4.9 · SLPaper 1Paper 2

Normal distribution

In a minute

The models measurements such as heights and masses, which cluster symmetrically around a mean. Its probabilities are areas under a bell-shaped curve. You estimate them with the 68–95–99.7 rule on Paper 1, and find them exactly with your GDC on Paper 2, forwards or backwards.

You'll be able to

  • Describe the normal curve, its properties and where it occurs, and read the notation X∼N(μ,σ2)X \sim \mathrm{N}(\mu, \sigma^{2}).
  • Use symmetry and a sketch to find normal probabilities without a calculator.
  • Use the 68–95–99.7 rule to estimate probabilities and expected numbers.
  • Find normal probabilities with your GDC, including given-that probabilities.
  • Find the value with a given probability below or above it, or the bounds of a middle region, using inverse normal.
Before you start3 quick questions on what this builds on
Check yourself
A data set has variance 3636. Find its standard deviation.
Check yourself
P(A∩B)=0.12\mathrm{P}(A \cap B) = 0.12 and P(B)=0.4\mathrm{P}(B) = 0.4. Find P(A∣B)\mathrm{P}(A \mid B).
Check yourself
A discrete random variable XX takes the values 11, 22 and 33, with P(X=1)=0.2\mathrm{P}(X = 1) = 0.2 and P(X=2)=0.45\mathrm{P}(X = 2) = 0.45. Find P(X=3)\mathrm{P}(X = 3).

Revise first: 4.3 Central tendency and dispersion, 4.6 Probability diagrams, rules and independence and 4.7 Discrete random variables and expected value

The normal curve

Weigh a thousand apples from one orchard. Most masses sit close to the average, and extreme ones are rare on both sides. A histogram with narrow bars then has a smooth bell-shaped outline: the normal curve. Heights, lengths, times and measurement errors often behave this way.

A height is a : it can take any value in a range, not just separate values as in 4.7. So its probabilities are areas under a curve, not sums of single probabilities.

In N(μ,σ2)\mathrm{N}(\mu, \sigma^{2}), μ\mu is the . The second number is the , not the . So X∼N(60,9)X \sim \mathrm{N}(60, 9) has σ=3\sigma = 3. Questions often write N(60,32)\mathrm{N}(60, 3^{2}) to make σ\sigma easy to see.

Bell-shaped normal curve, symmetric about a dashed vertical line at the mean mu. The area under the curve between two values a and b to the right of the mean is shaded and labelled as the probability that X lies between a and b. The whole area under the curve is 1. μ a b shaded area = P(a < X < b) total area = 1 mean = median = mode
The probability that XX lies between aa and bb is the area under the curve between them.

1. The curve is bell-shaped and symmetric about the vertical line through μ\mu. So the mean, and mode are all equal to μ\mu.

2. The total area under the curve is 11, because XX is certain to take some value.

3. The curve gets closer and closer to the horizontal axis on both sides but never touches it. Values far from μ\mu are possible, just very unlikely.

4. μ\mu fixes where the centre is. σ\sigma fixes the spread: a larger σ\sigma gives a wider, flatter curve, since the area must still be 11.

Symmetry gives you some probabilities with no calculator. Half the area lies on each side of the mean. Two tails that start the same distance dd from μ\mu have equal areas.

Learn this
P(X<μ)=P(X>μ)=0.5,P(X<μ−d)=P(X>μ+d)\mathrm{P}(X < \mu) = \mathrm{P}(X > \mu) = 0.5, \qquad \mathrm{P}(X < \mu - d) = \mathrm{P}(X > \mu + d)
Worked exampleThe masses MM grams of eggs from a farm are modelled by M∼N(62,52)M \sim \mathrm{N}(62, 5^{2}). Write down P(M≥62)\mathrm{P}(M \ge 62).

Try it first, then reveal one step at a time.

  1. Read the model: the mean is μ=62\mu = 62 and the standard deviation is σ=5\sigma = 5.
Check yourself
The lengths LL cm of ribbons cut by a machine satisfy L∼N(40,32)L \sim \mathrm{N}(40, 3^{2}). Find P(L=40)\mathrm{P}(L = 40).
Worked exampleThe heights HH cm of a type of tomato plant are normally distributed, with P(H<38)=P(H>54)=0.12\mathrm{P}(H < 38) = \mathrm{P}(H > 54) = 0.12. Find P(38<H<46)\mathrm{P}(38 < H < 46).

