CONTINUOUS

PROBABILIT Y
DISTRIBUTIONS
Y E A R 1 2 M AT H E M AT I C S A D V A N C E D
CONTINUOUS PROBABILITY
DISTRIBUTIONS
• When a variable can take any value in a particular interval, for example, when you measure rather than count it, you
have a continuous random variable.
• Quantities which can be modelled using continuous random variables include height, weight, time and mass.
EXAMPLE 1
This table gives the results of a survey of different times that runners take to complete a race.

a) Add a column of relative frequencies.

EXAMPLE 1 CONTINUED
b) Sketch a frequency histogram for the relative c) Estimate each probability
frequencies. i. 𝑃 𝑋 < 12
Solution Solution
6 8 11
(
P X < 12 = ) + +
32 32 32
25
=
32

ii. 𝑃(𝑋 ≥ 16)

Solution
2 1
(
P X ≥ 16 = ) +
32 32
3
=
32

iii. 𝑃(4 ≤ 𝑋 < 8)

Solution
1
(
P 4≤ X <8 = ) 4
CONTINUOUS PROBABILITY
DISTRIBUTIONS
• With continuous data we can’t really draw a histogram as we did in Example 1 or there would be an ‘infinite’ number
of columns with ‘zero’ widths. Instead, the probability distribution is a continuous curve.
PROBABILITY DENSITY FUNCTION

• A continuous probability distribution is represented by a function 𝑃 𝑋 = 𝑥 or 𝑝(𝑥) called a probability

density function (PDF) where 𝑋 is the random variable.
• As with discrete probability distributions, the sum of all probabilities must be 1.
• With a continuous probability distribution, we cannot calculate the probability for a single outcome, so 𝑃 𝑋 = 𝑥 = 0.
• Instead, we can only calculate the probability for a rage of values such as 𝑃 4 ≤ 𝑋 < 8 .
AREA UNDER A PROBABILITY DENSITY
FUNCTION
• The area under a probability density function is 1.

"
∫!" 𝑓 𝑥 𝑑𝑥 = 1

where 𝑓 𝑥 ≥ 0 (since 0 ≤ 𝑝 𝑥 ≤ 1)
EXAMPLE 2
⎧ 3x 2
⎪ for 1 ≤ x ≤ 3
A function is given by f (x) = ⎨ 26 .
⎪ 0 for all other x
⎩

Show that it is a continuous probability distribution.

Solution
For a continuous probability distribution, the area under the curve must be 1.
Drawing the graph, notice that the area will be 0 for all 𝑥 values outside 1 ≤ 𝑥 ≤ 3.
3 3x 2
Area = ∫
1 26
dx

1 3 2
=
26 1 ∫
3x dx

1 3 3
= ⎡x ⎤
26 ⎣ ⎦1
1 3 3
= ⎡3 − 1 ⎤⎦
26 ⎣
1
= ⎡ 26 ⎤
26 ⎣ ⎦
=1
So 𝑓(𝑥) is a continuous probability distribution.
EXAMPLE 3
A function is given by 𝑓 𝑥 = 𝑎𝑥 # defined for the domain [0, 5]. Find the value of 𝑎 for which this a probability density
function.
Solution
Drawing the graph gives the parabola in the domain 0, 5 .

A PDF has area 1.

5 125a
So ∫
0
ax 2 dx = 1 So
3
=1
5 5 3
∫0
ax 2 dx = a ∫
0
x 2 dx a=
125
5
⎡ x3 ⎤
= a⎢ ⎥
⎣ 3 ⎦0
⎡ 53 03 ⎤
= a⎢ − ⎥
⎣3 3⎦
⎡ 125 ⎤
= a⎢ ⎥
⎣ 3 ⎦
125a
=
3
CALCULATING PROBABILITIES

