bookSHelf
Introduction to Statistics
Independent or mutually exclusive?
Section 3.2 · Probability Topics
Introduction to Statistics · Section 3.2 Independent or mutually exclusive?
Two ideas that are frequently interchanged — and they are not two names for the same thing.
They are not two words for one idea. They answer two different questions.
S = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10}
Mutually exclusive — can both happen at once?
A = {1,2,3,4,5}    C = {7, 9}
A is the low half. C is two of the high numbers.
P(A AND C) = 0
Mutually exclusive means the overlap is empty.
Change one set, and the answer changes
B = {4,5,6,7,8}
B is the middle five — it reaches into A.
A AND B = {4, 5} = 2 ÷ 10 = 0.2
0.2 0 not mutually exclusive
Independent — does knowing one change the other?
P(A | B) = P(A)
Learning that B happened moves your estimate of A not at all.
What “given” actually does
delete every outcome outside the condition re-scale what is left
P(A) = 0.50
D = {4,5,6,7}
D is the middle four — it also reaches into A.
P(A | D) = 0.50 = P(A)
The world shrank. Your belief about A did not move. That is independence.
The same test, in arithmetic
P(A) × P(D) = 0.50 × 0.40 = 0.20
P(A AND D) = 0.20 they agree
Overlapping, like D — and yet
B = {4,5,6,7,8}    A AND B = {4, 5}
P(A | B) = 0.40 P(A)
The bar moved. B is not independent of A.
P(A) × P(B) = 0.50 × 0.50 = 0.25
P(A AND B) = 0.20 they disagree
Same overlap as D, opposite answer. Overlap alone decides nothing.
Same operation, different set
P(A | C) = 0
Your belief about A went from a half to nothing.
bar unchanged vs bar destroyed
Not neighbours on a scale. The two opposite ends of it.
Two questions, not one
overlap empty? vs does knowing change it?
Mutually exclusive events are the least independent events there are.
One test each
Mutually exclusive P(A AND B) = 0
Independent P(A AND B) = P(A) × P(B)
  • Ten equally likely outcomes — every claim here is countable.
  • A and C — the bands never share an outcome.
  • A and B — both bands sit under 4 and 5.
  • Given D: four outcomes left, two of them A — still half.
  • Given B: five outcomes left, two of them A — no longer half.
  • Given C: two outcomes left, and none of them A.

Key ideas

  • Mutually exclusive P(A AND B) = 0
    They cannot both happen.
  • Default assumption Assume not mutually exclusive until you check.
  • Independent P(A | D) = P(A)
    Knowing D changes nothing.
  • In arithmetic P(A AND D) = P(A) × P(D)
    Multiply, then compare.
  • Overlap decides nothing D and B share the same two outcomes — only one is independent.
  • The trap Mutually exclusive never independent.