one event is unrelated to the occurrence of another event. For example, the probability of
having red hair is independent of gender, with men being just as likely to have red hair as
women. Two events are considered dependent when the occurrence of one event is related
to the occurrence of another event. For example, the probability of being over six feet tall is
dependent on gender, with men being more likely than women to be over six feet tall.
The occurrence of two events can be expressed as either a joint probability or a conditional
probability. Joint probability refers to the probability that two events will occur at the same
time. For example, the probability of being male and being over six feet tall is a joint
probability. Typically, joint probability is denoted as p(A,B). Thus, p(over six feet tall,
male) is used to represent the joint probability of being a male and being over six feet tall.
Conditional probability refers to the probability that one event will occur given that some
other event has occurred. For example, the probability of being over six feet tall given being
a male is a conditional probability. Typically, conditional probability is denoted as p(A/B),
where / represents the word given. Thus, p(over six feet tall/male) is used to represent
the conditional probability of being over six feet tall given being a male. Conditional
probability is especially useful when two events are not independent, such as the
relationship between height and gender.18

Section IV.B.4. Multiplicative Rules of Probability

The multiplicative rule for independent events states that the probability that two
independent events will occur is equal to the product of their individual probabilities. For
example, imagine you are the chief of police in a major U.S. city. Determined to live up to
her campaign promise, the mayor of the city asks you to develop a comprehensive plan that
will reduce crime. After drafting a plan, you realize that the success of the plan depends on
both an increase in the number of police officers in the city and an improvement in the
economy. While you have little control over these factors, you do know that there is a 70%
chance that new officers will be hired, and a 30% chance that the economy will improve.
18

Logic Note. Notice that conditional probability expresses the relationship of a probabilistic conditional
statement discussed in Unit III. Also, while the symbol p is typically used to represent probability, you can
also use the abbreviation prob if you are using symbols to represent connectives such as the conditional. For
example, you could represent a conditional with probability as: If p, then q (prob=.50).

91
You also know that the state of the economy will have no impact on whether or not new
officers are hired. In probability, these events are represented as:
p(new officers) = 70%
p(improvement in the economy) = 30%

The joint probability that both new police officers are hired and the economy is improved is
determined by applying the multiplicative rule for independent events as follows:
p(new officers, improvement in the economy) = p(new police officers) X p(improvement in the economy)

Thus, according to the multiplicative rule for independent events, there is a 21% chance
(70% X 30%) that both new police officers are hired and the economy is improved.19
When two events are not independent, the multiplicative rule for dependent events is used
to account for the fact that the two events are related. According to the multiplicative rule
for dependent events, the probability of two dependent events occurring is equal to the
product of the individual probability that one event will occur and the conditional
probability that the second event will occur given that the first event has occurred. For
example, imagine you and your neighbor work in different departments for the same agency.
In order to save money, you and your neighbor carpool to work. Recently, your agency
announced that it is opening a new office complex closer to your home and intends to
transfer two of its six departments to the new office complex. You are interested in
transferring to the new complex, but would like to continue carpooling with your neighbor.
The probability that both you and your neighbor are transferred to the new office complex is
dependent because once one department is transferred, the probability that the second
department is transferred is reduced. The reduction in the probability that the second
department is transferred occurs because there is one less opening at the new office complex
and one less unit to fill it (1 opening / 5 departments). As a result, there is a 20% chance
that the second department is transferred to the new office complex given that one
department has already been transferred. In probability, these events are represented as:
p(a department is transferred) = 33%
p(a department is transferred / a department has already been transferred) = 20%

In this case, the probability that both you and your neighbor are transferred to the new office
complex is determined by applying the multiplicative rule of probability for dependent
events as follows:
p(your department is transferred, your neighbors department is transferred) =
p(a department is transferred) X p(a department is transferred / a department has already been transferred)

19

Logic Note. Notice that the multiplicative rule applies to a situation that has the logical form of the
conjunction, discussed in Unit II under deductive reasoning with connectives.

92
Thus, there is a 7% (33% X 20%) chance that both you and your neighbor are transferred to
the new office complex.
Self-Test: Section IV.B.4 (answers are given on page 132)
The problems below require you to apply the multiplicative rules of probability. Try to
determine the correct answer before reading the explanation given at the end of the manual.
1. While driving to work, Loukas car breaks down and is towed to a nearby garage. After
examining the car, a mechanic tells Loukas that three parts need to be replaced for the
car to run properly. After receiving the estimate, Loukas tells the mechanic that he
cannot afford to have the car repaired at the price quoted. Eager to get business, the
mechanic tells Loukas that he can save money by using rebuilt parts, but warns that each
rebuilt part has a 90% chance of working. After thinking about it for a while, Loukas
estimates that there is about a 70% chance that his car will run properly with the rebuilt
parts. As a result, he decides to borrow money and get new parts put in. Do you agree
with Loukas assessment that there is about a 70% chance that his car will run properly
with the rebuilt parts? Justify your answer by applying the multiplicative rule for
independent events.

2. Jennifer has recently graduated from college with a degree in marketing. She receives a
call from her friend Bob who tells her about a job he has applied for at a prestigious
marketing firm. The firm is hiring two associates at the entry level. Excited about the
prospect of working with Bob, Jennifer applies for the job. After one week goes by,
Jennifer calls the personnel department at the marketing firm to inquire about her
application. She is told that, including her and Bobs applications, four people have
applied for the two jobs and that no other applications will be accepted.
Applying the multiplicative rule of probability, Jennifer determines that there is a 25%
chance (50% X 50%) that she and Bob will be offered positions at the marketing firm.
Do you agree with Jennifers assessment that there is a 25% chance that she and Bob
will both get a job offer? Justify your answer applying the multiplicative rule for
dependent events.

93

Section IV.B.5. Additive Rules of Probability

The additive rule for independent events states that the probability that at least one of a
number of independent events will occur is equal to the sum of their individual probabilities.
For example, imagine you join a scuba diving group. As a practice, two members are
randomly paired each time the group meets to go diving. Understanding the dangers of
scuba diving, you are concerned about being paired with a novice diver. You find out that
35% of the group members are novice divers, 50% of the members are intermediate divers,
and 15% of the members are expert divers. In probability, these events are represented as:
p(novice) = 35%
p(intermediate) = 50%
p(expert) = 15%

The probability of being paired with either an intermediate diver or an expert diver is
determined by applying the additive rule for independent events as follows:
p(intermediate or expert) = p(intermediate) + p(expert)

Thus, there is a 65% chance (50% + 15%) that you will be paired with either an intermediate
or an expert diver.
When two events are not independent, the additive rule for dependent events is used to
determine the probability that at least one of a number of dependent events will occur. For
example, imagine you are in charge of recruiting for a local police force. An analysis shows
that the most successful officers either have a college degree or have military experience.
Concerned that there will not be enough qualified people to fill the openings, you decide to
survey people interested in becoming an officer to assess how many people would meet the
criterion of having a college degree or having military experience. You find that 60% of
those interested in becoming an officer have a college degree and 40% have military
experience.
Applying the additive rule for independent events would suggest that all people who are
interested in becoming officers would be qualified. However, because you are dealing with
two related events (that is, some people will have both a college degree and military
experience), it is necessary to subtract the joint probability of having a college degree and
having military experience. Imagine that the results of your survey suggest the following

94
joint probabilities for each possible combination of having or not having a college degree
and military experience.

College Degree
No College Degree
Total

Military Experience No Military Experience

30%
30%
10%
30%
40%
60%

Total
60%
40%
100%

Where:
p(college degree, military experience) = 30%
p(college degree, no military experience) = 30%
p(no college degree, military experience) = 10%
p(no college degree, no military experience) = 30%
Applying the additive rule for dependent events you find the following:
p(college degree or military experience) =
p(college degree) + p(military experience) - p(college degree, military experience)
Thus, there is a 70% chance (60% + 40% - 30%) that a person interested in becoming an
officer has either a college degree or military experience.20
Self-Test: Section IV.B.5 (answers are given on page 132)
The problems below require you to apply the additive rules of probability. Try to determine
the correct answer before reading the explanation given at the end of the manual.
1. Donna is not feeling well and decides to schedule a medical examination at one of two
doctors offices. Both offices have 20 doctors working at all times. In the past, Donna
has had a number of problems with general practitioners and residents. As a result, she
prefers to see internists or specialists whenever possible. Unfortunately, the policy at
both offices is that a patient is treated by the first available doctor. Donna decides to call
both offices to find out how many general practitioners, residents, internists, and
specialists work at each office. She gets the following information.

Office A
Office B

General Practitioner
45%
35%

Resident
5%
20%

Internist
35%
40%

Specialist
15%
5%

Given this information, at which office should Donna schedule an appointment given her
preference for internists and specialists? Justify your answer by applying the additive
rule for independent events.
20

Logic Note. Notice that situations in which the additive rule is used have the logical form of the alternation,
discussed in Unit II under deductive reasoning with connectives.

95

2. During a morning meeting, Sharons supervisor tells her that she must rent temporary
office space by the end of the business day. Sharon contacts a real estate agent and tells
her that she is interested in renting an office that either has ample parking space or is
near public transportation. In order to narrow the search, the real estate agent tells
Sharon the percentage of offices with ample parking space and the percentage of offices
near public transportation in each town in the area. After hearing her options, Sharon
decides that the best place to rent an office is in Milltown because 50% of the offices
have ample parking space and 50% of the offices are near public transportation. On the
basis of this information, Sharon determines that there is a 100% chance that the office
will either have ample parking space or be near public transportation. Sharon tells the
real estate agent she will rent an office in Milltown. Do you agree with Sharons
assessment that the office is guaranteed either to have ample parking space or to be near
public transportation? Justify your answer by applying the additive rule of probability
for dependent events.

Section IV.B.6. Summary

This part of the unit has concentrated on ways to increase the likelihood that you will make
accurate inferences in everyday life. You should go back and re-read relevant sections if
you feel unsure about any of the principles outlined above. The next part of the unit will
outline some of the common mistakes people make when estimating the likelihood of
events.

