Independence Versus Mutual Exclusivity

Two Different Questions About Events

When two events are being considered, there are two distinct questions: Can they occur together? and Does knowing that one occurred change the probability of the other? The first question concerns mutual exclusivity; the second concerns independence. Confusing these ideas can lead to incorrect calculations, because events that cannot occur together are generally dependent rather than independent.

For example, on one roll of a fair six-sided die, the events “roll a 1” and “roll a 2” cannot both happen. But if the roll is known to be a 1, the probability that it is a 2 is no longer $1/6$; it is zero. That change is evidence of dependence. By contrast, results on two separate fair coin tosses can include heads on both tosses, and learning the first toss was heads does not change the probability of heads on the second.

Definitions

An event is a set of outcomes in a probability model. For events $A$ and $B$, the event $A\cap B$ consists of the outcomes that belong to both $A$ and $B$. The event $A\cap B$ is empty exactly when there are no outcomes in common.

Definition: Events $A$ and $B$ are mutually exclusive (or disjoint) if $A\cap B=\varnothing$. They are independent if $$ P(A\cap B)=P(A)P(B). $$

The independence definition compares the probability that both events occur with the product of their separate probabilities. If the equality holds, the events do not alter each other's probabilities in this specific probabilistic sense. For events with positive probability, this is equivalent to saying that conditioning on either event leaves the probability of the other unchanged. For instance, if $P(B)>0$, then

P(A\mid B)=P(A) \quad\Longleftrightarrow\quad \frac{P(A\cap B)}{P(B)}=P(A) \quad\Longleftrightarrow\quad P(A\cap B)=P(A)P(B).

The conditional-probability version requires $P(B)>0$, because conditional probability given $B$ is defined by dividing by $P(B)$. The product definition of independence is the general definition and remains applicable even when one event has probability zero.

Mutual Exclusivity Is Not the Same as Independence

Mutual exclusivity is about whether the events share any outcomes. Independence is about whether a particular probability equality holds. Neither definition should be substituted for the other. In particular, mutual exclusivity is defined by $A\cap B=\varnothing$, not merely by $P(A\cap B)=0$. An intersection can contain outcomes that have probability zero.

The relationship between mutual exclusivity and independence follows directly from their definitions. If $A$ and $B$ are mutually exclusive, then $P(A\cap B)=0$. For them also to be independent, it must be true that

0=P(A\cap B)=P(A)P(B).

A product of two nonnegative probabilities is zero only if at least one of them is zero. Therefore, mutually exclusive events that both have positive probability are dependent. If one event has probability zero, mutual exclusivity alone does not make the events dependent; the product condition can still hold.

The converse distinction matters too: independent events need not be mutually exclusive. They can occur together, and when both have positive probability, independence requires their joint probability to equal the positive product of their probabilities.

Worked Examples

Worked Example: Mutually Exclusive Events That Are Dependent

A fair six-sided die is rolled once. Let $A$ be the event that the result is 1, and let $B$ be the event that the result is 2. Determine whether the events are mutually exclusive and whether they are independent.

The sample space is $\{1,2,3,4,5,6\}$, with each outcome having probability $1/6$. The events are $A=\{1\}$ and $B=\{2\}$. They have no outcome in common:

A\cap B=\varnothing,\qquad P(A\cap B)=0.

Thus, the events are mutually exclusive. To test independence, compare the joint probability with the product of the marginal probabilities:

P(A)P(B)=\left(\frac{1}{6}\right)\left(\frac{1}{6}\right) =\frac{1}{36}, \qquad P(A\cap B)=0.

Since $0\ne 1/36$, the events are not independent; they are dependent. In conditional terms, $P(A\mid B)=0$, whereas $P(A)=1/6$. Knowing that the result is 2 changes the probability of a 1.

Worked Example: Independent Events That Are Not Mutually Exclusive

A fair coin is tossed twice. Let $A$ be the event that the first toss is heads, and let $B$ be the event that the second toss is heads. Determine whether the events are independent and whether they are mutually exclusive.

The four equally likely outcomes are $HH, HT, TH,$ and $TT$, where the first letter is the first toss. Event $A=\{HH,HT\}$, event $B=\{HH,TH\}$, and their intersection is $A\cap B=\{HH\}$. Therefore,

P(A)=\frac{2}{4}=\frac{1}{2},\qquad P(B)=\frac{2}{4}=\frac{1}{2},\qquad P(A\cap B)=\frac{1}{4}.

The independence check gives

P(A)P(B)=\left(\frac{1}{2}\right)\left(\frac{1}{2}\right) =\frac{1}{4}=P(A\cap B).