Try it first, then reveal one step at a time.

  1. The two tails have equal areas, so 3838 and 5454 are the same distance from the mean. The mean is halfway between them: μ=38+542=46\mu = \frac{38 + 54}{2} = 46.
Check yourself
X∼N(75,σ2)X \sim \mathrm{N}(75, \sigma^{2}) and P(X>81)=0.3\mathrm{P}(X > 81) = 0.3. Find P(69<X<75)\mathrm{P}(69 < X < 75).

The 68–95–99.7 rule

Every normal curve has the same shape: μ\mu slides it along and σ\sigma stretches it. So the same fraction of values lies less than one σ\sigma from μ\mu, for every normal distribution, whatever the values of μ\mu and σ\sigma.

Learn this
P(μ−σ<X<μ+σ)≈0.68P(μ−2σ<X<μ+2σ)≈0.95P(μ−3σ<X<μ+3σ)≈0.997\begin{aligned} \mathrm{P}(\mu - \sigma < X < \mu + \sigma) &\approx 0.68 \\ \mathrm{P}(\mu - 2\sigma < X < \mu + 2\sigma) &\approx 0.95 \\ \mathrm{P}(\mu - 3\sigma < X < \mu + 3\sigma) &\approx 0.997 \end{aligned}

This is the . It is not in the formula booklet, so learn it.

Normal curve with vertical lines at one, two and three standard deviations either side of the mean. Brackets underneath show that about 68 percent of values lie within one standard deviation of the mean, about 95 percent within two, and about 99.7 percent within three. The two strips next to the mean each hold about 34 percent. μ − 3σ μ − 2σ μ − σ μ μ + σ μ + 2σ μ + 3σ 34% 34% 13.5% 13.5% about 68% about 95% about 99.7%
Split each band in half at μ\mu: the curve is symmetric.

Use symmetry to split the bands. Half of 68%68\% lies on each side of the , so 34%34\% lies between μ\mu and μ+σ\mu + \sigma. Half of 95%95\% is 47.5%47.5\%, so 47.5%−34%=13.5%47.5\% - 34\% = 13.5\% lies between μ+σ\mu + \sigma and μ+2σ\mu + 2\sigma.

The 5%5\% outside μ±2σ\mu \pm 2\sigma is shared by the two tails: 2.5%2.5\% in each. Likewise 0.15%0.15\% lies above μ+3σ\mu + 3\sigma.

Worked exampleThe masses MM grams of pears from a farm are modelled by M∼N(150,122)M \sim \mathrm{N}(150, 12^{2}). Use the 68–95–99.7 rule to estimate P(126<M<174)\mathrm{P}(126 < M < 174).

Try it first, then reveal one step at a time.

  1. Write each bound as a number of from the mean: 126=150−2×12=μ−2σ126 = 150 - 2 \times 12 = \mu - 2\sigma and 174=150+2×12=μ+2σ174 = 150 + 2 \times 12 = \mu + 2\sigma.
Worked exampleFor the pears in the last example, a crate holds 600600 pears. Estimate how many of them have a mass between 138138 g and 174174 g.

Try it first, then reveal one step at a time.

  1. Convert the bounds: 138=150−12=μ−σ138 = 150 - 12 = \mu - \sigma and 174=μ+2σ174 = \mu + 2\sigma. The interval is not symmetric, so split it at the mean.

Try it below. The shading starts at 150±12150 \pm 12 and shows 0.68270.6827. Drag the bounds to 126126 and 174174 to see 0.95450.9545. The rule rounds these exact areas.

Normal probabilitiesDrag the sliders
114126138150162174186X ~ N(150, 12²)
150
12

138
162

Probability = 0.6827 (P(138 < X < 162))

z-scores: a → −1.00 b → 1.00

The area under the bell curve between two values is a probability. The whole area is 1, and about 68% lies within one standard deviation of the mean.

The z-scores it lists count standard deviations from the mean. They belong to 4.12.

Check yourself
The lifetimes of a type of battery are normally distributed. About 95%95\% of lifetimes lie between 410410 hours and 490490 hours, an interval centred on the mean. Estimate the standard deviation σ\sigma, in hours.
Check yourself
X∼N(20,22)X \sim \mathrm{N}(20, 2^{2}). Use the 68–95–99.7 rule to estimate P(X<14)\mathrm{P}(X < 14).

Normal probabilities on your GDC

The rule only works at whole numbers of . For other bounds, your GDC's normal cdf finds the area between two bounds.