96

Part IV.C. Biases in Statistical Reasoning and Estimations of

Probability
While the rules of statistical reasoning and probability described above can increase the
likelihood that an estimate will be accurate, people often fail to apply them. Instead, people
apply heuristics or rules-of-thumb when making estimates that an event will occur. The
term heuristic refers to the speculative formulation of a solution to a problem. While the use
of heuristics can result in effective actions, heuristics can also lead people to ignore or
misinterpret useful information. This part of the unit describes the most common heuristics
and shows how the use of heuristics can result in inaccurate estimates when you perform
your job.
Section IV.C.1. Availability
Which is more likely, being killed in a car crash or being killed in a plane crash? If you use
the availability heuristic, your response to this question would be based on the event that is
most easily retrieved from memory. Since the media are more likely to report incidents of
plane crashes, people using the availability heuristic are more likely to retrieve memories of
a plane crash in which people were killed than they are of a car crash in which people were
killed. However, a person is more likely to be killed in a car crash than in a plane crash.
The use of the availability heuristic can result in violations of a number of the rules of
statistical reasoning. For example, imagine you are asked to estimate the probability that a
person will be married by the time he or she is 25 years old. If you use the availability
heuristic, your estimate would be based on the percentage of marriages among people you
know (that is, friends and relatives) in their early to mid-twenties because these examples
easily come to mind. A violation of the law of large numbers can occur if you only know a
few people who are in their early to mid-twenties. A violation of the rule of sample
representativeness can occur because your sample is likely to consist of people who are
similar on a number of characteristics, such as ethnicity, religion, educational level, and
geographic location. Each of these factors is related to the age at which a person gets
married.
Section IV.C.2. Representativeness
Representativeness refers to the tendency to assign an event to a category based on the
degree to which the event resembles other members of the category. For example, you may
assume that a person driving a Porsche is wealthy. It also can refer to the tendency to
believe that event A resulted from process B. For example, you may assume that a
person does not have a job because he or she is lazy.
Perhaps the most prevalent examples of the use of the representativeness heuristic are
stereotypes. Stereotypes are general descriptions of a category that allow us to make quick
decisions based on a limited amount of available information. When stereotypes lead to
inaccurate estimates, it is usually because two variables are perceived to go together even

97
though a relationship between them does not exist. This tendency has been termed illusory
correlation. For example, a common stereotype is that accountants are boring. Despite this
stereotype, there is no evidence that there is a correlation between being an accountant and
being boring.
The use of the representativeness heuristic can result in the violation of a number of the
rules of statistical reasoning. For example, imagine you are asked if John will be a good
student. You are told that John plays three sports during the school year. If you hold the
stereotype that athletes are poor students, you will probably predict that John will not be a
good student. In this example, a violation of the law of large numbers can occur if your
stereotype of athletes scholastic performance is based on a few observations of athletes in a
classroom. A violation of the rule of sample representativeness can occur because
stereotypes often result in selective perception when observing events of interest. That is,
people have a tendency to notice or remember events that are consistent with their
stereotype, and ignore events that are inconsistent with their stereotype. As a result, your
estimate may be based on a restricted sample consisting of observations confirming your
stereotype.
Section IV.C.3. Dilution
Dilution refers to the tendency to make a conservative estimate of the probability that an
event will occur when irrelevant information is provided along with relevant information.
For example, imagine you are searching for a primary care physician in your Health
Maintenance Organization (HMO). An HMO representative recommends a young doctor
whose office is located close to your home. The representative tells you that the doctor
graduated at the top of her class from a prestigious medical school. Given this information
you would probably predict that the doctor is very competent and make an appointment to
see her. When you arrive at her office, you find that the doctor is wearing clothes that do
not match and has not combed her hair. As a result of the doctors appearance, your
confidence in her ability to practice medicine is decreased. In this case, the doctors
appearance has diluted your estimate of her qualifications to practice medicine.
Dilution can result in a violation of the rule of sample validity. This occurs because
irrelevant information is used to make an estimate. In the example presented above,
considering the doctors physical appearance when making an estimate of her ability to
practice medicine may result in an erroneous conclusion. An evaluation of the doctors
ability to practice medicine should be based on factors related to the medical profession
rather than her physical appearance.

Section IV.C.4. Anchoring and Adjustment

Anchoring and adjustment refers to the tendency to make conservative adjustments from a
reference point as new information is provided. An example of anchoring and adjustment
can be found in the television game show The Price is Right. In this game, four
contestants are asked to estimate the price of a prize. The contestant with the closest

98
estimate without going over the actual price proceeds to a game in which he or she can win
other prizes. Typically, the remaining three contestants use the first contestants estimate of
the price as a reference point, adjusting their estimates only slightly from this point.
Anchoring and adjustment can result in a failure to recognize the regression to the mean
phenomenon when the initial reference point is extreme. For example, imagine you ask 50
people if they agree that 90% of teenagers in the United States play a musical instrument.
Most people would say they feel fewer than 90% of the teenagers in the United States play a
musical instrument. Now imagine that you ask another 50 people if they feel 1% of the
teenagers in the United States play a musical instrument. In this instance, most people
would say they feel that more than 1% of the teenagers in the United States play a musical
instrument. However, if people given a reference point of 90% are asked to give a specific
estimate of how many teenagers play a musical instrument, most will give an estimate less
than, but close to 90%. Similarly, the people given the reference point of 1% will most
likely give an estimate of more than, but close to 1%.

Section IV.C.5. Order Effect

Order effect refers to the tendency to make estimates based solely on initial or recent
observations. For example, imagine that Agent Smith has recently been assigned to work in
your sector, and makes a costly mistake on his first day on the job. When conducting a
performance appraisal, you may determine that Agent Smith performed less than adequately
even if his performance was acceptable for most of the review period. Now imagine you are
asked to conduct a performance appraisal for Agent Jones. During the review period, Agent
Jones has performed at an adequate level. However, in the past week, she has apprehended
an unusually high number of illegal aliens. You may be inclined to give Agent Jones a high
rating for the entire review period based on her most recent performance, even though her
performance for the entire period was adequate.
The order effect can lead people to violate the law of large numbers and the rule of sample
representativeness. It can also result in a failure to recognize the regression to the mean
phenomenon. Estimates based on early or recent observations violate the law of large
numbers because only a few observations are considered. For example, only considering
Agent Smiths initial level of performance when making your evaluation restricts your
sample of observations to one day. A violation of the rule of sample representativeness can
occur because your sample would not include performance in all areas, at different times,
and under different conditions. Finally, the order effect can result in a failure to recognize
the regression to the mean phenomenon when initial (or recent) observations are extreme.
For example, suggesting Agent Jones get a promotion on the basis of her performance over
the last week of the review period would be inappropriate because her performance will
most likely return to the level of the preceding weeks.

99
Section IV.C.6. Ignoring Relative Frequency
Ignoring relative frequency refers to the tendency to make estimates based on observed
frequencies rather than on relative frequencies. This occurs when information on
nonoccurrences is ignored or not available. For example, imagine you are interested in
buying a four-wheel drive vehicle. Due to an extremely harsh winter the previous year,
orders for the vehicle you are interested in purchasing are greater than the number of
vehicles produced. You go to two dealerships to see if they have the vehicle available.
Dealership A claims to be getting 10 vehicles in each month, while dealership B claims to be
getting 20 vehicles in each month.
At first, it may appear that you would have a better chance of getting the vehicle from
dealership B. However, an examination of the number of people purchasing the vehicle at
each dealership suggests that, on average, 20 people purchase the vehicle at dealership A,
while an average of 60 people purchase the vehicle at dealership B. Given this information,
you would determine that you are more likely to get the vehicle from dealership A (10
vehicles for 20 orders) than from dealership B (20 vehicles for 60 orders). Therefore, it is
important to consider the relative frequency rather than absolute frequency when
determining probabilities.
Ignoring relative frequency can result in an erroneous estimate of the base rate. In the
example above, the base rate of getting a vehicle at dealership A is 50%, while the base rate
of getting the vehicle at dealership B is 33%. If you ignore the relative frequency, you
would most likely estimate that the base rate of getting the vehicle is greater at dealership B
than at dealership A simply because dealership B will be receiving more vehicles during a
given month.

Section IV.C.7. Concrete Information

Concrete information refers to the tendency to give observed information more weight than
abstract information, even if the abstract information is more reliable. For example, imagine
you are the chief of police in a major U.S. city. You are interested in determining whether
the crime rate has changed over the past year. Anecdotal evidence from two of your top
officers suggests that crime has gone down. However, a recently released FBI report based
on all crime reported in the city suggests that the crime rate in your city has increased 5%.
You may be tempted to give the opinions of the officers more weight than the FBI report
simply because their reports are based on first-hand experience.
Favoring concrete information can result in violations of the law of large numbers and the
rule of sample representativeness. In the example above, a violation of the law of large
numbers can occur because the estimate is based on the experiences of a very small
percentage of all possible police officers in the city. A violation of the rule of sample
representativeness can occur because it is unlikely that the officers are exposed to all

100
neighborhoods in the city. On the other hand, the FBI report is based on all crime reported
in the city, regardless of neighborhood.

Section IV.C.8. Gamblers Fallacy

Gamblers fallacy refers to the tendency to overestimate the probability of the occurrence of
an event not previously observed. For example, imagine you and a friend have planned a
fishing trip during the upcoming weekend. You are concerned that it may rain, noting to
your friend that it rains 60% of the weekends during this time of year. In fact, it has rained
during each of the past five weekends. Always the optimist, your friend states it probably
will not rain because we are due for a sunny weekend.
In this example, the gamblers fallacy has resulted in your friends misuse of the
multiplicative rule for independent events. The gamblers fallacy occurs because people
focus on the probability of a sequence of events rather than the probability of the occurrence
of a single, independent event. Applying the multiplicative rule for independent events, you
would find that there is a 5% chance that it will rain during 6 consecutive weekends (.6 X .6
X .6 X .6 X .6 X .6). However, since the probability that it will rain during any individual
weekend is independent of whether it has rained during previous weekends, there is still a
60% chance that it will rain during the weekend of your fishing trip.
The gamblers fallacy is also referred to as the Monte Carlo fallacy. It got this name in 1913
when a roulette wheel at a Monte Carlo casino came up red 26 times in a row. After the
13th spin, people began betting large sums of money on black, working under the
assumption that the probability of the ball landing on a red number 14 times in a row is
infinitesimal. Unfortunately, they failed to consider that the probability of an outcome
during each spin of the wheel is independent of previous outcomes. As a result, many
people lost large sums of money by betting on black during the next 13 spins of the wheel.

Section IV.C.9. Misperceptions about Equiprobability

Misperceptions about equiprobability refers to the tendency to define incorrectly the
probability that two or more events will occur as equal when in fact they are not. For
example, imagine you have a friend coming in from out of town for the weekend. You plan
a number of outdoor activities and want to determine the probability that it will be sunny
during both days of the weekend. On any given day, there is a 50% chance that it will be
sunny and a 50% chance that it will rain.
When determining the probability that it will be sunny on both days, many people consider
the possible combinations of weather during the two days without respect to the order in
which the sunny and rainy days occur. Using this approach, they assume there is an equal
chance that it will be sunny both days, raining both days, and sunny one day and raining one
day. This results in an estimate that there is a 33% chance that it will be sunny on both days
during the weekend. However, the belief that each of the combinations has the same

101
probability of occurring is a faulty assumption. In all, there are four possible combinations
of sunny and rainy days over a period of two days:
1.
2.
3.
4.

First Day
Sun
Sun
Rain
Rain

Second Day
Sun
Rain
Sun
Rain

Since there are four possible combinations, there is a 25% chance (1/4) that any individual
combination will occur. Examining the table above, you will notice that there is a 25%
chance that it is sunny during both days (#1), and a 25% chance that it is raining during both
days (#4). In order to determine the probability of having one day of sun and one day of
rain, you must apply the additive rule for independent events. There is a 25% chance that it
is sunny during the first day and raining during the second day (#2), and a 25% chance that
it is raining during the first day and sunny during the second day (#3). Applying the
additive rule for independent events, you find that probability of having one sunny day and
one rainy day during the weekend is equal to 50% (25% + 25%). In this case, the
misperception of equiprobability would result in an underestimate of the probability that it is
sunny one day and raining on one day. Notice that the probability that it is sunny one day
and raining one day (50%) is twice as large as the probability that it is sunny both days
(25%) or raining both days (25%).