The events are independent. They are not mutually exclusive, because $HH$ belongs to both events. The result of the first toss does not prevent heads on the second toss.

Worked Example: A Nonempty Intersection With Probability Zero

Consider a probability model with outcomes $\{z,r,s\}$, where $P(\{z\})=0$, $P(\{r\})=0.3$, and $P(\{s\})=0.7$. Let $A=\{z\}$ and $B=\{z,r\}$. Determine whether the events are mutually exclusive and whether they are independent.

The events share the outcome $z$, so their intersection is not empty:

A\cap B=\{z\}\ne\varnothing.

Thus, the events are not mutually exclusive. However, the shared outcome has probability zero. The event probabilities are

P(A)=0,\qquad P(B)=0.3,\qquad P(A\cap B)=P(\{z\})=0.

Using the product definition,

P(A)P(B)=(0)(0.3)=0=P(A\cap B).

So $A$ and $B$ are independent, despite sharing an outcome. This example shows why “the intersection has probability zero” is not, in general, the definition of mutual exclusivity. It also illustrates why the conditional form $P(B\mid A)$ cannot be used here: $P(A)=0$, so that conditional probability is undefined.

A Reliable Way to Classify Two Events

It helps to answer the two questions separately. First check whether the events share outcomes; then check the probability-product equality. Do not decide that events are independent just because they seem unrelated in words, and do not decide they are mutually exclusive merely because their joint probability is zero.

Identify the event intersection.
Write $A\cap B$, the outcomes that satisfy both event descriptions. The events are mutually exclusive exactly when this set is empty.

Find the relevant probabilities.
Calculate $P(A)$, $P(B)$, and $P(A\cap B)$ from the probability model or data provided.

Test independence.
Compare $P(A\cap B)$ with $P(A)P(B)$. Equality means independent; inequality means dependent.

State both conclusions.
Report mutual exclusivity and independence separately. The events may be mutually exclusive, independent, neither, or—in cases involving zero-probability events—both.

Why the Distinction Matters

The multiplication rule for independent events is $P(A\cap B)=P(A)P(B)$. It is valid when independence has been established, not merely because two events have different names or describe different features. For mutually exclusive events with positive probabilities, the joint probability is zero, while the product $P(A)P(B)$ is positive. Using the independence rule in that situation would therefore give the wrong joint probability.

Conversely, if a problem says two events are mutually exclusive, it tells you that they cannot both occur: $A\cap B=\varnothing$. It does not tell you that they are independent. When both events have positive probabilities, mutual exclusivity tells you they are dependent, since observing one rules out the other.

Key takeaway: Mutual exclusivity is a statement about an empty intersection: $A\cap B=\varnothing$. Independence is a statement about probabilities: $P(A\cap B)=P(A)P(B)$. If mutually exclusive events both have positive probability, they are dependent.

Common Mistakes

Calling mutually exclusive events independent. The mistaken reasoning is that if events are separate, one has no effect on the other. But if mutually exclusive events both have positive probability, occurrence of one rules out the other. Their joint probability is zero, not the positive product of their marginal probabilities.
Defining mutual exclusivity by zero intersection probability. The condition $P(A\cap B)=0$ says that the intersection has probability zero. It does not guarantee that the intersection contains no outcomes. Mutual exclusivity requires $A\cap B=\varnothing$.
Using conditional probability when the conditioning event has probability zero. The expression $P(A\mid B)$ is defined by division by $P(B)$, so it is unavailable when $P(B)=0$. Use the product definition of independence in that case.
Checking only one relationship. “Independent” does not mean “mutually exclusive,” and “mutually exclusive” does not by itself state whether an event has probability zero. Check the set intersection and the probability equality separately.

Check Your Understanding

For each question, distinguish the event-set relationship from the probability relationship.

On one fair six-sided die roll, $A$ is “the result is odd” and $B$ is “the result is greater than 4.” Are the events mutually exclusive? Test whether they are independent using their probabilities.
Suppose $A$ and $B$ are mutually exclusive and $P(A)=0.25$, $P(B)=0.40$. Find $P(A\cap B)$ and determine whether the events are independent.
Two events satisfy $P(A)=0.3$, $P(B)=0.5$, and $P(A\cap B)=0.15$. Are they independent? Can you determine from these probabilities alone whether their intersection is empty?
Explain why $P(A\cap B)=0$ does not always establish that $A$ and $B$ are mutually exclusive.
If $A\cap B=\varnothing$ and $P(A)=0$, must $A$ and $B$ be dependent? Use the product definition to justify your answer.