Events: and, or, not

37 min · 5 drills

In the textbook: Ch. 5.1 Grundbegriffe

This article opens the chapter on probability. In the exam the chapter brings almost four of the twenty tasks on average, more than any other chapter (table in the article on the exam). The tasks use the words “and”, “or”, “not”. We will turn each of them into a step with numbers: what to add and what to subtract. The most common mistake comes with “or”: two chances cannot simply be added when the events overlap, or the common part is counted twice.

A probability says how often something happens if the same thing is repeated many times: a throw of a die, picking a random customer, testing a part.

The game. An ordinary die is thrown, and the token moves forward if the number is even or greater than 3, that is 4, 5 or 6. Otherwise it stays put.

With what probability does the token move forward?

It is tempting to add the two chances. Even faces are 3 of 6, the faces 4, 5 and 6 are also 3 of 6, and 36+36=1: the token would seem to move every time. But if a one comes up, it stays put, so the sum counted something extra.

Where exactly? Let us go through all six faces and answer both questions for each:

FaceEven?Greater than 3?Token moves?
1nonono
2yesnoyes
3nonono
4yesyesyes
5noyesyes
6yesyesyes

The token moves on 2, 4, 5 and 6, that is in 4 cases out of 6, and the probability is 46=23. The sum gave 6 out of 6 because the faces 4 and 6 answer “yes” to both questions and sit in both chances. The extra 26 are exactly these two faces.

This is the addition rule: for the probability of “one or the other”, add the two chances and subtract once the shared part where both hold. On the die: 36+36−26=46. Every “or” in the exam is computed the same way. Below we write the rule with letters and go through the traps.

Check yourself. If the rule is already familiar, here is a task at exam level: in a shop, 50 % of the customers buy bread, 40 % buy milk and 25 % buy both. What share of the customers buys at least one of the two?

Outcomes and events

A throw of the die ends in one of six results: 1, 2, 3, 4, 5 or 6 comes up. Each such result is called an outcome (German: Ergebnis). The die is fair, no face comes up more often than another, so the outcomes are equally likely.

All outcomes together are written with the Greek letter Ω (omega). The outcomes are listed, separated by commas, in curly brackets:

Ω={1,2,3,4,5,6}

The rule of the game asks two questions about a throw: is the number even, and is it greater than 3? The answer “yes” to each question comes with some outcomes and not with others. The set of outcomes for which the answer is “yes” is called an event (German: Ereignis) and is written with a capital letter.

A dieEach outcome is drawn as a face of the die, a circle gathers the faces of one event.

all six outcomes of the throw: Ω

A die. Outcomes as dots in a box, events as regions over them. A holds 2, 4, 6.Ω
Step 1 of 2All outcomes

We throw the die: 1, 2, 3, 4, 5 or 6 can come up. These are all the outcomes, and their set is written Ω.

If you throw the die very often, each face comes up in about a sixth of the throws. An even number comes up in about three sixths, that is in half. This share is called the probability of the event and is written P(A), “the probability of the event A”: P for probability.

For equally likely outcomes the probability of an event is the number of its outcomes, ∣A∣, divided by the number of all outcomes, ∣Ω∣:

P(A)=∣A∣∣Ω∣=36=0.5

A probability always lies between 0 and 1:

  • zero belongs to an event without a single outcome;
  • one belongs to an event that contains all outcomes.

And, or, not

The words “and”, “or”, “not” build a new event out of two. The new event is again a set of outcomes. The example below lights them up one by one on the same die.

A dieCompare where the lit faces lie: in the common part of the circles, in at least one circle, or outside.

a 4, 5 or 6 comes up, that is B: three of six faces

A die. Outcomes as dots in a box, events as regions over them. A holds 2, 4, 6; B holds 4, 5, 6. Lit: B, 3 of 6 outcomes.ΩAB|B| = 3
Step 1 of 5B: a 4, 5 or 6 comes up

The same outcomes Ω and the same circle A, “an even number comes up”. The rule's second question: a number greater than 3 comes up: 4, 5 or 6. This is the event B, B={4,5,6}, the second circle. B holds three outcomes too, ∣B∣=3, and P(B)=36=0.5.