Section IV.C.10. Overestimating Conjunctive Events

Overestimating conjunctive events refers to the tendency to overestimate the probability that
two events will occur. For example, Tversky and Kahneman (1982) presented people with
the following description:
Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a
student, she was deeply concerned with issues of discrimination and social justice, and also
participated in antinuclear demonstrations. Please check off the most likely alternative:
Linda is a bank teller.
Linda is a bank teller and is active in the feminist movement.
Nearly 90% of the respondents said that Linda is more likely to be a feminist bank teller
than a bank teller. However, according to the multiplication rule for independent events, the
probability that two events will occur (that is, being a bank teller and a feminist) cannot be
greater than the probability of either event alone. Thus, the probability that Linda is a bank
teller is necessarily greater than the probability that she is a feminist bank teller.
The conjunction fallacy results in a violation of the multiplicative rule, and tends to become
more profound as the number of events involved increases. This can be shown using the
example of the rebuilt car parts presented in problem 1 of the self-test for section IV.B.4

102
covering the multiplicative rules of probability. Remember that in the problem, the three
rebuilt car parts each had a 90% chance of working correctly, resulting in a 73% chance that
the car will work correctly with all three rebuilt parts. However, if five rebuilt parts were
used (each with a 90% chance of working correctly) there is a 59% chance that the car will
work correctly.
Section IV.C.11. Summary
In this part of the unit, heuristics that often lead to erroneous estimates of probability were
presented. When making an evaluation, consider whether you have used one of the
heuristics described above. If you have used one of the heuristics, think about the rules of
statistical reasoning and probability that it violated, and determine its impact on the accuracy
of your assessment. Below are tables which list heuristics described above and the rules of
statistical reasoning and probability they violate.
Table 1. Biases in Statistical Reasoning
Law of
Large Numbers

Bias

Sample
Representativeness

Regression
to the Mean

Anchoring and Adjustment

Sample
Validity

Availability

Concrete Information

Dilution

Order Effect

Representativeness

Table 2. Biases in Estimating Probabilities

Bias

Base Rate

Multiplicative Rules

Gamblers Fallacy
Ignoring Relative Frequency

Additive Rules

X
X

Misperceptions about Equiprobability

Overestimating Conjunctive Events

References for Unit IV

Kahneman, D., & Tversky, A. (1982). On the study of statistical intuitions. Cognition, 11,
237-251.

103

Unit V - Posttest for Manual of Job-Related Thinking Skills

The following passage describes a set of facts. The passage is followed by eight
conclusions. Read the passage and then decide whether each conclusion is:
T) true, which means that you can infer the conclusion from the facts given; or
F) false, which means that the conclusion is contrary to the facts given; or whether there is
I) insufficient information to decide, which means that there is insufficient information for
you to determine whether the facts imply the conclusion or are contrary to the
conclusion.
A staff member who is responsible for scheduling meetings at a district office needs to be
aware of certain practices and requirements within the district. For example, all
naturalization ceremonies in the district are scheduled for Wednesdays and Thursdays. Of
course, no naturalization ceremonies are closed to the public. However, some meetings that
are closed to the public are held in the district directors conference room.
1.

The only events held on Wednesdays and Thursdays are naturalization ceremonies.
(T / F / I)

2.

Any event that is scheduled for Tuesday is not a naturalization ceremony. (T / F / I)

3.

Some naturalization ceremonies are not held on Wednesday or Thursday. (T / F / I)

4.

All naturalization ceremonies are open to the public. (T / F / I)

5.

Some meetings held in the district directors conference room are closed to the public.
(T / F / I)

6.

Some events held in the district directors conference room are not naturalization
ceremonies. (T / F / I)

7.

No events that are held on Wednesdays or Thursdays are closed to the public. (T / F / I)

8.

Some naturalization ceremonies are not held in the district directors conference room.
(T / F / I)

104
The following passage describes a set of facts. The passage is followed by eight conclusions.
Read the passage and then decide whether each conclusion is:
T) true, which means that you can infer the conclusion from the facts given; or
F) false, which means that the conclusion is contrary to the facts given; or whether there is
I) insufficient information to decide, which means that there is insufficient information for
you to determine whether the facts imply the conclusion or are contrary to the
conclusion.
When an Adjudications Officer interviews an applicant for a benefit, such as a green card,
certain prescribed procedures are followed. If the information provided by the applicant
meets the legal criteria for approval, and only if it meets these criteria, the application is
approved on the spot. If the information does not meet the criteria, either the applicant is
asked to supply additional information or the application is denied. If the Adjudications
Officer suspects fraud, a Special Agent is asked to join the interview.
9.

If a Special Agent is not asked to join the interview, then the applicant is not suspected
of fraud. (T / F / I)

10. If the information provided by the applicant does not meet the legal criteria, the
application is not approved on the spot. (T / F / I)
11. If the applicant is not asked to supply additional information, then the information
already provided meets the legal criteria. (T / F / I)
12. If the application is approved on the spot, then the information provided meets the legal
criteria. (T / F / I)
13. Even if the information provided by the applicant meets the legal criteria, the
application often is not approved on the spot. (T / F / I)
14. A certain individuals application was not approved on the spot; as a result, a Special
Agent was asked to sit in on the interview with this individual. (T / F / I)
15. If the Adjudications Officer does not suspect fraud, a Special Agent is not asked to join
the interview. (T / F / I)
16. If the information provided by an applicant does not meet the legal criteria for approval
but the application is not denied, then the Adjudications Officer has asked for more
information. (T / F / I)

105
The following passage describes a set of facts. The passage is followed by eight conclusions.
Read the passage and then decide whether each conclusion is:
T) true, which means that you can infer the conclusion from the facts given; or
F) false, which means that the conclusion is contrary to the facts given; or whether there is
I) insufficient information to decide, which means that there is insufficient information for
you to determine whether the facts imply the conclusion or are contrary to the
conclusion.
Security measures have been developed to protect laptop computers used at DHS.
Unauthorized access is one security concern. If individuals without proper DHS clearance
obtain access to DHS computers or data, it is considered to be unauthorized access. As one
form of protection, all DHS laptops should have access control software. The physical
security of laptops, particularly during travel, is another concern. If a laptop computer is
checked as baggage, it must be secured in a specialized storage container.
17. Some of the computers that should have access control software are DHS laptops.
(T / F / I)
18. If a computer is an DHS laptop, then it should have access control software. (T / F / I)
19. The only computers at DHS that need to have access control software are laptops.
(T / F / I)
20. If a person who did not have a proper DHS clearance obtained access to data from a
laptop, but not to the laptop itself, this would not be considered unauthorized access.
(T / F / I)
21. If there was no unauthorized access to an DHS laptop, that means that no individuals
without an DHS clearance obtained access to the laptop or its data. (T / F / I)
22. If an DHS laptop is not being checked as baggage, it does not need to be secured in a
specialized storage container. (T / F / I)
23. All computers that must be secured in specialized storage containers also have access
control software. (T / F / I)
24. No DHS laptop should be without access control software. (T / F / I)

106
The following passage describes a set of facts. The passage is followed by eight conclusions.
Read the passage and then decide whether each conclusion is:
T) true, which means that you can infer the conclusion from the facts given; or
F) false, which means that the conclusion is contrary to the facts given; or whether there is
I) insufficient information to decide, which means that there is insufficient information for
you to determine whether the facts imply the conclusion or are contrary to the
conclusion.
In Sector X, illegal immigrants enter the country through one of two areas--either through a
border crossing on one of several highways or by crossing areas of rough terrain away from the
highways. A study of activity in this sector revealed that nearly all of the illegal immigrants
under the age of 16 that were apprehended were intercepted at the highway border crossings. It
was also found that nearly all illegal immigrants apprehended while crossing the areas of rough
terrain were 16 years of age or older. During the last month of the study, the number of illegal
aliens apprehended in rough terrain was twice as great as the number intercepted at highway
border crossings.
25. An illegal alien intercepted in rough terrain was most likely to be 16 years of age or older.
(T / F / I)
26. If an illegal alien was not apprehended in an area of rough terrain, then the alien was not
under the age of 16. (T / F / I)
27. An illegal alien apprehended at a highway border crossing was more likely to be under the
age of 16 than to be 16 years of age or older. (T / F / I)
28. Of the illegal aliens who were 16 years of age or older and who were apprehended while
trying to enter this sector, nearly all attempted to cross the area of rough terrain.(T / F / I)
29. If an illegal immigrant under the age of 16 is apprehended, it is most likely to be at a
highway border crossing. (T / F / I)
30. During the last month of the study, an illegal alien apprehended while entering this sector
was more likely to be 16 years of age or older than to be under the age of 16. (T / F / I)
31. During the last month of the study, an illegal immigrant 16 years of age or older who was
apprehended was more likely to have been apprehended at a highway border crossing than in
rough terrain. (T / F / I)
32. During the last month of the study, an illegal alien who was not apprehended was most
likely to have crossed the border at a highway border crossing. (T / F / I)

End of Posttest. Check your answers against those on pages 134-138.

107

Unit I - Answers to Self-Test and Manual of Job-Related Thinking Skills Pretest

Unit I
1.
2.
3.
4.
5.
6.

basic
certain
complete
probabilistic
incomplete
induction

Manual of Job-Related Thinking Skills Pretest

1. True. This conclusion can be inferred from the second sentence of the paragraph, which
says that none of the criminal cases resulting from apprehensions by the Border Patrol at
Station X last month were narcotics cases. Since none of the criminal cases was a
narcotics case, it follows that no narcotics cases were among the criminal cases resulting
from apprehensions.
2. Insufficient information. The third sentence states that there were a few criminal cases
involving alien smuggling that were not referred to another agency. The paragraph does
not tell us anything about criminal cases that were referred to another agency. The
paragraph does not give us enough information to conclude that there were such cases.
3. True. The first sentence says that only narcotics cases are referred to DEA. The second
sentence says that none of the criminal cases resulting from apprehensions last month
were narcotics cases. From that information, we can conclude that none of the criminal
cases resulting from apprehensions at Station X last month were referred to the DEA.
4. True. Since all narcotics cases are referred to DEA, it must be that any case that is not
referred to DEA is not a narcotics case.
5. False. This statement contradicts the second sentence of the paragraph, which says that
none of the criminal cases resulting from apprehensions by the Border Patrol in Station
X last month were narcotics cases. If the second sentence is true, then it cannot be true
that some criminal cases apprehended last month by the Border Patrol in Station X were
narcotics cases.
6. True. This answer can be inferred from the first sentence of the paragraph. Since all
narcotics cases resulting from apprehensions by the Border Patrol are referred to the
DEA, it must be that none of them remain unreferred to DEA.

108
7. Insufficient information. The paragraph does not provide us any information about cases
that were referred to other agencies. It only tells us about cases that were not referred to
other agencies. We cannot tell from the paragraph whether or not any cases were
referred to another agency.
8. True. This conclusion can be inferred from information in the second and third
sentences. The third sentence says that a few criminal cases resulting from
apprehensions last month involved alien smuggling. We know that these cases did not
involve narcotics because the second sentence says that none of the criminal cases
resulting from apprehensions last month were narcotics cases.
9. True. This conclusion can be inferred from the second sentence, which says that an alien
must have either a border crossing card or a resident alien card or a passport. If an alien
does not have one of these documents, he or she must present one of the other two types
of document.
10. True. This conclusion can be inferred from the first sentence of the paragraph, which
states that if aliens do not present themselves for inspection at a designated port of entry,
they cannot enter the U.S.A. legally. From this statement, we can conclude that
presenting oneself for inspection is an absolute requirement for entry. Therefore, if an
alien was allowed to enter the U.S.A., then the alien must have met that requirement.
11. Insufficient information. Although the first sentence states that presenting oneself for
inspection is a requirement for entry, we cannot conclude that M.N. was allowed to enter
simply because she presented herself at the designated port of entry. She might not have
met other requirements for entry, as stated in the second sentence. Therefore, the
paragraph does not give us sufficient information to conclude that she would be allowed
to enter.
12. Insufficient information. The paragraph does not give us information about all aliens
who entered at this port of entry for the purpose of shopping. We know that all people
who entered with a border crossing card were either visiting family or shopping.
However, some people who came for shopping may have presented another type of
document (for example, a passport).
13. True. The paragraph tells us that in the week J.T. entered, every alien who had a border
crossing card entered either to visit family or to shop. J.T. had a border crossing card
and he was not coming to shop. Therefore, he must have entered to visit his family.
14. False. This conclusion is false because it contradicts the information in the first
sentence, which says that if aliens do not present themselves at a designated port of
entry, they cannot enter the U.S.A. legally.