• Since 𝑃 𝑋 = 𝑥 = 0 for continuous probability distributions, we can only find the probability of a range of values
𝑃 𝑎≤𝑋≤𝑏 .
• Also, since 𝑃 𝑋 = 𝑎 = 0 and 𝑃 𝑋 = 𝑏 = 0, it makes no difference whether we use ≤ or <, ≥ or >.
• 𝑃 𝑎<𝑋<𝑏 =𝑃 𝑎≤𝑋≤𝑏
EXAMPLE 1
For the probability density function, find:

a) 𝑃 𝑋≤2 b) 𝑃 1<𝑋<4
Solution Solution
𝑃 𝑋 ≤ 2 is the shaded area between 𝑥 = 0 and 𝑃 1 < 𝑋 < 4 is the shaded area between 𝑥 = 1
𝑥 = 2. and 𝑥 = 4.
Notice that 𝑃 1 < 𝑋 < 4 = 𝑃 1 ≤ 𝑋 ≤ 4 since
1 𝑃 𝑋 = 1 = 𝑃 𝑋 = 4 = 0.
P( X ≤ 2) = 2 ×
5
2 1
= P(1 < X < 4) = 3×
5 5
3
=
5
PROBABILITIES IN PROBABILTY
DENSITY FUNCTIONS
%
• 𝑃 𝑋 ≤ 𝑥 = ∫$ 𝑓 𝑥 𝑑𝑥 where 𝑦 = 𝑓 𝑥 is a PDF
defined in the domain [𝑎, 𝑏].

&
• 𝑃 𝑎 ≤ 𝑋 ≤ 𝑏 = ∫$ 𝑓 𝑥 𝑑𝑥 where 𝑦 = 𝑓(𝑥) is a PDF
and 𝑎 and 𝑏 are in the defined domain.
EXAMPLE 2
'% !
A function is given by 𝑓 𝑥 = (() defined in the domain [2, 5]. Find:

a) 𝑃(𝑋 ≤ 4) b) 𝑃(3 ≤ 𝑋 ≤ 4)
Solution Solution
2
3x4
3x 2
4
P( X ≤ 4) = ∫
2 117
dx (since domain is [2, 5]) P(3 ≤ X ≤ 4) = ∫
3 117
dx

1 4 2 1 4 2
=
117 2 ∫
3x dx =
117 3 ∫
3x dx

1 4 1 4
= ⎡ x3 ⎤ = ⎡ x3 ⎤
117 ⎣ ⎦ 2 117 ⎣ ⎦3
=
1
117
(
43 − 23 ) =
1
117
(43 − 33 )

56 37
= =
117 117
UNIFORM CONTINUOUS PROBABILITY
DISTRIBUTIONS
• For a uniform probability distribution, every outcome has the same probability.

⎧ 1
⎪ for all a ≤ x ≤ b
f (x) = ⎨ b − a
⎪⎩ 0 for all other values of x

• Equal intervals along the 𝑥 – axis will have the same probability.
EXAMPLE 3
A continuous probability function y = 𝑓(𝑥) is uniform in the domain [5, 15].
a) Sketch the probability density function.
Solution
A uniform distribution has all equal probabilities so will have the same height. This gives a rectangle.

PDF has area 1:

A = bh
= (15 − 5)h
= 10h
So 10h = 1
1
h=
10
EXAMPLE 3 CONTINUED
b) Find:

i. 𝑃 𝑋≥8 ii. 𝑃 7 ≤ 𝑋 ≤ 10 iii. 𝑃 8 < 𝑋 < 11

Solution Solution Solution

1 1 1
P( X ≥ 8) = (15 − 8) × P(7 ≤ X ≤ 10) = (10 − 7) × P(8 < X < 11) = (11− 8) ×
10 10 10
7 3 3
= = =
10 10 10
CUMULATIVE DISTRIBUTION
FUNCTION
• The cumulative distribution function is a general formula for finding 𝑃 𝑋 ≤ 𝑥 directly.
• The cumulative distribution function (CDF) is given by
%
𝐹 𝑥 = @ 𝑓(𝑥) 𝑑𝑥
$
where 𝑓(𝑥) is a PDF defined in the domain [𝑎, 𝑏].
• We can use the cumulative distribution function to find probabilities such as 𝑃 𝑋 ≥ 𝑎 or 𝑃(𝑎 ≤ 𝑋 ≤ 𝑏).
EXAMPLE 1
*% "
A continuous probability function is given by 𝑓 𝑥 = #++ defined in the domain [1, 4].
a) Find the cumulative distribution function. b) Use the CDF to find:
Solution i. 𝑃 𝑋≤3
x
Solution
F(x) = ∫
a
f (x) dx
34 − 1
For 𝑃 𝑋 ≤ 3 : F(3) =
x 4x 3 255
= ∫
1 255
dx
81− 1
=
1 x 3 255
=
255 1 ∫
4x dx
80
=
1 x 255
= ⎡⎣ x 4 ⎤⎦
255 1 16
=
=
1
255
(
x4 − 1 ) (,
51