An event has occurred if one of its outcomes came up. If a 4 comes up, both A and B have occurred. If a 5 comes up instead, only B has occurred.

German exam tasks write these events in words. These phrases come up most often:

In the taskEventOn the dieIn words
“A und B”, “sowohl A als auch B”, “beides”A∩B4, 6both happen
“A oder B”, “mindestens eines”A∪B2, 4, 5, 6at least one happens
“nicht A”A′1, 3, 5A does not happen
“weder A noch B”, “keines”A′∩B′1, 3neither happens
“genau eines”A∪B without A∩B2, 5exactly one happens

The last row of the table, “genau eines” (exactly one): one of the two events occurs, but not both at once. Take A∪B and remove the common part A∩B; on the die that leaves 2 and 5.

Your turn1 of 5

The task says: “Mit welcher Wahrscheinlichkeit kauft ein Kunde weder Brot noch Milch?” Let A be “the customer buys bread” and B “the customer buys milk”. Which event is asked for?

Overlapping or not: disjoint events

Let us add a third rule to the game: the token goes back to the start on a 1 or a 2, which is the event C={1,2}. Let us compare it with the two earlier ones, first with B.

A dieThe circles B and C do not overlap: no face lies in both.

all six outcomes: Ω

A die. Outcomes as dots in a box, events as regions over them. B holds 4, 5, 6; C holds 1, 2.Ω
Step 1 of 4All outcomes

Again one throw of the die and the same six outcomes Ω: 1, 2, 3, 4, 5, 6.

Now let us compare C with A.

A dieDo circles A and C have a common part?

A: a 2, 4 or 6 comes up; C: a 1 or 2 comes up

A die. Outcomes as dots in a box, events as regions over them. A holds 2, 4, 6; C holds 1, 2.ΩAC
Step 1 of 2A and C

The same outcomes Ω, now with circles A “an even number comes up” and C “a 1 or a 2 comes up”.

A and C: the two lies in both circles. There is only one common outcome, but there is one, so A and C can occur together.

An event and its complement are always disjoint: A and A′ split all outcomes into two parts with nothing in common.

In a task you recognise disjoint events from the wording: “everyone pays in exactly one way”, “the customer chooses one plan”, “a part goes into exactly one class”. If there is no such sentence and a number greater than zero is given for “beides”, the events overlap.

Your turn2 of 5

A die is thrown once. Which two events are disjoint?

Probability of A or B: the addition rule

Now we compute the probability of “or”: how to add two chances when the events share outcomes. The exam often asks for the rule we are about to derive in reverse as well. This is how such a task looks:

Exam-style task

Eine Prüfung besteht aus einem Test und einer Hausarbeit, die unabhängig voneinander bestanden werden. Den Test bestehen 62 % der Teilnehmer. Mindestens eines von beiden bestehen Teilnehmer mit der Wahrscheinlichkeit 0.8062. Mit welcher Wahrscheinlichkeit besteht ein Teilnehmer die Hausarbeit?

  1. a)0.61
  2. b)0.19
  3. c)0.49
  4. d)0.68
  5. e)0.30

“At least one of the two” is the union A∪B, and this time its probability is given. What is wanted is one of the two single chances, the one for the term paper. The task does not state the intersection A∩B as a number; it hides in the words “independently of each other”. We first derive the rule on the die and on a hotel, and solve the task itself afterwards in example 4.

At the start we saw on the die that the sum 36+36 counts the faces 4 and 6 twice. Now we write the same thing with events.

A dieWhere are the faces that the sum 3 + 3 counted twice?

an even number came up, A: three faces

A die. Outcomes as dots in a box, events as regions over them. A holds 2, 4, 6; B holds 4, 5, 6. Lit: A, 3 of 6 outcomes.ΩAB|A| = 3
Step 1 of 4First A

The same outcomes Ω and the same circles: A “an even number comes up”, B “a 4, 5 or 6 comes up”.

