Probability
S1 Topic 3, spec point by spec point (Probability). Each spec point has the method, worked examples written the way the mark scheme wants them, how the marks go and the mistakes that cost the most marks.
Spec3.1Elementary probability
Use elementary probability.
The basics
- Probabilities are between 0 (impossible) and 1 (certain).
- For equally likely outcomes, \(P(A) = \dfrac{\text{number of outcomes in } A}{\text{total number of outcomes}}\).
- The probabilities of all possible outcomes add up to 1. So \(P(\text{not } A) = 1 - P(A)\).
- From data, a relative frequency estimates a probability: \(\frac{27}{48}\) of the beans weighed between 1.1 and 1.6 g, so \(P \approx 0.5625\).
complement and a missing value
Jeff wins a game with probability 0.74. \(P(\text{not win}) = 1 - 0.74 = 0.26\).
A Venn diagram's regions are 0.36, 0.24, \(x\), \(y\), 0.09 and \(z\) with \(x + z = 0.15\). Then \(y = 1 - 0.36 - 0.24 - 0.09 - 0.15 = 0.16\).
- B1 for a complement (WST01A Jun 2026 Q1c); M1 A1 for using "total = 1" to find a missing probability (WST01 Oct 2025 Q6).
Often asked as: one-mark starters; missing values in Venn diagrams and tables.
Probabilities that add to more than 1.
Spec3.2Sample spaces, exclusive and complementary events, conditional probability
Use sample spaces, mutually exclusive and complementary events, and conditional probability, including \(P(A \cup B) = P(A) + P(B) - P(A \cap B)\) and \(P(A \cap B) = P(A)P(B \mid A)\).
The key rules
\[P(A') = 1 - P(A) \qquad P(A \cup B) = P(A) + P(B) - P(A \cap B) \qquad P(B \mid A) = \frac{P(A \cap B)}{P(A)}\]
- Mutually exclusive: can't both happen, \(P(A \cap B) = 0\). Their circles don't overlap.
- Conditional probability \(P(B \mid A)\): the probability of \(B\) given that \(A\) has happened. The sample space shrinks to \(A\), so divide by \(P(A)\).
Venn diagrams
Fill them from the inside out: the triple overlap first, then the pairwise overlaps, then the "only" regions, then the outside. Every region must have a number (write 0 where empty).
a Venn diagram with three sets
Of 40 families, 12 have a dog (\(D\)), 14 a cat (\(C\)), 11 a rabbit (\(R\)), 9 have none, no family has a dog and a rabbit, and 2 have a dog and a cat. Complete the Venn diagram and find \(P(C \mid R)\).
\(D\) only \(= 10\), \(D \cap C = 2\). In total \(40 - 9 = 31\) families have at least one.
\(31 = 12 + 14 + 11 - 2 - n(C \cap R)\), so \(n(C \cap R) = 4\). Then \(C\) only \(= 14 - 2 - 4 = 8\) and \(R\) only \(= 11 - 4 = 7\).
\(P(C \mid R) = \frac{4}{11}\).
- Venn diagrams: separate B marks for the right overlaps (non-overlapping circles for exclusive events), the outside value, and each region (WST01 Jan 2020 Q2a).
- Conditional: M1 for the correct ratio \(\frac{P(\ldots \cap \ldots)}{P(\ldots)}\) with numerator smaller than denominator, A1 (WST01A Jan 2026 Q4d).
- "Which events are mutually exclusive?" needs the events named, not just \(P(B \cap C) = 0\) (WST01 Oct 2017 Q2a).
Often asked as: the most examined S1 probability point (97 questions). Venn diagrams with three events; conditional probabilities from Venn diagrams, tables or trees; trees with algebra (in terms of \(n\)).
\(P(B \mid A) = \frac{P(A \cap B)}{P(B)}\) (dividing by the wrong one).
Putting the total of \(C\) in the "\(C\) only" region.
Spec3.3Independent events
Use and test for independence: \(P(B \mid A) = P(B)\), \(P(A \mid B) = P(A)\), \(P(A \cap B) = P(A)P(B)\).