109
15. Insufficient information. The paragraph does not give us any information about the
characteristics of aliens who entered without a border crossing card.
16. Insufficient information. The paragraph tells us that aliens must present themselves at a
designated port of entry and have one of three documents in order to enter the country
legally. This is not sufficient information for us to conclude that if an alien does not
present himself or herself at a port of entry, then he or she does not have one of the three
documents.
17. True. The third sentence says that there were several apprehensions that involved
extensive pursuit by local authorities. From this sentence we can conclude that some
cases that involved pursuit by local authorities resulted in apprehensions. We know
from the second sentence that all apprehended persons took voluntary departure.
Therefore, we can conclude that some cases that involved extensive pursuit by local
authorities resulted in the voluntary departure of aliens.
18. False. This information contradicts the information in the first and second sentences.
The first sentence says that an illegal alien cannot both take voluntary departure and be
deported. According to the second sentence, all apprehended aliens took voluntary
departure, which leads to the conclusion that none was deported.
19. Insufficient information. The third sentence says that several apprehensions involved
extensive pursuit by local authorities. This does not give us sufficient information to
conclude that all apprehensions involved such pursuit.
20. True. Three cases involved pursuit to a crossroads where the individual either turned left
or was picked up by a vehicle, or both. Two of the individuals were not picked up by a
vehicle. Therefore, they must have taken the left turn.
21. True. This conclusion is equivalent in meaning to the third sentence of the paragraph,
which says that there were several apprehensions that involved extensive pursuit by
local authorities.
22. Insufficient information. The paragraph does not tell us if all apprehensions were the
result of extensive pursuit by local authorities (see explanation for question 19).
Therefore, we cannot conclude that everyone who took voluntary departure was
apprehended as the result of such pursuit.
23. True. This conclusion is equivalent in meaning to the third sentence of the paragraph.
Since some apprehensions did involve extensive pursuit by local authorities we can say
that these same apprehensions did not occur without such pursuit.
24. False. This conclusion contradicts the eighth sentence, which says that in these three
cases, the individuals had either turned left or been picked up by a vehicle, or both.

110
25. Insufficient information. There is insufficient information to estimate the probability
that an illegal alien crossing the bridge when the train is present will be apprehended.
According to the last sentence of the first paragraph, although aliens are apprehended
when the train is present, the sensor is of no use in signaling the presence of people
under such circumstances. Furthermore, the passage does not tell the proportion of aliens
that are apprehended and thus we cannot estimate the probability that an individual alien
will be apprehended while crossing the bridge if a train is present.
26. True. The third sentence states that if there is no train, 70% of sensor signals lead to
apprehensions of illegal aliens. Therefore, if an Agent responds to such a signal, there is
a 70% chance that an alien will be apprehended.
27. Insufficient information. The passage does not give us information about the number of
aliens who cross the bridge without being apprehended. Therefore, we do not know the
likelihood that people who crossed without being apprehended did so when the train was
present.
28. True. This conclusion is correct because the information in the passage allows us to
conclude that in an average week at least 70 people are apprehended crossing when the
train is not present and 10 people are apprehended crossing when the train is present.
Therefore, at least seven-eighths of the people are apprehended while crossing in the
absence of the train. Since the overwhelming majority are apprehended crossing while
the train is absent, it is very likely that this was the case for group A.
29. Insufficient information. There is no information that allows us to conclude how often
trains are present. Although many more aliens are apprehended when the train is absent,
it could be simply that many more try to cross the bridge when the train is absent.
30. True. The last sentence of the first paragraph states that apprehensions are made when
the train is present. Therefore, it cannot be true that no apprehensions occur when a train
is present.
31. False. This is an incorrect statement of the likelihood that a sensor signal will lead to an
apprehension of an illegal alien when a train is not present. From the information in the
paragraph we must conclude that there is only a 70% chance of an apprehension.
32. Insufficient information. The passage does not give information about the number of
illegal aliens who attempt to cross the bridge and that number cannot be inferred from
any other information in the passage. Therefore, there is no way to estimate whether
more aliens attempted to cross when no train was present or when a train was present.

111

Unit II - Answers to Self Tests

1. True
2. True
3. Some weapons are guns.
Anything that is not a weapon is not a gun.
4. No weapons are guns.
Some guns are not weapons.
5. Invalid
6. Valid
7. Invalid
8. Valid

Part II.A
Section II.A.1
1. (not a set) rewritten as a set: individuals who are listening
2. a set
3. (not a set) rewritten as a set: hazardous materials (or acts)
4. (not a set) rewritten as a set: forms that are completed and filed
5. a set
6. computer software, user friendly products
7. law enforcement personnel, people who carry guns
8. Mexicans, people who were deported
9. Border Patrol Agents, people who speak Spanish
10. Service employees, people who are responsible for immediately reporting any allegation of
misconduct
Section II.A.2
1.
courses filled to
capacity for the next
three months
supervisory
training
courses

Every supervisory training

course is filled to capacity
for the next three months.

112
2.

very old cases

Some of the cases being
handled by this office are
very old.

cases
handled
by this
office

3.

old computers
computers
equipped with
Windows
software

None of the old computers

is equipped with Windows
software.

4.

working
computers

computers
on our
system

A few computers on our

system are not working.

113
Section II.A.3
1a. False. This conclusion contradicts the information in the original statement.
1b. True. If all members of our staff have mailboxes at the front desk, then it must be true
that there are some people who have mailboxes at the front desk who are members of
our staff.
1c. Insufficient information. The original statement does not tell us if the members of our
staff are the only people who have mailboxes at the front desk. It may be that members
of some other staff also have mailboxes in the same place. (This form is the converse,
which in this case, you may recall, is a common reasoning mistake, or illogical bias.)
1d. True. Since all members of our staff have mailboxes at the front desk, it must be true
that anyone who does not have a mailbox there is not a member of our staff. (This is
the form called the contrapositive, which was referred to in Logic Note 2.)
2a. Insufficient information. The original statement tells us about every supervisory
training course in the next three months, but it does not tell us about every course that is
filled to capacity. Therefore, we do not have enough information to conclude that every
course that is filled to capacity for the next three months is a supervisory training
course. (This form is again the converse, which in this case is a reasoning error.)
2b. False. This conclusion is contrary to the information in the original statement.
2c. True. Since all the supervisory training courses are filled to capacity for the next three
months, it must follow that none of them is filled to less than capacity. This conclusion
represents a double negation of the original statement. (This form is called the obverse,
which was referred to in Logic Note 3.)
2d. True. Since all the supervisory training courses are filled to capacity for the next three
months, it must follow that any course that is not filled to capacity is not a supervisory
training course. (This form is called the contrapositive, which was referred to in Logic
Note 2.)
Section II.A.4
1a. False. This conclusion contradicts the information in the original statement.
1b. True. If none of the new Jeeps are being deployed to Station 1, it must be true that none
of the vehicles being deployed to Station 1 is a new Jeep. The two sets new Jeeps and
vehicles being deployed to Station 1 are completely excluded from each other. This
is the form known as the converse.

114
1c. True. Since none of the new Jeeps are being deployed to Station 1, it must be true that
all of the new Jeeps are among the vehicles that are not being deployed to Station 1.
This conclusion represents a double negation of the original statement. (This form is
called the obverse, which was referred to in Logic Notes 3 and 4.)
1d. False. This conclusion is contrary to the information in the original statement.
2a. False. This conclusion is contrary to the information in the original statement.
2b. True. Since none of the Assistant Chiefs will be able to attend the meeting, it follows
that none of the people who will attend the meeting are Assistant Chiefs. This
conclusion is valid because the two sets (Assistant Chiefs, people who will be able to
attend the meeting) are completely separate from each other. This form is called the
converse.
2c. False. This conclusion contradicts the information in the original statement.
2d. True. Since none of the Assistant Chiefs will be able to attend the meeting, it follows
that all of the Assistant Chiefs are unable to attend the meeting. This conclusion
represents a double negation of the original statement. (This form is called the obverse,
which was referred to in Logic Notes 3 and 4.)
Section II.A.5
1a. True. Since some DHS employees are trained in the use of firearms, it must be true that
there are some people trained in the use of firearms that are employees of DHS. This is
the form known as the converse.
1b. Insufficient information. The original statement referred to only some DHS employees.
It does not give us enough information to draw a conclusion about all DHS employees.
1c. False. This conclusion contradicts the information in the original statement.
1d. True. Since some DHS employees are trained in the use of firearms, it must be true that
there are some DHS employees that are not untrained in the use of firearms. This
conclusion represents a double negation of the original statement (the obverse).
2a. Insufficient information. The original statement only gives information about some
Canadian citizens. It does not tell us about whether or not there are some Canadian
citizens who are ineligible to work in the United States.

115
2b. True. Since some Canadian citizens are eligible to work in the United States, it must be
true that these Canadian citizens are not ineligible to work in the United States. This
conclusion represents a double negation of the original statement (the obverse) and is
synonymous with it.
2c. Insufficient information. The original statement does not give information about all
Canadian citizens, so we do not have enough information to know if this conclusion is
true or not.
2d. True. Since some Canadian citizens are eligible to work in the United States, it must be
true that some people who are eligible to work in the United States are Canadian
citizens. This form is the converse.
Section II.A.6
1a. False. This conclusion clearly contradicts the information in the original statement.
1b. Insufficient Information. The original statement said that I received some documents
that were not in acceptable form. It said nothing about whether or not I also received
some documents that were in acceptable form.
1c. True. This statement is equivalent to the original statement. It represents a double
negation of the original statement (the form called the obverse).
1d. Insufficient Information. The original statement only said I received some documents
that were not in acceptable form. We do not know whether or not that refers to all the
documents that were received. It might be that I received some documents that were in
acceptable form.
2a. Insufficient information. The original statement said that there were some computers in
our office that cannot run the new software. That statement does not tell us which
computers can run the software or indeed if there are any computers that can run it. If
you concluded that this statement is true, you have fallen into an illogical bias (a
prevalent reasoning error). This form is the converse of the original statement.
2b. Insufficient information. The original statement does not give us information about all
the computers in our office. The statement only mentions some computers.
Therefore, we cannot draw a conclusion that applies to all of the computers in the
office.
2c. Insufficient information. The original statement says that some computers in our office
cannot run the new software. That information is not sufficient in itself to let us
conclude that there are some computers in our office that can run the software. If you

116
chose True for this conclusion, you committed an illogical bias, a common reasoning
mistake.
2d. False. This conclusion contradicts the information in the original statement.
Section II.A.7
1a. Insufficient information. We know it is not true that none of the computers can run the
new software. Therefore, it must be true that at least some of them can run the
software. However, we do not have enough information to conclude that all of the
computers in our office can run the new software.
1b. True. See explanation for a above.
1c. Insufficient information. See explanation for a above. Just as we do not have enough
information to know if a is true, we also do not have enough information to know if c is
true. Given the initial statement, either a or c must be true. However, we do not have
enough information to know which is true.
2a. False. If it is not true that some of the employees in our office received awards last
year, then it must be true that none did. It certainly is false to conclude that all of them
did.
2b. True. See a above.
3a. Insufficient information. Since it is not true that all of the passengers boarding the
plane have proper documentation of U.S. citizenship, it must be that some of them do
not have such documentation. However, we do not have enough information to
conclude that none of the passengers have proper documentation.
3b. Insufficient information. See a above. Even though it is not true that all passengers
boarding the plane have proper documentation, it may be that some of them do.
However, we do not have sufficient information to draw that conclusion.
3c. True. See a above.
4a. True. If it is not true that some DHS managers have not completed the training course,
then it must be true that all of them have completed it.
4b. False. If it is not true that some DHS managers have not completed the training course,
then it is clearly also not true that none of them have completed the course.

