Two-way tables and multistage events
Reading probabilities from two-way tables, tree diagrams for multistage events, independent events and the product rule, and drawing with and without replacement.
Real questions rarely involve one coin. A survey sorts people by two things at once. A weekend has two days that might be wet. A pen comes out of a pencil case and doesn't go back in. Two-way tables and tree diagrams are the tools that keep these multistage situations organised, so you count everything once and nothing twice.
Two-way tables
A two-way table sorts a group by two categories. The totals at the ends of each row and column are your best friends: they tell you what to divide by.
A school surveyed 200 Year 12 students about whether they have a part-time job and whether they drive to school.
| Drives | Doesn't drive | Total | |
|---|---|---|---|
| Has a job | 46 | 64 | 110 |
| No job | 24 | 66 | 90 |
| Total | 70 | 130 | 200 |
Using the survey, one student is chosen at random. Find the probability that the student (a) has a job and drives, (b) has a job or drives (or both), (c) drives, given that they have a job.
Tree diagrams
A tree diagram shows a multistage experiment one stage at a time. Each set of branches from one point must add to 1, because one of them has to happen. Each complete path from the start to the end is one outcome.
Why multiply? Suppose Saturday has a $0.3$ chance of rain. Of all the wet Saturdays, Sunday is also wet $0.3$ of the time. So both days are wet in $0.3$ of $0.3$ of all weekends, and "of" means multiply: $0.3 \times 0.3 = 0.09$.
When the result of one stage doesn't change the chances at the next stage, the events are independent, and the branch probabilities are the same each time.
Set the chance of rain each day and the number of days. Pick an event and the matching paths light up. The Check readout shows the shortcut calculation.
With a $30\%$ chance of rain, how likely is at least one wet day over 3 days? Now push the chance up to $50\%$. Notice how quickly "at least one" heads towards certain as you add days.
The forecast gives a $30\%$ chance of rain on each of Saturday and Sunday, independently. Find the probability that (a) both days are wet, (b) exactly one day is wet, (c) at least one day is wet.
Check: the four paths are $0.09 + 0.21 + 0.21 + 0.49 = 1$. If your paths don't add to 1, something is wrong.
With and without replacement
If you draw an object, put it back and draw again, the second draw is exactly like the first: with replacement gives independent stages. If you keep the first object, there is one fewer object for the second draw and the probabilities change: without replacement gives dependent stages.
A pencil case holds 4 blue pens and 2 black pens. Two pens are taken out, one after the other, without replacement. Find the probability that (a) both pens are blue, (b) the pens are different colours.
With replacement, the second draw would still be $\frac{4}{6}$ for blue, so $P(BB) = \frac{4}{6} \times \frac{4}{6} = \frac{4}{9}$. That is a little higher, because putting the blue pen back keeps more blue pens in the case.
Your turn
A bag has 5 green and 3 white lollies. Two are eaten one after the other. Find the probability that both are green.
Show the answer
Eating is definitely without replacement. $P(GG) = \frac{5}{8} \times \frac{4}{7} = \frac{20}{56} = \frac{5}{14} \approx 0.36$.
Using the job and driving survey on this page, find the probability, as a decimal, that a randomly chosen student has a job.
The 'Has a job' row total is 110 out of 200. $\frac{110}{200} = 0.55$.
Using the same survey, find the probability, as a decimal, that a randomly chosen student drives to school.
The 'Drives' column total is 70. $\frac{70}{200} = 0.35$.
A health survey of 300 people found: vaccinated and got sick 12; vaccinated and stayed well 188; not vaccinated and got sick 30; not vaccinated and stayed well 70. Of the vaccinated people, what proportion got sick?
"Of the vaccinated people" makes the vaccinated group the total: $12 + 188 = 200$. So the proportion is $\frac{12}{200} = 0.06$, compared with $\frac{30}{100} = 0.3$ for the unvaccinated.
Events $A$ and $B$ are independent with $P(A) = 0.25$ and $P(B) = 0.4$. Find $P(A \text{ and } B)$.
For independent events, multiply: $0.25 \times 0.4 = 0.1$.
On a tree diagram, the branches coming out of one point have probabilities $0.35$ and $x$. What is $x$?
Branches from one point cover every possibility at that stage, so they add to 1. $x = 1 - 0.35 = 0.65$.
The chance of rain on any day this week is $0.2$, independently. Find the probability of at least one wet day in the next 3 days.
Use the complement. $P(\text{all 3 dry}) = 0.8 \times 0.8 \times 0.8 = 0.512$, so $P(\text{at least one wet}) = 1 - 0.512 = 0.488$.
A bag has 3 red and 5 blue marbles. Two are drawn without replacement. What is $P(\text{both red})$?
$P(RR) = \frac{3}{8} \times \frac{2}{7} = \frac{6}{56} = \frac{3}{28}$. The second fraction is out of 7, with only 2 reds left. $\frac{9}{64}$ would be the answer with replacement.
A basketballer makes $60\%$ of free throws, and each shot is independent. Find the probability, as a decimal, that she makes both of her next two free throws.
$P(\text{make, make}) = 0.6 \times 0.6 = 0.36$.
Exam style: $70\%$ of a Year 12 class study for a test. Of those who study, $90\%$ pass. Of those who don't study, $40\%$ pass. Use a tree diagram to find the probability, as a decimal, that a randomly chosen student passes.
Two paths end in a pass. Study then pass: $0.7 \times 0.9 = 0.63$. No study then pass: $0.3 \times 0.4 = 0.12$. Add them: $0.63 + 0.12 = 0.75$.
Exam style: For the same class, find the probability, as a decimal, that a randomly chosen student did not study and failed.
Did not study: $0.3$. Of those, $1 - 0.4 = 0.6$ fail. Multiply along the path: $0.3 \times 0.6 = 0.18$.
Coaching Centre