The idea
\(A\) and \(B\) are independent if knowing one happened doesn't change the probability of the other. Any one of these shows it:
\[P(A \cap B) = P(A) \times P(B) \qquad P(A \mid B) = P(A) \qquad P(B \mid A) = P(B)\]
To test: calculate both sides of \(P(A \cap B) = P(A)P(B)\) and compare, then state the conclusion.
Mutually exclusive and independent are different: exclusive events (with non-zero probabilities) can't be independent, because \(P(A \cap B) = 0 \neq P(A)P(B)\).
using independence
\(P(H) = \frac38\), \(P(H \cup W) = \frac34\), and \(H\) and \(W\) are independent. Find \(P(W)\).
\(\frac34 = \frac38 + P(W) - \frac38P(W) \Rightarrow \frac38 = \frac58P(W) \Rightarrow P(W) = \frac35\).
exclusive and independent together
\(A\) and \(C\) are mutually exclusive, \(B\) and \(C\) are independent, \(P(B) = 0.6\) and \(P(C) = 0.15\). Find \(P(B \cap C)\) and \(P(A \cap C)\).
\(P(B \cap C) = 0.6 \times 0.15 = 0.09\). \(P(A \cap C) = 0\).
- M1 for the addition law, M1 for \(P(H \cap W) = P(H)P(W)\) (must be used), A1 for an equation in \(P(W)\), A1 cso (WST01 Jun 2022 Q4a, WST01 Jan 2018 Q2c).
- For independent \(F\), \(G\): \(P(F \mid G) = P(F)\) (B1).
Often asked as: show events are (or aren't) independent with numbers; find a missing probability using independence; fill a Venn diagram with independence conditions.
Concluding without calculating both sides.
Treating "mutually exclusive" as "independent".
Spec3.4Sum and product laws: tree diagrams and Venn diagrams
Use the sum and product laws, tree diagrams and Venn diagrams, including sampling with and without replacement.
Tree diagrams
- Multiply along branches (AND).
- Add between final outcomes (OR).
- Each set of branches from one point adds to 1.
- Second-stage probabilities are conditional on the first stage.
Without replacement
The numbers change after each pick: e.g. 4 red and 3 blue, take a red, now 3 red and 3 blue.
Turn-taking games
"A wins on her second turn" means A fails, B fails, A succeeds: multiply those three.
tree diagram with an unknown
It rains with probability \(p\). If it rains, the probability Tom takes the bus is 0.4; if not, 0.15. Overall, \(P(\text{bus}) = 0.26\). Find \(p\) and \(P(\text{rain} \mid \text{bus})\).
\(0.4p + 0.15(1 - p) = 0.26 \Rightarrow 0.25p = 0.11 \Rightarrow p = 0.44\).
\(P(\text{rain} \mid \text{bus}) = \dfrac{0.4 \times 0.44}{0.26} = 0.677\).
without replacement
A bag has 4 red and 3 blue counters. Two are taken without replacement. Find \(P(\text{same colour})\) and \(P(\text{first red} \mid \text{same colour})\).
\(P(RR) = \frac47 \cdot \frac36 = \frac{12}{42}\), \(P(BB) = \frac37 \cdot \frac26 = \frac{6}{42}\). \(P(\text{same}) = \frac{18}{42} = \frac37\).
\(P(\text{first red} \mid \text{same}) = \dfrac{12/42}{18/42} = \dfrac23\).
- M1 for an equation from the tree with at least two terms in \(p\), dM1 for simplifying, A1 (WST01 Jun 2017 Q4a).
- "At least" questions: M1 for \(1 - P(\ldots)\), M1 for the right expression, A1 (WST01A Jan 2026 Q4b).
- Turn-taking: M1 for a product of three probabilities, A1 (WST01 Jun 2022 Q5e).
Often asked as: complete a tree (often in terms of \(n\), WST01 Oct 2024 Q7a), then find probabilities and conditional probabilities; sampling socks or counters until a condition is met.
Not reducing the numbers when sampling without replacement.
Adding along a branch instead of multiplying.