117

Part II.B
Section II.B.1
For each pair of statements below, underline the middle term and then write a valid
conclusion relating the other two terms.
1.

Premise 1: All DHS employees are Federal Government employees.

Premise 2: All ICE employees are DHS employees.
Possible Conclusion: All ICE employees are Federal Government employees. or
Some Federal Government employees are ICE employees.
Explanation: The set ICE employees is completely included in the set DHS
employees, which is in turn completely included in the set Federal Government
employees. The second conclusion can be deduced from the first.

2.

Premise 1: No Canadian citizens are U.S. citizens.

Premise 2: All citizens of Quebec are Canadian citizens.
Possible Conclusion: No citizens of Quebec are U.S. citizens. or
No U.S. citizens are citizens of Quebec.
Explanation: The first premise says that the set of Canadian citizens is completely
separate from the set of U.S. citizens. The second premise says that all citizens of
Quebec are completely included in the set of Canadian citizens. Therefore, all
citizens of Quebec must be completely excluded from the set of U.S. citizens. The
converse of this statement (see Section II.A.4) is also valid.

3.

Premise 1: No District 1 staff has completed the new training.

Premise 2: All recently hired adjudicators have completed the new training.
Conclusion: No recently hired adjudicators are on the District 1 staff. or
No one on the District 1 staff is a recently hired adjudicator.
Explanation: According to the second premise, all recently hired adjudicators have
completed the new training. However, from the first premise, it can be deduced that
no one who has completed the new training is on the District 1 staff. Therefore, it
must be true that none of the recently hired adjudicators are on the District 1 staff.
The converse of this statement (see Section II.A.4) is also valid.

4.

Premise 1: Some agents are trained in Advanced CPR.

Premise 2: All employees trained in Advanced CPR passed the Basic CPR course.
Conclusion: Some agents have passed the Basic CPR course. or
Some people who passed the Basic CPR course are agents.

118
Explanation: From the second premise we know that all employees trained in
Advanced CPR passed the Basic CPR course, and from the first premise we know that
some agents are trained in Advanced CPR. Therefore, we can conclude that some
agents have passed the Basic CPR course. We can also conclude the valid converse
(see Section II.A.5), that some people who passed the Basic CPR course are agents.
5.

Premise 1: All ICE vehicles are listed in the vehicle database.

Premise 2: Some Department of Homeland Security vehicles are not ICE vehicles.
Conclusion: From the information contained in these premises, no conclusion is
possible about the relationship between DHS vehicles and vehicles listed in the ICE
vehicle database.
Explanation: No valid conclusion can be drawn from these two premises. The second
premise tells us that some DHS vehicles are not ICE vehicles. The first premise tells us
that all ICE vehicles are listed in the ICE vehicle database, but it gives us no
information about non-ICE vehicles. Therefore, we can draw no further conclusion
about DHS vehicles.

6.

Premise 1: All persons born in the U.S.A. are U.S. citizens.

Premise 2: All persons born in the U.S.A. can get social security cards.
Conclusion: Some people who can get social security cards are U.S. citizens. or
Some U.S. citizens can get social security cards.
Explanation: From the second premise, we can conclude that some people who can get
social security cards are born in the U.S. When that information is combined with the
information in the first premise--that all persons born in the U.S. are U.S. citizens--we
can conclude that some people who can get social security cards are U.S. citizens. We
can also conclude the valid converse (see Section II.A.5), that some U.S. citizens can
get social security cards.

Section II.B.2
1.

Invalid. The term Federal Government employees is distributed in the conclusion but
not in the premises--a violation of rule 5.

2.

Valid. The conclusion is negative in accordance with rule 2. The middle term is
distributed in premise 2 (rule 4). Both sets are distributed in the conclusion and in the
premises (rule 5).

3.

Valid. Premise 1 refers to only part of a set, so the conclusion refers to only part of a
set (rule 7). Both premises are affirmative, so the conclusion is affirmative (rule 3).
The middle term is distributed in premise 2 (rule 4). Neither term is distributed in the
conclusion or in the premises (rule 5).

119
4.

Invalid. The premises are both affirmative but the conclusion is negative--a violation of
rule 3; also, the term under the jurisdiction of the DHS is distributed in the conclusion
but not in the premises--a violation of rule 5.

Section II.B.3
1.

There is insufficient information to draw this conclusion. From the converse of the
second premise (see Section II.A.3), assuming that there were people who failed to
qualify last week, we can conclude that there were some employees who must
qualify with their firearm this week. These people are the employees who failed to
qualify last week. However, there may be other employees who need to qualify this
week (for example, employees who were not tested last week). Therefore, we can
only conclude that some of the employees who have to qualify with their firearm this
week are not Special Agents.
A valid conclusion is: Some employees who have to qualify with their firearm this
week are not Special Agents.

2.

There is insufficient information to draw this conclusion. It could be that there are
other entries into the service system besides requests for computer assistance. This
erroneous conclusion stems from committing the illogical bias of assuming that the
converses of premise 1 and premise 2 are valid (see Section II.A.3). We only have
enough information to conclude that some entries in the computer system are Help
Desk requests and that some Help Desk requests are for computer assistance.
A valid conclusion is: Some entries into the service system are requests for computer
assistance.

3.

Valid.

4.

Invalid. No valid conclusion about the space assigned to Jim and the space on this
floor can be drawn because the two premises only refer to undefined parts of sets.
There is no way to tell if the space assigned to Jim is the same Logistics office space
that is on this floor.

Part II.C
Section II.C.1
1a. The Department of Homeland Security both confers benefits and enforces laws.
1b. Not a conjunction

120
1c. To be successful, a teletraining instructor must maintain eye contact with the viewers
and must involve viewers in interactive exercises.
1d. A complete computer set-up includes a processor, a display, a keyboard, and a mouse.
1e. Not a conjunction
2a. The new vehicles do not have four-wheel drive.
2b. It is not the case that both Agent Lpez and Agent Johnson are on annual leave today.
2c. It is not true that the agency will use both classroom training and computer-based
training in the future.
2d. Neither you nor I will attend the meeting.
2e. It is not true that the ceremony will be canceled if it rains.
Section II.C.2
1a. If you contribute to the Combined Federal Campaign through payroll deduction, then
(antecedent)
there is a record of a deduction on your biweekly earnings statement.
(consequent)
1b. If an international flight arrives, Inspectors process the arriving passengers.
(antecedent)
(consequent)
1c. You can take the advanced supervisory course only if you have taken the basic
(antecedent)
(consequent)
supervisory course.
1d. This is not a conditional. It is an alternation, which will be described in the next
section.
1e. Whenever an entire office undergoes a move, there is an inevitable period of disruption.
(antecedent)
(consequent)

121

2a. If you are a CBP Inspector, then you work for DHS.
2b. When someone is hired into the Border Patrol, that person always attends training in
Glynco.
2c. A person can receive e-mail messages only if his or her computer is connected to a
network.
2d. If you study a foreign language enthusiastically, you will develop a good vocabulary in
that language.
3a. Insufficient information. This conclusion is based on the invalid converse of the
conditional. This conclusion rests on the assumption that if a computer has a modem, it
was purchased for someone on the A-Team. The original conditional sentence does not
permit us to draw that conclusion.
3b. True. Because Johns computer was purchased for the A-Team, it does have a modem.
3c. True. Because all computers purchased for the A-Team have modems, any computer
that does not have a modem was not purchased for the A-Team. (This form is the
contrapositive; see Logic Note 1.)
3d. Insufficient information. The original conditional statement only refers to computers
that were purchased for the A-Team. It says nothing about computers purchased for
other teams. They may or may not have modems.
4a. Insufficient information. The conditional sentence only tells what will happen if we
request the same spending level as last year. It does not say what will happen if we
request a higher level.
4b. True. The original conditional statement told us that our budget would be approved if
we requested the same spending level as last year. Since our budget was not approved,
we must not have requested the same spending level. (This form is the contrapositive;
see Logic Note 1.)
4c. False. This statement contradicts the original conditional statement.
4d. True. If the original conditional statement is true, then this statement is true. This
statement says that the antecedent of the conditional is true. Therefore, the consequent
must also be true.

122

Section II.C.3
1a. Conditional
1b. Disjunction
1c. Alternation
1d. Biconditional
1e. Conditional
2a. Insufficient information. It is possible that the Jeep has both a faulty starter and bad
spark plugs. The fact that it has a faulty starter does not rule out the possibility that it
has bad spark plugs.
2b. Insufficient information. The Jeep might have only one of the two problems. Having
bad spark plugs does not necessarily mean that it also has a faulty starter.
2c. True. If the Jeep does not have bad spark plugs, it must have a faulty starter. The
original statement said that the Jeep had either one or the other of these two problems.
If it does not have one of the problems, it must have the other one.
2d. True. See explanation for c.
3a. True. The branch does not have both a color printer and a black-and-white printer. If
the branch has a color printer, it does not have a black-and-white printer.
3b. Insufficient information. Although the branch does not have both a color printer and a
black-and-white printer, we do not know for sure if the branch has either of these two
pieces of equipment. Therefore, if a branch does not have one piece of equipment, the
same branch may or may not have the other piece.
3c. Insufficient information. See explanation for b.
3d. False. This must be false because a branch does not have both a color printer and a
black-and-white printer.

123

Section II.C.4
1.

Insufficient information. The form represents the invalid converse of the compound
conditional (If p, then q and r; q and r, therefore p). The original statement said that if
the consultants recommendations were implemented, then we would achieve two
results. However, the statement did not say that the only way to achieve these results
was to implement the consultants recommendations. It could be that we would achieve
the same results by pursuing another course of action. Therefore, knowing that we
achieved these results is not enough to let us conclude that the consultants
recommendations were implemented.

2.

True. According to the conditional statement, implementing the consultants

recommendations would lead to certain results. If the results are not achieved, then the
consultants recommendations must not have been implemented.

3.

True. The antecedent of the conditional is true. Therefore, the consequent must be
true.

4.

Insufficient information. Just stating that a result was achieved is not sufficient to let us
conclude that the consultants recommendations were implemented. That result could
have been achieved for another reason.

5.

True. This says that the plan to be implemented will not achieve the results expected
from implementing the consultants recommendations. Therefore, the plan must not
include implementing these recommendations. (This conclusion is the contrapositive;
see note 3.)

124

Unit III - Answers to Self-Tests

Part III.A
1.
2.
3.
4.
5.
6.
7.

True
False
True
False
False
True
True

8a.
8b.
8c.
8d.

Deductive, good reasoning

Inductive, poor reasoning
Deductive, good reasoning
Inductive, good reasoning

Part III.B
Section III.B.1
1a. Unknown. Since many illegal aliens are not observed, the staff officer will need to
draw conclusions about these unobserved individuals and also project future trends,
which cannot be known at present.
1b. Known. Personnel records contain this information for all DHS employees.
1c. Known. Total expenditures and expenditures for travel are documented for each fiscal
year.
1d. Unknown. The recruitment specialist has no way to find out about the reading habits of
all college seniors. She must draw conclusions based on college seniors who apply for
DHS jobs.
2a. Correct inference. Ninety percent of the new agents in Sector X are new hires.
Therefore, Agent A.B. is very likely to be a new hire.
2b. Incorrect inference. Eighty percent of the new agents in Sector Y are new hires.
Therefore, it is not very likely that C.D. transferred from another sector. There is only a
20% chance that this happened.