So 𝑃 𝑋 ≤ 3 = +(
x4 − 1
=
255 ii. 𝑃 𝑋 < 1.6
Solution
For 𝑃 𝑋 < 1.6 : F(1.6) = 1.6 − 1
4

255
= 0.0218
So 𝑃 𝑋 < 1.6 = 0.0218
EXAMPLE 2
'% !
A continuous probability function is given by 𝑓 𝑥 = ''+ defined in the domain [2, 7].
a) Find the cumulative distribution function.
Solution
x
F(x) = ∫
a
f (x) dx
x 3x 2
= ∫
2 335
dx

1 x 2
=
335 2 ∫
3x dx

1 x
= ⎡⎣ x 3 ⎤⎦
335 2

=
1
335
(
x 3 − 23 )
=
1
335
x3 − 8( )
x3 − 8
=
335
EXAMPLE 2 CONTINUED
b) Use the CDF to find:

i. 𝑃 𝑥≥4
Solution

To find 𝑃 𝑥 ≥ 4 , first find 𝑃 𝑋 ≤ 4 : Area under the curve is 1:

The shaded part of the PDF is 𝑃 𝑋 ≤ 4 . P( X ≥ 4) = 1− P( X ≤ 4)
43 − 8 56
P( X ≤ 4) = = 1−
335 335
56 279
= =
335 335
EXAMPLE 2 CONTINUED

ii. 𝑃(3.5 ≤ 𝑋 ≤ 6.2)

Solution
𝑃(𝑋 ≤ 3.5) is the area to the left of 𝑥 = 3.5
𝑃 𝑋 ≤ 6.2 is the area to the left of 𝑥 = 6.2
So P(3.5 ≤ X ≤ 6.2) = P( X ≤ 6.2) − P( X ≤ 3.5)
6.23 − 8 3.53 − 8
= −
335 335
= 0.688 − 0.104
= 0.583
MODE OF A CONTINUOUS
PROBABILITY DISTRIBUTION
• We sometimes want to know what the highest probability is. This is the mode.
• The mode is the maximum point of the probability density function.
EXAMPLE 3
Find the mode of the continuous probability distribution shown below.

Solution

The highest point of the PDF is at 𝑥 = 7.7

So the mode is 7.7.
EXAMPLE 4
'% ,!%
A continuous probability distribution is defined on the interval 1 ≤ 𝑥 ≤ 5 and has equation 𝑓 𝑥 = -#
. Find the
mode.
Solution
3x(6 − x)
f (x) =
92
3
= (6x − x 2 )
92
The function is a parabola with 𝑎 < 0 so will have a maximum turning point.
We use calculus to see if this point lies within the defined domain [1, 5].
&
(We could also use 𝑥 = − #$ for the axis of symmetry of a parabola).
3
f '(x) = ( 6 − 2x ) x = 3 lies in the domain [1, 5].
92 3
For stationary points: f "(x) = (−2)
92
f '(x) = 0 3
3 = −
92
( 6 − 2x ) = 0 46
<0
6 − 2x = 0
6 = 2x Concave down so a maximum turning point.
So the mode is 3.
3= x
QUANTILES

Median
• For a continuous probability distribution, the median is the value of 𝑥 that splits the distribution into halves.
(
• Because the PDF has an area of 1, the area on each side of the median is #.
• The median lies at the point 𝑥 where

%
@ 𝑓 𝑥 𝑑𝑥 = 0.5
$

given 𝑦 = 𝑓(𝑥) is a PDF defined in the domain [𝑎, 𝑏].