The sum 36+36=1 added the outcomes of A and those of B. A has 3 outcomes, ∣A∣=3.

In a mountain hotel, 60 % of the guests book breakfast, 75 % the spa and 41 % both. What share booked at least one of the two? The outcomes are now the guests. Imagine 100 guests: 60 book breakfast, 75 the spa, 41 both. Percentages are shares of these 100 guests, and the rule works just as it did with the faces of the die.

Hotel guestsHow many guests booked at least one service?

A: breakfast, 60 guests; B: spa, 75 guests

Hotel guests. A number in each region of a box, events as circles over it. Events: A, B.ΩAB100
Step 1 of 4Name the events

All 100 guests together are Ω in the picture. Circle A is the guests who booked breakfast, circle B the guests who booked the spa. The number of guests in an event is written between vertical bars: ∣A∣=60 and ∣B∣=75, so P(A)=0.60 and P(B)=0.75.

“Both” is A∩B, the part the two circles share, 0.41. “At least one of the two” is A∪B, everything inside the circles.

Why is 41 subtracted once and not twice? Hint: how many times should a guest with both count in “at least one”?

For disjoint events there is nothing to subtract: the intersection is empty, P(B∩C)=0. The same rule simply gives the sum, P(B∪C)=36+26−0=56. Check by the outcomes: B∪C={1,2,4,5,6}, five outcomes. Disjoint events need no rule of their own, it is the same rule with a zero.

In a café 35 % of the guests pay cash, 50 % by card, the rest by phone. Every guest pays in exactly one way.

With what probability does a guest pay cash or by card?

Everyone pays in exactly one way, so “cash and card” does not happen: the events are disjoint, the intersection is 0.

35+50−0=85 %.

By phone pay 100−85=15 %. Together with the 85 % these are all guests.

This is the task from the sheet above. This time “at least one” is given, and the addition rule runs backwards: from the union we recover one of the two single probabilities.

Let A be “passes the test” and B “passes the term paper”. Given are P(A)=0.62 and P(A∪B)=0.8062; wanted is P(B).

Independent means that passing the test does not change anyone's chance on the term paper. Among the 62 % who pass the test, the same share P(B) passes the term paper as among all participants. So the intersection is a share of a share, P(A∩B)=0.62⋅P(B). Where this product rule comes from is shown in the section “Independent events: ‘and’, ‘or’ and an unknown share”.

0.8062=0.62+P(B)−0.62 P(B)=0.62+0.38 P(B)

The two terms with P(B) combine into (1−0.62) P(B)=0.38 P(B). Hence 0.38 P(B)=0.8062−0.62=0.1862 and P(B)=0.1862/0.38=0.49: option (c).

Computed forwards: 0.62+0.49−0.62⋅0.49=1.11−0.3038=0.8062, exactly the value in the task.

Try it yourself. A group has 30 students. 18 attend English, 12 Spanish, 5 both languages. How many students attend at least one language?

Your turn3 of 5

P(A)=0.3 and P(B)=0.4. In the first case A and B overlap, P(A∩B)=0.1. In the second case they are disjoint. When is P(A∪B) larger?

Your turn4 of 5

45 % of a bank's customers have a credit card, 30 % a mortgage, 20 % both. With what probability does a randomly chosen customer have at least one of the two?

Traps. The wrong options of the exam task with the test and the term paper come about like this:

  • Forgetting the intersection, option (b), 0.19: 0.8062−0.62=0.1862≈0.19 treats the test and the term paper as if they excluded each other. Yet one can pass both, and whoever passes both is already inside the 62 %.
  • Dividing by the wrong number, option (e), 0.30: 0.1862/0.62≈0.30. When the terms are combined, the factor next to P(B) is 1−0.62=0.38, not 0.62.
  • The product only with “unabhängig”: in the hotel example, 0.60⋅0.75=0.45 instead of the given 41 % would be wrong and would give 60+75−45=90. The intersection is taken as a product only when the task says “unabhängig voneinander”, as the exam task does. When the intersection is given as a number, use it.
  • Check by substituting: put an option into P(A)+P(B)−P(A) P(B). Only 0.49 gives 0.8062; (a) 0.61 gives 0.8518, (d) 0.68 gives 0.8784.