125
2c. Incorrect inference. The paragraph does not give information about what percent of all
new Border Patrol hires are in Sector Y. Therefore, we cannot say that it is likely that
E.F. works there.
2d. Correct inference. Ten percent of the new agents in Sector X transferred from another
sector.
3a. Correct inference. Since three-fourths of the nonofficer crew members are not from
Norway, it must be that one-fourth are from Norway. Therefore, any nonofficer crew
member has a one-fourth chance of being from Norway.
3b. Correct inference. See explanation for a.
3c. Incorrect inference. We do not know the proportion of the total crew that are officers.
Therefore, we cannot estimate the chances that a Norwegian crew member is an officer.
3d. Incorrect inference. Since all of the officers are from Norway, any crew member who is
not from Norway is also not an officer (this is a deductive conclusion).
4a. Correct inference. There is a 50% chance that there would be a tour for a new
employee, but there is also a chance (probability unknown) that there would be a tour
for official visitors.
4b. Incorrect inference. The paragraph does not tell us about the relative frequency of tours
for new employees and for official visitors. Therefore, we cannot say if a tour on
Monday is more likely to be for a new employee or for official visitors.
4c. Correct inference. Since the probability was 50% that there would be a tour for a new
employee, an upper limit of 50% is set on the chance that there would not be a tour.
Depending on the frequency of official visitors, the chance that there would be no tours
might be even lower.
4d. Correct inference. On any Monday there was a 50% chance that there would be a new
employee. Therefore, there was a 50% chance that there would not be a new employee.
5a. Correct inference. The paragraph tells us that all of the computer equipment, which
constitutes 30% of the items, has not been declared surplus. It is possible, perhaps even
likely, that there are other items that have not been declared surplus. Therefore, there is
at least a 30% chance that a randomly selected piece of equipment has not been
declared surplus.

126
5b. Correct inference. Since the probability is at least 30% that a randomly selected piece
of equipment has not been declared surplus, then the probability that it has been
declared surplus cannot exceed 70%.
5c. Incorrect inference. None of the computer equipment has been declared surplus.
Therefore, any piece of equipment that has been declared surplus could not be a piece of
computer equipment.
5d. Incorrect inference. There is at least a 30% chance that this piece of equipment is
computer equipment, given that 300 pieces of computer equipment had not been
declared surplus.
6a. Not an inductive generalization. Squares are defined as having four sides.
6b. An inductive generalization based on the observations of patterns of moss growth.
6c. An inductive generalization based on observations of the harm done to bodies by
exposure to radiation.
6d. Not an inductive generalization. As of 1996, 50 governmental entities had been legally
declared states.
Section III.B.2
1a. Correct inference. A randomly selected employee from the work site had a 30% chance
of being an undocumented alien. If the person was an undocumented alien, then he or
she was not from a North, South, or Central American country. In addition, other
workers from the site might not be from a North, South, or Central American country.
Therefore, there is at least a 30% chance that a randomly selected employee at that site
was not from a North, South, or Central American country.
1b. Incorrect inference. Based on the second sentence in the paragraph, one must conclude,
by deduction, that no employee from that site who was from a North, South, or Central
American country was an undocumented alien.
1c. Correct inference. Since at least 30% of the employees (the undocumented aliens) are
not from a North, South, or Central American country, the probability that a randomly
selected employee was from the Americas was not more than 70%.
1d. Incorrect inference. Although none of the undocumented aliens was from a North,
South, or Central American country, the country of origin of the other 70% of the
employees is unknown.

127
2a. Incorrect inference. The paragraph does not give us sufficient information to draw a
conclusion about the exact distribution of new and holdover furniture in the
redecorated area.
2b. Correct inference. Only 20% of the items of furniture in the visitors area before
redecoration were made of wood, which was required for use after redecoration.
Therefore, there was only a 20% chance that an item of furniture that was used before
the redecoration could have been used after the redecoration.
2c. Correct inference. Eighty percent of the furniture in the visitors area before the
redecoration did not have wood frames and thus did not meet the requirement for
furniture in the redecorated area. In addition, pieces of furniture from before the
redecoration may not be used for other reasons. Thus, there is at least an 80% chance
that an item of furniture in the room before redecoration would not be used after the
redecoration.
3a. Correct inference. A case from Agent Zs caseload had a 50% chance of being a case of
document counterfeiting and thus being investigated vigorously. In addition, it is
possible, no doubt very likely, that there were other types of cases that were
investigated vigorously. Thus, the chance that a randomly selected case is investigated
vigorously is 50% or greater.
3b. Correct inference. All cases that are vigorously investigated have not failed to yield a
successful prosecution. Since this case has at least a 50% chance of being investigated
vigorously, it also has at least a 50% chance of not failing to produce a successful
prosecution.
3c. Incorrect inference. The chance that this case was not investigated vigorously is no
more than 50%, but it could be less than 50% (compare to the explanation for a).
3d. Incorrect inference. The chance that this case was prosecuted unsuccessfully is not
more than 50%, but it could be less than 50% (compare to explanation for b).

Part III.C
Section III.C.1
1a. Correct conclusion. Vehicle A is a typical vehicle apprehended while transporting
illegal aliens. Therefore, according to the paragraph, there is a 20% chance that it was
stopped for speeding.

128
1b. Incorrect conclusion. Vehicle B was transporting illegal aliens but was not
apprehended. The paragraph does not give us any information about the probability that
such a vehicle was not stopped for speeding.
1c. Correct conclusion. If vehicle A had a 20% chance of being stopped for speeding, it
had an 80% chance of not being stopped for speeding. These two probabilities are the
complements of each other.
1d. Incorrect conclusion. The paragraph does not give us information about the universe of
vehicles stopped for speeding. We only know that some of them were transporting
illegal aliens. We do not know what the percentage is.
2a. Incorrect conclusion. The paragraph does not give us information about the record of
returns for all books signed out for purposes other than specific work projects.
Therefore, we cannot assign a definite value to the probability that such a book will be
returned on time.
2b. Correct conclusion. The paragraph tells us that of all books returned on time, 70% were
signed out for use on specific work projects. Therefore, we can estimate that 700 of
1000 such books were returned on time.
2c. Correct conclusion. The paragraph tells us that if a book was signed out for use on a
specific work project, there is a 70% chance that it will be returned on time. Therefore,
we can estimate that 350 of 500 such books will be returned on time.
2d. Incorrect conclusion. The paragraph does not give us information on all the books that
were not returned on time. Therefore, we cannot estimate how likely it is that such a
book was signed out for a reason other than a work project.
Section III.C.2
1a. Correct conclusion Since one-tenth of the cases are criminal aliens, all of whom are not
detained at the DHS facility, there is a lower limit on the percentage of cases that would
not be detained there and hence on the probability that any case involved someone who
was not detained there.
1b. Correct conclusion. This conclusion is the complement of the conclusion in a. If the
chances of not being detained in an DHS detention facility are at least 10%, then the
chances of being detained at such a facility can be no more than 90%.
1c. Incorrect conclusion. This conclusion is false because the probability that a case
concerns a noncriminal alien is exactly .90, which is 1 minus .10 (the probability that a
case concerns a criminal alien).

129
2a. Correct conclusion. Since there is a 90% chance that there was a record of a special
deduction on an employees earnings statement, there is a 10% chance that there is no
special deduction. This sets a lower limit on the chance that an employee did not
contribute to the Combined Federal Campaign through payroll deduction. It is assumed
that there could be a special deduction on the earnings statement for some other reason
(such as an allotment for a bond purchase), so the probability of not contributing could
be greater than .10).
2b. Incorrect conclusion. This is not an accurate statement of the probability that a
randomly selected employee did contribute to the Combined Federal Campaign through
payroll deduction. That probability is somewhere in the range of 0 to .90.
2c. Correct conclusion. This is the complement of the probability that a randomly selected
employee did have a special deduction on the earnings statement.
3a. Incorrect conclusion. This conclusion is false because the correct probability is exactly
25%.
3b. Correct conclusion. This is the complement of the correct version of answer a.
3c. Correct conclusion. The employee worked more than eight hours on one-fourth of her
work days and these are the only days eligible for overtime.
4a. Incorrect conclusion. The 40% of the work stations that were not fixed had received
neither the new hardware nor the upgraded software. If they had received either of
those two items, the problems would have been corrected.
4b. Incorrect conclusion. There is insufficient information to draw this conclusion. The
60% of the work stations that were fixed would have been fixed if they had received
either of these new items. There is no way to tell if any of them had received both
items.
4c. Correct conclusion. The 40% of work stations that were not fixed clearly had received
neither the new hardware nor the upgraded software. There is also the possibility that
some of the work stations whose problems had been corrected had been fixed through
some other means (for example, reconfiguration by a skilled technician). That is why
the probability is at least .40.

130

Unit IV - Answers to Self-Tests

Part IV.A
Section IV.A.1
1. While it is tempting to give the information provided by the two coworkers more weight
than the results of the survey, it is unlikely that the coworkers friends represent all
people who own the car John is interested in purchasing. According to the law of large
numbers, a survey of 1,000 car owners is more likely to be representative of the
population of car owners than are the coworkers friends. Since the results of the survey
suggest that most people are satisfied with the car, John should consider purchasing the
car.
2. In this problem, Center B is more likely to have days on which at least 7.5% of the
claims processed are incorrect. According to the law of large numbers, small samples
are more likely to provide extreme observations than large samples. Therefore, over the
course of the year, Center B is more likely to have days in which 7.5% of the claims
processed are incorrect. Similarly, Center B is also more likely to have days on which
2.5% or fewer of the claims processed are incorrect. However, when the number of
incorrect claims processed at each plant is averaged over the course of a year, both
Center A and Center B are likely to process incorrect claims 5% of the time. This occurs
because days on which very few incorrect claims are processed at Center B will be offset
by days on which many incorrect claims are processed.
Section IV.A.2
1. The sample used in the newspaper poll is more likely to be accurate for a number of
reasons. First, even though the poll conducted by the newspaper excluded voters who do
not have telephones, it is more likely to be representative of the population of voters
because it was randomly selected. On the other hand, the local news program poll only
included viewers of the program. Second, the local news program poll required viewers
to pay $.50 in order to participate. This restricted the local news program poll sample to
viewers who were willing to pay $.50 to participate in the poll.
This example highlights the fact that we are often forced to draw conclusions from
flawed samples. Understanding the rules of statistical reasoning helps us to maximize
our chance of drawing an accurate conclusion even when we are provided with less than
ideal samples (in some circumstances you may feel you cannot draw an accurate
conclusion given the samples provided).

131
2. You may have noticed that both samples are small and violate the law of large numbers.
Ideally, Lisa should continue to collect information before making an evaluation.
However, in order to focus on sample representativeness, assume that Lisa cannot collect
more information. In this case, Lisa should use her coworkers sample to estimate the
percentage of employees in the district willing to work overtime because his sample
included employees working during many shifts in key areas within the district. On the
other hand, Lisas sample was restricted to employees working during one shift in a
specific area within the district. This reduces the probability that her sample reflected
the population of employees in the district.
Section IV.A.3
1. Laura was right to be skeptical of the effectiveness of the course. Since the class
consisted of a group of students who performed poorly on the test, the increased scores
may be the result of regression to the mean. Remember, a students test score is a
function of two factors: his or her actual ability and chance factors. Even if his or her
ability remains constant, a students score will most likely be different on the second
administration of the test because chance factors impact his or her score on both
administrations. The first time a student who scored extremely low took the test, chance
factors were likely to have had an negative effect on performance (e.g., the student may
have been sick when taking the test). The impact of chance factors during the second
administration of the test is not likely to be as negative, resulting in the students
receiving a higher test score.
Stronger evidence would have been provided if the class had consisted of low, average,
and high scoring students. Laura could be confident in the effectiveness of the course if
all students scored higher on the second administration of the test, regardless of their
score on the first administration. On the other hand, if only low scoring students
improved their test scores, the increase could be attributed to regression to the mean.
2. Since the relationship between the comedians stand-up routine and his acting ability is
not perfect, Greg and Jane should not be surprised by the comedians performance. For
the most part, good stand-up comedians will be good actors. Occasionally, a good standup comedian will be only a fair actor. Therefore, Greg and Jane should be disappointed
(they just spent $7.00 each to see the movie), but not surprised that the stand-up
comedian was not a great actor.
Section IV.A.4
1. Julias claim that the most expensive stereo system is best violates the rule of sample
validity. The difference in price could be due to a number of factors other than quality.
For example, a stereo system may be more expensive because of higher shipping costs.
Therefore, Julias claim that most expensive stereo system is better may be incorrect.
She should have based her assessment on attributes known to be related to quality.