EXAMPLE 1
%!
Find the median of the continuous probability distribution defined as 𝑓 𝑥 = #( in the domain [1, 4].
Solution
%!
For 𝑓 𝑥 = (# defined in the domain [1, 4], first find the cumulative distribution function (CDF):

x x2 1 x
For the median:
∫
1 21
dx =
21 ∫1
x 2 dx
x

1 ⎡ x3 ⎤
x
∫a
f (x) dx = 0.5
= ⎢ ⎥ x3 − 1
21 ⎣ 3 ⎦1 = 0.5
63
1 ⎛ x 3 13 ⎞
= ⎜ − ⎟ x 3 − 1 = 31.5
21 ⎝ 3 3 ⎠
x 3 = 32.5
1 ⎛ x − 1⎞ 3
= x = 3 32.5
21 ⎜⎝ 3 ⎟⎠
≈ 3.2
x −1
3
=
63
QUARTILES, DECILES AND
PERCENTILES
• Quartiles, deciles and percentiles are values that separate a proportion of a set of data.
• For example, 𝑄( > bottom 25% of scores, 𝑄' > bottom 75% of scores, 2nd decile > bottom 20% of scores and
67th percentile > bottom 67% of scores.
EXAMPLE 2
%#
A continuous probability distribution is defined as 𝑓 𝑥 = ((,.+ in the domain [4, 9].
Find, correct to 2 decimal places:
a) The 1st quartile
Solution
First find the CDF: 1st quartile: 25%
xx4 1 x

∫ ∫
x

4 11605
dx =
11605 4
x 4 dx
∫a
f (x) dx = 0.25
x
1 ⎡ x5 ⎤ x 5 − 1024
= ⎢ ⎥ = 0.25
11605 ⎣ 5 ⎦ 4 58025
x 5 − 1024 = 24506.25
1 ⎛ x 5 45 ⎞
= − x 5 = 15530.25
11605 ⎜⎝ 5 5 ⎟⎠
x = 5 15530.25
1 ⎛ x 5 − 1024 ⎞
= ≈ 6.89
11605 ⎜⎝ 5 ⎟⎠

x 5 − 1024 So the 1st quartile is 6.89.

= This means 𝑃 𝑋 ≤ 6.89 = 0.25.
58025
EXAMPLE 2 CONTINUED

b) The 38th percentile c) The 7th decile

Solution Solution
38th percentile: 38% 7th decile: 70%
x x

∫a
f (x) dx = 0.38 ∫a
f (x) dx = 0.7

x 5 − 1024 x 5 − 1024
= 0.38 = 0.7
58025 58025
x 5 − 1024 = 22049.5 x 5 − 1024 = 40617.5
x 5 = 23073.5 x 5 = 41641.5
x = 5 23073.5 x = 5 41641.5
≈ 7.46 ≈ 8.39

So the 38th percentile is 7.46. So the 7th decile is 8.39.

This means 𝑃 𝑋 ≤ 7.46 = 0.38 This means 𝑃 𝑋 ≤ 8.39 = 0.7.
NORMAL DISTRIBUTION
• The normal distribution is a special continuous probability distribution.
• Its probability density function is often called a bell curve because of its shape.
• The normal distribution is a symmetrical bell-shaped function.
• The mean, mode and median are equal, at the centre of the probability density function.

• We use the population mean 𝜇 and standard deviation 𝜎 for the normal distribution.
GRAPHING A NORMAL DISTRIBUTION
• In a normal distribution, most of the data lies within 3 standard deviations of the mean, with the mean in the centre.
EXAMPLE 1
a) A set of data is normally distributed with mean 8.3 and standard deviation 1.2.
Sketch the probability distribution function.
Solution
µ + σ = 8.3+ 1.2
= 9.5
µ + 2σ = 8.3+ 2(1.2)
= 10.7
µ + 3σ = 8.3+ 3(1.2)
= 11.9
µ − σ = 8.3− 1.2
= 7.1
µ − 2σ = 8.3− 2(1.2)
= 5.9
µ − 3σ = 8.3− 3(1.2)
= 4.7
EXAMPLE 1 CONTINUED
b) A normal distribution has the probability density function below. Find its mean and standard deviation.