On the formula sheet: P(A∪B)=P(A)+P(B)−P(A∩B) (addition rule).

Disjoint does not mean independent

Disjoint events are often confused with independent ones. Here independence is needed only to tell the two apart. It is covered in full in the article “Independence: when probabilities multiply”. Toss a coin and roll a die. Whatever the coin shows, a six comes up with the same chance 16: the coin does not affect the die. Such events are called independent (German: stochastisch unabhängig): if you know that one has occurred, the chance of the other does not change. For them P(A∩B)=P(A)⋅P(B). Take the disjoint events B “a 4, 5 or 6 comes up” and C “a 1 or 2 comes up”.

A dieThe common part of the circles B and C is empty, but the product of their probabilities is not zero.

B and C share no face, yet the product is greater than zero

A die. Outcomes as dots in a box, events as regions over them. B holds 4, 5, 6; C holds 1, 2. Lit: B and C.ΩBC
Step 1 of 2Intersection and product

The same outcomes Ω, circle B around the faces 4, 5, 6 and circle C around the faces 1 and 2.

  • B and C never occur together, the intersection is empty: P(B∩C)=0;
  • the product: P(B)⋅P(C)=36⋅26=16, greater than zero.

A counterexample on the same die: A “even” and C “a 1 or a 2” overlap in the two.

A dieCompare which share of the faces in circle C is even and which share of all six faces.

A and C meet only in the two

A die. Outcomes as dots in a box, events as regions over them. A holds 2, 4, 6; C holds 1, 2. Lit: A and C.ΩAC
Step 1 of 2Intersection and product

The same outcomes Ω, circles A (the even faces) and C (the 1 and the 2).

  • A and C occur together only on a 2, the intersection: P(A∩C)=16;
  • the product: P(A)⋅P(C)=36⋅26=16.

In tasks this is confused in both directions. If the task says “unabhängig voneinander”, the intersection equals the product P(A)⋅P(B) and is greater than zero. With P(A)=0.6 and P(B)=0.5, “at least one” is 0.6+0.5−0.6⋅0.5=0.80, not 1.1 as it would come out without subtracting. A probability above 1 shows at once that the intersection was lost.

Your turn5 of 5

P(A)=0.3, P(B)=0.5, and the events A and B are disjoint. Which statement is true?

At a glance

The topic in a minute

  • Event: a set of outcomes. With equally likely outcomes P(A)=∣A∣/∣Ω∣.
  • And, or, not: A∩B both at once (“und”, “sowohl … als auch”), A∪B at least one (“oder”, “mindestens eines”), A′ everything except A (“nicht”), A′∩B′ none (“weder … noch”).
  • Disjoint events (“unvereinbar”) have no common outcomes, P(A∩B)=0. If both probabilities are greater than zero, they are dependent.
  • Addition rule: P(A∪B)=P(A)+P(B)−P(A∩B), because the intersection entered the sum twice. For disjoint events 0 is subtracted.
  • The addition rule for A or B

    Two events and “both” are given; the question asks “or”, “at least one”, “mindestens eines”.
    P(A∪B)=P(A)+P(B)−P(A∩B)

    Trap: multiplying P(A)⋅P(B) instead of using the given intersection.

When the question is “neither of the two”, it is easy to subtract the common part twice. How to avoid that is shown in “Neither of the two: probability through the complement”.

Now practise it

Events: and, or, not

Trainer · Probability

Practise
Next: Neither of the two: probability through the complement
0/5

Was this article helpful?