1. Sketch the curve, mark μ\mu and the bounds, and shade the area you want.

2. Write the probability statement and the parameters, such as P(40<T<55)\mathrm{P}(40 < T < 55) with μ=48\mu = 48, σ=7\sigma = 7.

3. Enter the lower bound, upper bound, μ\mu and σ\sigma. Enter σ\sigma, not σ2\sigma^{2}.

4. For less than bb, use a huge negative lower bound, −1E99-1\text{E}99. For more than aa, use 1E991\text{E}99 as the upper bound.

Worked exampleThe time TT minutes a bus takes to complete its route is modelled by T∼N(48,72)T \sim \mathrm{N}(48, 7^{2}). Use your calculator to find the probability that a journey takes between 4040 and 5555 minutes. Give your answer to 3 s.f.

Try it first, then reveal one step at a time.

  1. Sketch: 4848 in the centre, 4040 to its left and 5555 to its right. Shade between them.
Check yourself
The lengths FF cm of fish in a lake are modelled by F∼N(31.5,2.82)F \sim \mathrm{N}(31.5, 2.8^{2}). Find the probability that a fish is longer than 3535 cm. Give your answer to 3 s.f.

For a given-that question, use P(A∣B)=P(A∩B)P(B)\mathrm{P}(A \mid B) = \frac{\mathrm{P}(A \cap B)}{\mathrm{P}(B)} from 4.6. Here A∩BA \cap B is where the two intervals overlap: sketch them.

Worked exampleThe masses WW kg of parcels at a depot follow W∼N(2.3,0.42)W \sim \mathrm{N}(2.3, 0.4^{2}). A parcel is chosen at random. Given that it weighs less than 2.52.5 kg, find the probability that it weighs less than 22 kg.

Try it first, then reveal one step at a time.

  1. Write the condition down: P(W<2∣W<2.5)=P(W<2 and W<2.5)P(W<2.5)\mathrm{P}(W < 2 \mid W < 2.5) = \frac{\mathrm{P}(W < 2 \text{ and } W < 2.5)}{\mathrm{P}(W < 2.5)}.
Check yourself
The volume VV litres of paint in a tin follows V∼N(3.2,0.252)V \sim \mathrm{N}(3.2, 0.25^{2}). Given that a tin holds more than 33 litres, find the probability that it holds more than 3.53.5 litres.

Inverse normal: from a probability to a value

Sometimes you know the area and need the boundary: the mark the lowest 15%15\% fall below, or the mass only 2%2\% of boxes exceed. This is an calculation.

Inverse normal works with the area to the left of kk. So turn every condition into a left-hand area first.

1. P(X<k)=p\mathrm{P}(X < k) = p: enter the area pp.

2. P(X>k)=p\mathrm{P}(X > k) = p: the area to the left of kk is 1−p1 - p, so enter 1−p1 - p.

3. A middle region of probability pp, centred on μ\mu: the tails share 1−p1 - p, so each holds 1−p2\frac{1 - p}{2}. The lower bound has area 1−p2\frac{1 - p}{2} to its left, and the upper bound has 1+p2\frac{1 + p}{2}.

Learn this
P(X>k)=p  ⟺  P(X<k)=1−p\mathrm{P}(X > k) = p \iff \mathrm{P}(X < k) = 1 - p

In this lesson μ\mu and σ\sigma are always given. When one of them is unknown, you work with the standardised value zz instead, in 4.12.

Worked exampleMarks MM in a test are modelled by M∼N(58,112)M \sim \mathrm{N}(58, 11^{2}). The lowest 15%15\% of candidates fail. Find the pass mark mm, to 3 s.f.

Try it first, then reveal one step at a time.

  1. Sketch: a small shaded tail on the left, so mm must be below 5858.
Worked exampleThe masses BB grams of cereal in boxes are modelled by B∼N(507,42)B \sim \mathrm{N}(507, 4^{2}). The heaviest 2%2\% of boxes are sent back for checking. Find the mass mm above which a box is sent back, to 3 s.f.

Try it first, then reveal one step at a time.

  1. Write the condition: P(B>m)=0.02\mathrm{P}(B > m) = 0.02.
Worked exampleThe lengths LL mm of leaves on a tree are modelled by L∼N(84,92)L \sim \mathrm{N}(84, 9^{2}). The middle 70%70\% of lengths lie between pp and qq, symmetric about the mean. Find pp and qq.