Solution
µ = 15.6 (centre of the distribution)
µ + σ = 19
15.6 + σ = 19
σ = 19 − 15.6
= 3.4 (check that all the differences between all the values on the graph are 3.4)
GRAPHING A NORMAL DISTRIBUTION
• The normal distribution can have different shapes depending on the size of the standard deviation.
• In the diagram, the green curve shows the normal distribution with the highest standard deviation.
STANDARD NORMAL DISTRIBUTION
• The standard normal distribution is a normal curve that has been transformed so that that the mean is 0 and the
standard deviation is 1.
• The values of a standard normal distribution are called 𝑧 rather than 𝑥, also known as 𝒛 – scores or standardised
scores.

µ=0
σ =1
Area under the curve is 1.
PROBABILITY TABLES FOR THE
STANDARD NORMAL DISTRIBUTION
• Probability tables for the standard normal distribution can be found on page 633-634 of the Maths in Focus textbook.
• The values represent the area to the left of (or less than) the 𝑧 – score.
• Row labels show the 𝑧 – score to one decimal place.
• Column labels show the second decimal place.
EXAMPLE 2
For a standard normal distribution, use the table to find:

a) 𝑃 𝑧 ≤ 0.6 b) 𝑃 𝑧 ≤ −1.83
Solution Solution
The table gives the area under the PDF for the
standard normal distribution.

For −1.83, find −1.8 in the left column of the

table and the entry under 0.03 in the row.
Find 0.6 in the left column of the table. 𝑃 𝑧 ≤ −1.83 = 0.0336
𝑃 𝑧 ≤ 0.6 = 0.7257
EXAMPLE 2 CONTINUED

c) 𝑃 𝑧 < 2.34 d) 𝑃(𝑧 ≥ −2.7)

Solution Solution

Find 2.34, find 2.3 in the left column of the table Find −2.7 in the left column of the table.
and the entry under 0.04 in this row. P(z ≤ −2.7) = 0.0035
𝑃 𝑧 < 2.34 = 0.9904 P(z ≥ −2.7) = 1− 0.0035
= 0.9965
EXAMPLE 2 CONTINUED

e) 𝑃(−0.3 ≤ 𝑧 ≤ 1.4)
Solution

P(−0.3 ≤ z ≤ 1.4) = P(z ≤ 1.4) − P(z ≤ −0.3)

= 0.9192 − 0.3821
= 0.5371
EXAMPLE 3
For a standard normal distribution, use the tables to find:

a) The median b) The lower quartile

Solution Solution
The median separates the bottom half of the The lower quartile separated the bottom 0.25 of
scores. the data.
We want to find the score with the cumulative In the table there are 2 values close to 0.25:
probability of 0.5 in the table.. 𝑃 𝑧 ≤ −0.67 = 0.2514 and
Median = 0 (this is the same value as the mean) 𝑃 𝑧 ≤ −0.68 = 0.2483
0.2514 is closer to 0.25 so the lower quartile is
approximately −0.67.
(Note: 0.2514 − 0.25 = 0.0014,
and 0.25 − 0.2483 = 0.0017.)
EXAMPLE 3 CONTINUED
c) The upper quartile d) the 84th percentile
Solution Solution
The upper quartile separates the bottom 0.75 of The 84th percentile separates the bottom 0.84 of
the data. the data.
In the table there are 2 values close to 0.75: In the table there are 2 values close to 0.84:
𝑃 𝑧 ≤ 0.67 = 0.7486 and 𝑃 𝑧 ≤ 0.99 = 0.8389 and 𝑃 𝑧 ≤ 1 = 0.8413
𝑃 𝑧 ≤ 0.68 = 0.7517 0.8389 is closer to 0.84 so the 84th percentile is
0.7486 is closer to 0.75 so the upper quartile is approximately 0.99.
approximately 0.67.
Notice that the values for the lower and upper
quartiles are ±0.67 because the normal
distribution is symmetrical.
EXAMPLE 3 CONTINUED

e) The 6th decile

Solution
The 6th decile separates the bottom 0.6 of the data.
In the table there are 2 values close to 0.6:
𝑃 𝑧 ≤ 0.25 = 0.5987 and 𝑃 𝑧 ≤ 0.26 = 0.6026
0.5897 is closer to 0.6 so the 6th decile is approximately 0.25
EXAMPLE 4
Shade the area of the normal distribution where the values are:

a) Above the top 20% of data.