132
2. While the sample used in the survey of overall satisfaction satisfies the law of large
numbers and sample representativeness, David should not use the results to draw
conclusions about employee satisfaction with pay. Overall satisfaction can be attributed
to a number of factors other than pay including the nature of the job, relationships with
coworkers, and the quality of supervision. Therefore, using overall job satisfaction
ratings to assess satisfaction with pay may lead David to draw an erroneous conclusion.

Part IV.B
Section IV.B.4
1. While each individual part has a 90% chance of working, all three parts must be working
correctly simultaneously for the car to run properly. According to the multiplicative rule
for independent events, the probability of the car running properly is determined by
multiplying the individual probability that each part will work (90% X 90% X 90%). As
a result, there is a 73% chance that the car will run properly when using the three rebuilt
parts. Therefore, Loukas estimate was correct.
2. Jennifer failed to take into account the fact that the probability of a persons being
offered a position decreases after a job offer has been given to another person. As a
result, Jennifer incorrectly applied the multiplicative rule for independent events to
determine the probability that she and Bob will be offered positions at the marketing
firm. In this problem, there is a 50% chance of getting a job offer given that no job
offers had been given. However, there is only a 33.33% chance of getting a job offer
given that one job offer has already been made (3 applicants for 1 job opening).
Applying the multiplicative rule for dependent events would have resulted in Jennifers
determining that there is a 16.7% (50% X 33.33%) chance that she and Bob will both get
job offers from the marketing firm.
Section IV.B.5
1. Using the additive rule for independent events, Donna should choose to make an
appointment at Office A. Given the information provided, there is a 50% chance (35% +
15%) that Donna will be assigned to either an internist or a specialist at Office A. On
the other hand, there is only a 45% chance (40% + 5%) that Donna will be assigned to
either an internist or a specialist at Office B.

133
2. Much to Sharons dismay, she may find that her assessment was wrong. Remember that
the additive rule for dependent events must be applied when events are related. While
there are many possible combinations of the office having ample parking space and/or
being near transportation, below is a table containing one.21

Ample Parking
Limited Parking
Total

Near Transportation
25%
25%
50%

Not Near Transportation

25%
25%
50%

Total
50%
50%
100%

Where:
p(ample parking, near transportation) = 25%
p(ample parking, near transportation) = 25%
p(limited parking, not near transportation) = 25%
p(limited parking, not near transportation) = 25%

Sharon was correct in estimating that there is a 50% chance the office has ample parking
and a 50% chance the office is near public transportation. However, there is a 25%
chance that the office has both ample parking and is near public transportation, a
probability common to both events. In order to account for the overlap between the two
events, it is necessary to subtract the probability of the offices having ample parking
and being close to public transportation. This results in the following formula:
p(ample parking or near transportation) =
p(ample parking) + p (near transportation) - p(ample parking , near transportation)

Thus, there is a 75% chance (50% + 50% - 25%) that the office has ample parking space
or is near public transportation.

21

There are a number of possible combinations of an office having ample parking space and/or being near
transportation. For example, if most offices that provide ample parking are located near public transportation,
a table such as the following might result:

Ample Parking
Limited Parking
Total

Near Transportation
10%
40%
50%

Not Near Transportation

40%
10%
50%

Total
50%
50%
100%

Notice that, while the percentages inside the table have changed, the total percentages have remained the same
(50% of offices are located near public transportation and 50% of the offices provide ample parking).

134

Unit V - Answers to Manual of Job-Related Thinking Skills Posttest22

1. Insufficient information. The second sentence in the paragraph states that all
naturalization ceremonies in the district are scheduled for Wednesdays and Thursdays.
This is not sufficient information to allow us to conclude that naturalization ceremonies
are the only events held on Wednesdays and Thursdays. It could be that other events are
also held on those days. (Deductive reasoning with two sets)
2. True. This conclusion follows from the second sentence of the paragraph. Since all
naturalization ceremonies in the district are held on Wednesdays and Thursdays, an
event that is held on any day other than Wednesday or Thursday is not a naturalization
ceremony. (Deductive reasoning with two sets)
3. False. This conclusion contradicts the information in second sentence. (Deductive
reasoning with two sets)
4. True. The third sentence in the paragraph states that no naturalization ceremonies are
closed to the public. This is the same as saying that all naturalization ceremonies are
open to the public. (Deductive reasoning with two sets)
5. True. This conclusion is equivalent in meaning to the fourth sentence, which states that
some meetings that are closed to the public are held in the district directors conference
room. (Deductive reasoning with two sets)
6. True. From the fourth sentence we can conclude that some meetings held in the district
directors conference room are closed to the public. In the third sentence, we learn that
no naturalization ceremonies are closed to the public. From these two statements we can
conclude that at least some events held in the district directors conference room -- the
ones that are closed to the public -- are not naturalization ceremonies. (Deductive
reasoning with three sets)
7. Insufficient information. This conclusion goes beyond the information in the paragraph.
The paragraph permits us to conclude that some events held on Wednesday or Thursday
-- the naturalization ceremonies -- are not closed to the public. It does not permit us to
conclude that no events held on those days are closed to the public. You might be led to
draw this erroneous conclusion if you interpreted the second sentence of the paragraph
as meaning that the only events held on Wednesdays and Thursdays are naturalization
ceremonies. (Deductive reasoning with three sets)

22

The type of thinking skill used in each question is given in parentheses at the end of each question.

135
8. Insufficient information. Question 6 represented the valid conclusion that some events
held in the district directors conference room are not naturalization ceremonies.
However, we are not justified by this conclusion or by any other information in the
paragraph in concluding that some naturalization ceremonies are not held in the district
directors conference room. (Deductive reasoning with three sets)
9. True. The last sentence of the paragraph states that if an Adjudications Officer suspects
fraud when interviewing an applicant for a benefit, a Special Agent is asked to join the
interview. From this statement we can conclude that if no Agent is asked to join the
interview, the applicant is not suspected of fraud. (Deductive reasoning with
connectives)
10. True. The second sentence says that an application is approved on the spot if and only if
the information provided by the applicant meets the legal criteria. From this statement
we can conclude that if the information does not meet legal criteria, the application is not
approved on the spot. (Deductive reasoning with connectives)
11. Insufficient information. We cannot draw the conclusion that, if an applicant is not
asked to supply additional information, the information already provided meets the legal
criteria. As stated in the third sentence, another course of action -- denial of the
application -- is also possible when the information does not meet the criteria.
(Deductive reasoning with connectives)
12. True. The second sentence tells us that an application is approved on the spot only if the
information provided meets legal criteria. Therefore, we can conclude that if an
application is approved on the spot, the information provided must meet the legal
criteria. (Deductive reasoning with connectives)
13. False. This contradicts the second sentence of the paragraph, which states that if the
information provided meets legal requirements, the application is approved on the spot.
(Deductive reasoning with connectives)
14. Insufficient information. The paragraph does not give us enough information to draw
this conclusion. The last sentence says that a Special Agent is asked to join the
interview if fraud is suspected. The paragraph does not indicate that an Agent would be
called anytime an individuals application fails to be approved on the spot. (Deductive
reasoning with connectives)
15. Insufficient information. The paragraph does not say that an Agent is asked to join an
interview only in cases of suspected fraud. There may be other situations in which a
Special Agent is asked to join an interview. Therefore, we do not have enough
information to decide that this conclusion is either true or false. (Deductive reasoning
with connectives)

136
16. True. The third sentence states that if the information provided by an applicant does not
meet legal criteria, then one of two actions is taken: either the applicant is asked to
provide additional information or the application is denied. Therefore, if the information
does not meet legal criteria for approval and the application is not denied, it must be that
the Adjudications Officer has asked for more information. (Deductive reasoning with
connectives)
17. True. The fourth sentence of the paragraph says that all DHS laptops should have access
control software. From this statement, we can conclude that some of the computers that
should have access control software are DHS laptops. (Deductive reasoning with two
sets)
18. True. This conclusion is equivalent to saying that all DHS laptops should have access
control software, which is the same as the information in the fourth sentence of the
paragraph. (Deductive reasoning with two sets and connectives)
19. Insufficient information. The paragraph says that all DHS laptops should have access
control software. It does not state that laptops are the only type of computers that should
have this software. It could be that desktop computers also should have access control
software. (Deductive reasoning with two sets)
20. False. The third sentence of the paragraph clearly says that unauthorized access occurs
when individuals without proper clearance obtain access to computers or data. Access to
data alone is thus considered unauthorized access. (Deductive reasoning with
connectives)
21. True. The third sentence says that unauthorized access occurs when individuals without
proper clearance obtain access to computers or data. Therefore, if there has been no
unauthorized access, it must mean that no individuals without proper clearance obtained
access either to DHS computers or to DHS data. (Deductive reasoning with connectives)
22. Insufficient information. The last sentence of the paragraph says that if a laptop
computer is being checked as baggage, it must be secured in a specialized storage
container. However, the paragraph does not tell us whether or not a specialized
container is also needed when the computer is not being checked as baggage. (Deductive
reasoning with connectives)
23. Insufficient information. We are able to conclude that some computers that must be
secured in specialized storage containers also have access control software; these are
DHS laptops that are being checked as baggage. However, we do not have enough
information to make the general conclusion that all computers that must be secured in
specialized storage containers also have access control software. We might reach this
overly general conclusion if we erroneously concluded that all computers that must be

137
secured in specialized storage containers are DHS laptops. But the paragraph does not
give us the information to draw that conclusion. (Deductive reasoning with sets and
connectives)
24. True. The fourth statement says that all DHS laptops should have access control
software. This is equivalent to saying that no DHS laptop should be without access
control software. (Deductive reasoning with connectives)
25. True. The third sentence states that nearly all illegal immigrants apprehended while
crossing the areas of rough terrain were 16 years of age or older. Therefore, an illegal
alien intercepted in rough terrain is most likely to be 16 years of age or older. (Inductive
reasoning with two sets)
26. Insufficient information. The second sentence says that nearly all illegal immigrants
under the age of 16 were intercepted at highway border crossings rather than in areas of
rough terrain. Therefore, it is clear that some illegal immigrants under the age of 16
were intercepted away from areas of rough terrain. If an illegal alien was not
apprehended in an area of rough terrain, we cannot necessarily conclude that the alien
was not under the age of 16. (Inductive reasoning with sets and connectives)
27. Insufficient information. Although we know that nearly all illegal immigrants under the
age of 16 were intercepted at highway border crossings, we do not know if individuals
under the age of 16 constitute the majority of illegal aliens intercepted at these crossings.
Therefore, we do not have enough information to conclude that an illegal alien
apprehended at a highway border crossing was more likely to be under the age of 16
than to be 16 years of age or older. (Inductive reasoning with two sets)
28. Insufficient information. Although the third sentence tells us that nearly all the illegal
immigrants apprehended while trying to cross the area of rough terrain were 16 years of
age or older, we do not know if this group constituted the majority of individuals 16
years of age or older who were apprehended while trying to enter the sector. Therefore,
we do not have enough information to conclude that nearly all illegal aliens 16 years of
age or older apprehended while trying to enter this sector tried to cross the area of rough
terrain. (Deductive reasoning with two sets)
29. True. The second sentence says that nearly all of the illegal immigrants under the age of
16 who were apprehended were intercepted at the highway border crossings. Therefore,
we can conclude that if an illegal immigrant under the age of 16 is apprehended, it is
most likely to be at a highway border crossing. (Inductive reasoning with two sets)
30. True. During the last month of the study, according to the last sentence of the
paragraph, the number of illegal immigrants apprehended in rough terrain was twice
as great as the number apprehended at highway border crossings. Since those