Try it first, then reveal one step at a time.

  1. Outside the middle 70%70\% lies 30%30\%, shared equally: 15%15\% in each tail.
Check yourself
X∼N(40,52)X \sim \mathrm{N}(40, 5^{2}) and P(X>k)=0.3\mathrm{P}(X > k) = 0.3. Find kk.
Check yourself
X∼N(12,1.52)X \sim \mathrm{N}(12, 1.5^{2}). Find cc such that P(12−c<X<12+c)=0.8\mathrm{P}(12 - c < X < 12 + c) = 0.8. Give your answer to 3 s.f.

Common mistakes

  • Typing the variance into the GDC's standard deviation field. In N(μ,σ2)\mathrm{N}(\mu, \sigma^{2}) the second number is σ2\sigma^{2}, so take its square root first.
  • Using the area pp in inverse normal when P(X>k)=p\mathrm{P}(X > k) = p. Inverse normal wants the area to the left, so enter 1−p1 - p. A sketch shows kk must be above the mean.
  • Putting all the leftover area in one tail for a middle region. For the middle 80%80\%, each tail holds 10%10\%, so the bounds have areas 0.10.1 and 0.90.9 to their left.
  • In a given-that question, stopping at P(A∩B)\mathrm{P}(A \cap B). Divide by the probability of the condition, P(B)\mathrm{P}(B).
  • Using normal pdf to find a probability. For a continuous variable P(X=a)=0\mathrm{P}(X = a) = 0; normal pdf gives the height of the curve, not an area.
  • Rounding a probability to 3 s.f. and then using it in a later step. Store the full value, and round only the final answer.

Exam tips

  • Sketch: draw a symmetric bell, mark the mean, label the bounds and shade the region. It shows at once whether the answer should be above or below 0.5.
  • On Paper 2, write down the probability statement and the parameters, such as P(40<T<55)\mathrm{P}(40 < T < 55) with μ=48\mu = 48, σ=7\sigma = 7, before the answer. That line earns the method mark.
  • Paper 1 has no calculator, so normal questions there use symmetry and the 68–95–99.7 rule. Write down needs no working, but check which side of the mean each bound is on.
  • Give probabilities to 3 s.f. unless told otherwise. If a later part uses an earlier probability, use the stored value; a wrong earlier answer used correctly can still earn follow-through marks.

Key takeaways

Learn thisP(X<μ)=P(X>μ)=0.5,P(X<μ−d)=P(X>μ+d)\mathrm{P}(X < \mu) = \mathrm{P}(X > \mu) = 0.5, \qquad \mathrm{P}(X < \mu - d) = \mathrm{P}(X > \mu + d)
Learn thisP(μ−σ<X<μ+σ)≈0.68P(μ−2σ<X<μ+2σ)≈0.95P(μ−3σ<X<μ+3σ)≈0.997\begin{aligned} \mathrm{P}(\mu - \sigma < X < \mu + \sigma) &\approx 0.68 \\ \mathrm{P}(\mu - 2\sigma < X < \mu + 2\sigma) &\approx 0.95 \\ \mathrm{P}(\mu - 3\sigma < X < \mu + 3\sigma) &\approx 0.997 \end{aligned}
Learn thisP(X>k)=p  ⟺  P(X<k)=1−p\mathrm{P}(X > k) = p \iff \mathrm{P}(X < k) = 1 - p
Learn thisX∼N(μ,σ2)X \sim \mathrm{N}(\mu, \sigma^{2}): a bell-shaped curve, symmetric about μ\mu, with total area 11. Mean, median and mode all equal μ\mu. The second number is the variance.
Learn thisFor a continuous variable P(X=a)=0\mathrm{P}(X = a) = 0, so << and ≤\le give the same probability.
Learn thisNormal cdf takes lower bound, upper bound, μ\mu and σ\sigma. Use −1E99-1\text{E}99 for less than and 1E991\text{E}99 for more than.
Learn thisFor a middle region of probability pp, each tail holds 1−p2\frac{1 - p}{2}.
Learn thisGiven that: P(A∣B)=P(A∩B)P(B)\mathrm{P}(A \mid B) = \frac{\mathrm{P}(A \cap B)}{\mathrm{P}(B)}, where A∩BA \cap B is the overlap of the two intervals.

Practise this

15 past-paper questions on 4.9 in the AA HL bank, each with a mark scheme and a worked explanation.

Practise 4.9 →