Solution
𝑃 𝑧 > 𝑎 = 20%, so 𝑃 𝑧 ≤ 𝑎 = 80% (0.8 or 8th decile)
From the table:
𝑃 𝑧 ≤ 0.84 = 0.7995 and 𝑃 𝑧 ≤ 0.85 = 0.8023
0.7995 is closer to 0.8 so the 8th decile is approximately 0.84.
This means all values to the left of 0.84 lie below 80%.
So the top 20% lies to the right of 0.84.
EXAMPLE 4 CONTINUED

b) Above the top 10% of data.

Solution
𝑃 𝑧 > 𝑎 = 10%, so 𝑃 𝑧 ≤ 𝑎 = 90% (0.9 or 9th decile)
From the table:
𝑃 𝑧 ≤ 1.28 = 0.8997 and 𝑃 𝑧 ≤ 1.29 = 0.9015
0.8997 is closer to 0.9 so the 9th decile is approximately 1.28.
This means all values to the left of 1.28 lie below 90%.
So the top 10% lies to the right of 1.28.
EMPIRICAL RULE
EXAMPLE 1
A data set is normally distributed with mean 20 and standard deviation 3.
What percentage of scores lie between:

a) 17 and 23? b) 14 and 26?

Solution Solution
We can draw the PDF for the normal curve. Scores between 14 and 26 are within 2 standard
deviations of the mean.
So about 95% of scores lie between 14 and 26.

c) 11 and 29?
Solution
Scores between 11 and 29 are within 3 standard
deviations of the mean.
Scores between 17 and 23 are within 1 standard
deviation of the mean. So about 99.7% of scores lie between 11 and 29.

So about 68% of scores lie between 17 and 23.

EXAMPLE 2
A normal distribution has 𝜇 = 65.2 and 𝜎 = 1.3. What percentage of scores lie between:

a) 61.3 and 65.2? b) 65.2 and 66.5?

Solution Solution
We can draw the PDF for the normal curve.

Scores between 63.9 and 66.5 are within 1

Scores between 61.3 and 69.1 are within 3 standard deviation of the mean (68% of scores).
standard deviations of the mean (99.7% of data). Scores between 65.2 and 66.5 are half this area.
Scores between 61.3 and 65.2 are half this area. 68% ÷ 2 = 34%
99.7% ÷ 2 = 49.85% So 34% of scores lie between 65.2 and 66.5.
So 49.85% of scores lie between 61.3 and 65.2.
EXAMPLE 2 CONTINUED
c) 62.6 and 69.1?
Solution

(
Scores between 62.6 and 65.2 are half the area within 2 standard deviations of the mean #
×95% .
(
Scores between 65.2 and 69.1 are half the area within 3 standard deviations of the mean #
×99.7% .
1 1
Total area = × 95% + × 99.7%
2 2
= 47.5% + 49.85%
= 97.35%
So 97.35% of scores lie between 62.6 and 69.1.
𝑧 - SCORES

• We can transform any normal distribution into a standard normal distribution by using 𝑧 –
scores.
• To convert a raw score, 𝑥, into a 𝑧 – score, use the formula:

𝑥−𝜇
𝑧=
𝜎

where 𝜇 is the mean and 𝜎 is the standard deviation of the distribution.