138
apprehended while crossing areas of rough terrain are nearly all 16 years of age or older,
it follows that an illegal alien apprehended while entering this sector was more likely to
be 16 years of age or older than to be under the age of 16 during the last month of the
study. (Inductive reasoning with three sets)
31. False. Since apprehensions were twice as high in the rough terrain as at border
crossings, and since those apprehended in rough terrain were nearly all 16 years of age
or older, it follows that an illegal immigrant 16 years of age or older was more likely to
be apprehended in rough terrain. (Inductive reasoning with three sets)
32. Insufficient information. The paragraph does not give any information about illegal
aliens who were not apprehended. (Inductive reasoning with two sets)

139

Glossary of Key Terms

Additive rule for dependent events refers to a rule of probability that is used to determine
the probability that at least one of a number of dependent events will occur.
Additive rule for independent events refers to a rule of probability that is used to determine
the probability that at least one of a number of independent events will occur.
Alternation refers to a form in the logic of connectives which says Either statement p is
true or statement q is true or both statements p and q are true.
Ambiguity refers to the situation in which a word or set of words is used or is understood to
have more than one meaning.
Anchoring and adjustment heuristic refers to the tendency to make slight adjustments from
an initial reference point as new information is provided.
Antecedent refers to the part of a conditional sentence which, if true, necessitates that the
other part of the sentence (the consequent) be true. When a conditional sentence is
stated in if...then form, the antecedent is the if part of the sentence.
Availability heuristic refers to the fact that people often base their evaluations on the event
that is most easily retrieved from memory.
Base rate (also called the prior probability) refers to the number of times an event has
occurred in the past.
Biased sample refers to samples that do not reflect the population on relevant
characteristics.
Biconditional refers to a form in the logic of connectives which says If statement p is true,
then statement q is true; and if statement q is true, then statement p is true. Another
way of stating the biconditional is: Statement p is true if and only if statement q is true.
Complements are two numbers which when added together equal 1.
Compound conditional refers to a conditional form in which another basic connective is
embedded as either the antecedent or consequent. An example of a compound
conditional is: If either statement p or statement q is true, then statement r is true.
Concrete information refers to the tendency to give observed information more weight than
abstract information, even if the abstract information is more reliable.
Conditional probability refers to the probability that one event will occur given that some
other event has occurred.

140
Conditional refers to a form in the logic of connectives which says If statement p is true,
then statement q is true.
Conjunction refers to a form in the logic of connectives which says that two or more
statements are true at the same time.
Connectives are words such as and, if...then, and either...or, that express relationships
between simple statements.
Consequent refers to the part of a conditional sentence which must be true if the other part
of the sentence (the antecedent) is true. When a conditional sentence is stated in
if...then form, the consequent is the then part of the sentence.
Contrapositive refers to a logical form in which the positions of two parts of a sentence are
reversed and the two parts are negated. The contrapositive of the statement All
members of set A are members of set B is All nonmembers of set B are nonmembers
of set A. The contrapositive of the conditional sentence If statement p is true then
statement q is true is If statement q is not true then statement p is not true. The
contrapositive is a valid conclusion for both of these examples.
Converse refers to a logical form in which the positions of two parts of a sentence are
reversed. The converse of the statement All members of set A are members of set B is
All members of set B are members of set A. The converse of the conditional sentence
If statement p is true then statement q is true is If statement q is true then statement p
is true, but this is not a valid conclusion for these two forms. It is valid for No
members of set A are members of set B and for Some members of set A are members
of set B.
Deductive reasoning refers to reasoning that leads to a conclusion that must be true if the
evidence on which it is based is true.
Degree of confirmation refers to the interpretation of probability in inductive logic. The
probability of a conclusion is the degree of confirmation provided for the conclusion by
the premises.
Dilution refers to the tendency to make a conservative estimate of the probability that an
event will occur when irrelevant information is provided along with relevant
information.
Disjunction refers to a form in the logic of connectives which says Statement p and
statement q cannot both be true.
Distributed term refers to a term or set that is referred to in its entirety in a premise. For
example, in the statement All members of set A are members of set B, set A is a
distributed term because the statement refers to all members of the set.

141
Event refers to the phenomenon you are interested in (e.g., performance of trainees, number
of employees satisfied with their job).
Extended conditional refers to a form in the logic of connectives which has two conditional
premises. The antecedent of the conclusion is from one of the premises and the
consequent is from the other premise. An example of an extended conditional is: If
statement q is true, then statement r is true; and if statement p is true, then statement q is
true; therefore, if statement p is true, then statement r is true.
Fallacy refers to a mistake in reasoning, that is to say, drawing a conclusion that is not
justified by the evidence.
False conclusion (see Invalid conclusion, below).
Forgetful induction refers to the inductive fallacy of neglecting to use some the available
evidence when drawing an inductive conclusion.
Gamblers fallacy refers to the tendency to overestimate the probability of the occurrence of
an event not previously observed.
Hasty induction refers to the inductive fallacy of drawing an inductive conclusion on the
basis of very limited evidence.
Heuristic refers to a speculative formulation of a solution to a problem. In statistical
reasoning it refers to a rule of thumb that people use when making estimates of the
likelihood that an event will occur.
Ignoring relative frequency refers to the tendency to make estimates based on observed
frequencies rather than on relative frequencies. This occurs when information on
nonoccurrences is ignored or not available.
Illogical bias refers to any very common reasoning mistake.
Illusory correlation refers to the tendency to perceive that two variables are related even
though a relationship between them does not exist.
Inductive analogy refers to a type of informal reasoning that involves drawing a conclusion
about a member of a set based on its similarity to members of other sets.
Inductive generalization refers to the use of specific observations as the basis for making a
general rule.
Inductive reasoning refers to reasoning with incomplete information to draw a conclusion
that is probabilistic, that is, a conclusion that may be false even though it is compatible
with the evidence.
Insufficient information (see Invalid conclusion, below).

142
Invalid conclusion refers to a conclusion that is not justified, given the available evidence.
Invalid conclusions either are false or are conclusions for which there is insufficient
information. False conclusions are contrary to the available evidence. Conclusions for
which there is insufficient information are conclusions that cannot be drawn because
there is not enough information to know if they are true.
Inverse refers to a logical form in which the two parts of a sentence are negated. The
inverse of All members of set A are members of set B is All nonmembers of set A
are nonmembers of set B. The inverse of If statement p is true then statement q is
true. is If statement p is not true then statement q is not true. In both of these
examples, the inverse is not equivalent in meaning to the original statement.
Middle term refers to the set that is found in both of the premises in a syllogism. Because
the middle term is found in both premises, it is the term of comparison for deducing the
relationship between the other two sets.
Misperceptions about equiprobability refers to the tendency to define incorrectly the
probability that two or more events will occur as equal when in fact they are not.
Multiplicative rule for dependent events refers to a rule of probability that is used to
determine the probability that two or more dependent events will occur.
Multiplicative rule for independent events refers to a rule of probability that is used to
determine the probability that two or more independent events will occur.
Negate a complex connective sentence refers to the action of stating that a complex
connective sentence is not true. For example, the negation of the sentence If you
complete classroom training and on-the-job training, you will have finished the basic
requirements is It is not true that if you complete classroom training and on-the-job
training, you will have finished the basic requirements.
Negate a conjunction refers to the action of stating that a conjunction is not true. For
example, the negation of the sentence Both the Attorney General and the
Commissioner will attend the ceremony is It is not true that both the Attorney General
and the Commissioner will attend the ceremony.
Negate a simple sentence refers to the action of stating that a simple sentence is not true.
For example, the negation of the sentence All members of set A are members of set B
is It is not true that all members of set A are members of set B.
Obverse refers to a logical form in which the verb of a sentence is negated and its predicate
term is negated. For example, the obverse of the sentence All members of set A are
members of set B is All members of set A are not non-members of set B (or its
equivalent No members of set A are nonmembers of set B). The obverse of a
statement is equivalent in meaning to the original statement.

143
Order effect refers to the tendency to make estimates based solely on initial or recent
observations.
Overestimating conjunctive events refers to the tendency to overestimate the probability
that two events will occur.
Population refers to the entire collection of the events of you are interested in collecting
information about. In everyday language, the term population refers to people; however
in statistics it can refer to any set of events or things in which you are interested (e.g.,
computers and cars).
Premise refers to any evidence that is used to support a conclusion. In reasoning with three
sets, the two statements from which the conclusion is drawn are usually called the
premises.
Probability refers to a number between 0 and 1 that indicates the likelihood that an event
will happen or that a conclusion is true. See also degree of confirmation.
Pure fallacies of induction are the fallacies of forgetful, hasty, and slothful induction. They
represent errors of induction because they entail not using all of the available evidence
when drawing a conclusion and/or not attributing the degree of probability to the
conclusion which is justified by the evidence.
Random sampling refers to an approach to collecting a sample where all possible events in
the population have an equal chance of being observed.
Regression to the mean refers to the fact that, on average, an extreme observation of an
event is likely to be followed by a less extreme observation of the same event in the
future. As a result, subsequent observations of an event are likely to be more similar to
the typical occurrence in the population. Regression to the mean can also refer the fact
that, on average, an extreme observation of one event will be associated with a less
extreme observation of a different, but related event.
Relative frequency refers to the number of times an event of interest occurs relative to other
possible events. Relative frequency provides an estimate of the probability that the
event of interest will occur.
Representativeness heuristic refers to the tendency to assign an event to a category based on
the degree to which the event resembles other members of the category. For example,
you may assume that a person driving a Porsche is wealthy. It also can refer to the
tendency to believe that event A resulted from process B. For example, you may
assume that a jobless person does not have a job because he or she is lazy.

144
Sample representativeness refers to the fact that samples that are similar to the population in
terms of characteristics relevant to the event you are measuring are more likely to be
representative of the population.
Sample validity refers to the extent to which a sample of observed events reflects accurately
what interests you about the population.
Selective perception refers to the tendency to notice or remember events that are consistent
with a stereotype, and ignore events that are inconsistent with a stereotype.
Set refers to a grouping of individuals or things that share one or more characteristics.
Slothful induction refers to the inductive fallacy of failing to draw an inductive conclusion
in spite of strong evidence in support of the conclusion.
Stereotype refers to a general description of the characteristics that are used to define
members of a category.
Stratified random sampling refers to an approach to collecting a sample events are sampled
randomly from sub-populations identified on the basis of key characteristics of the
population.
Syllogism refers to a logical form in reasoning with three sets. The syllogism consists of
two premises and a conclusion. Each premise is a statement about the relationship
between two sets. The two premises have one set in common (the middle term or term
of comparison) which is used for deducing the relationship between the other two sets.
Term of comparison (see Syllogism, above).
Total available evidence refers to the requirement in inductive reasoning that a conclusion
should be based on all the available evidence that is relevant to the conclusion.
True conclusion (see Valid conclusion, below).
Universe of discourse refers to the range of individuals, events, or things about which a
conclusion is drawn in inductive or deductive reasoning.
Valid conclusion refers to a conclusion that is justified, given the evidence.