EXAMPLE 1
A data set is normally distributed with mean 15 and standard deviation 2.
a) Draw the probability density function for this
distribution
Solution
ii. 𝑥 = 15
Solution
x−µ
z=
σ
15 − 15
=
2
=0
b) Use 𝑧 – scores to convert each 𝑥 value to
standardised scores: iii. 𝑥=9
i. 𝑥 = 19 Solution
Solution x−µ
z=
x−µ σ
z=
σ 9 − 15
=
19 − 15 2
=
2 = −3
=2
EXAMPLE 2
A data set is normally distributed with mean 23.8 and standard deviation 1.2.
a) Find the 𝑧 – score for each raw score and describe where it is on the standard normal distribution.
i. 25.7 iii. 28.3
Solution Solution
x−µ x−µ
z= z=
σ σ
25.7 − 23.8 28.3− 23.8
= =
1.2 1.2
≈ 1.58 ≈ 3.75

A score of 25.7 lies 1.58 standard deviations A score of 28.3 lies 3.75 standard deviations
above (to the right of) the mean. above (to the right of) the mean
ii. 20.6
Solution
x−µ
z=
σ
20.6 − 23.8
=
1.2
≈ −2.67 A score of 20.6 lies 2.67 standard deviations below (to the left of) the mean.
EXAMPLE 2 CONTINUED
b) Which of the scores from part a) are very unlikely? c) Find the value of a raw score whose 𝑧 – score is
Solution − 1.82, correct to 2 decimal places.

We can draw each score on the standard normal Solution

x−µ
distribution. z=
σ
x − 23.8
−1.82 =
1.2
−2.184 = x − 23.8
21.616 = x

So the raw score is 21.62.

99.7% of scores lie within 3 standard deviations of
the mean.
So a 𝑧 – score of 3.75 is very unlikely.
This means that a score of 28.3 is very unlikely.
APPLICATIONS OF THE NORMAL
DISTRIBUTION
• The normal distribution is often used in quality control and predicting outcomes.
• Using 𝑧 – scores also allows us to compare data sets.
EXAMPLE 1
A company produces 1 kg packets of sugar. A quality control check found that the weight of the packets was normally
distributed with a mean weight of 0.995 kg and standard deviation 0.03 kg. The company policy is to reject any packet
with a weight outside 2 standard deviations from the mean.

a) What is the smallest weight allowed by the c) What percentage of packets will be rejected?
company? Solution
Solution About 95% of weights lie within 2 standard
µ − 2σ = 0.995 − 2 × 0.03 deviations of the mean. So 5% will lie outside this
= 0.935 kg area.
The company rejects 5% of the packets of sugar.

b) What is the largest weight allowed?

Solution d) What percentage of large packets will be rejected?
µ + 2σ = 0.995 + 2 × 0.03 Solution
= 1.055 kg Since the normal distribution is symmetrical, the
5% is made up of 2.5% of larger and 2.5% of
smaller packets.
So the company rejects 2.5% of larger packets.
EXAMPLE 2
The mean shelf life of a spice is 13.4 weeks and the standard deviation is 1.8 weeks.

a) Would a shelf life of 20 weeks be unusual? Why? c) What percentage of shelf lives would be expected
Solution to be between 13.4 and 15.5 weeks?
µ − 3σ = 13.4 − 3× 1.8 Solution
=8
µ + 3σ = 13.4 + 3× 1.8
= 18.8
So the shelf life of spices almost certainly lies
between 8 and 18.8 weeks.
A shelf life of 20 weeks is outside this range, so it
would be unusual. 𝜇 = 13.4 so its 𝑧 – score = 0.
b) Find the 𝑧 – score for a shelf life of 15.5 weeks. P(13.4 ≤ X ≤ 15.5) = P(0 ≤ z ≤ 1.17)
Solution = P(z ≤ 1.17) − P(z ≤ 0)
x−µ = 0.8790 − 0.5000 (using probability table)
z=
σ
= 0.379
15.5 − 13.4 So 37.9% of shelf lives would be expected to be
=
1.8 between 13.4 and 15.5 weeks.
= 1.17
EXAMPLE 3
In Year 7 at a school the means weight of students was 59.4 kg and the standard deviation was 3.8 kg. In Year 8, the mean
was 63.5 kg and the standard deviation was 1.7 kg.
John in Year 7, and Deng in Year 8, both weighed 68 kg. Which student was heavier in relation to his Year?
Solution
For John: For Deng:
x−µ x−µ
z= z=
σ σ
68 − 59.4 68 − 63.5
= =
3.8 1.7
≈ 2.263 ≈ 2.647

A 𝑧 – score of 2.647 is higher than a 𝑧 – score of 2.263.

So Deng weighed more in relation to his Year group then John.