S1 5.

1 - Standard Normal Random Variables

Continuous Random Variables and the Normal Distribution
The normal distribution is the most important of all probability distributions. It can be used to
model an enormous range of situations, such as people’s heights and weights, errors in
measurements, and I.Q. scores.

The normal distribution is

symmetrical, with the mean,
median and mode all at its centre.
As values move away from the
mean so their probabilities
decrease. 68% of the distribution
lies within ±1 standard deviation
of the mean, 95% within ±2
standard deviations, 99.7% within
±3 standard deviations.

The normal distribution is a continuous distribution, unlike the discrete distributions we were
looking at in the previous section. This means that the probability that a normal variable
takes a single value is always zero. Also, when considering a range of values, the symbols
< and ≤ can be used interchangeably.

P(Z = k) = 0 for all k

P(Z ≤ k) = P(Z < k) P(Z  k) = 1 – P(Z < k)

Standard Normal Variable, Z ~ N(0, 1)

The probability density function for a normal distribution is a complicated function which
involves integration. To make calculations easier, we use standard normal tables, an
extract from which is shown below. These tables apply to what is known as a standard
normal variable, Z, which has mean zero and standard deviation equal to one; we write Z ~
N(0, 1).

Looking at the left hand side of the table below, the numbers in the first column are the values
of z correct to 1 decimal place, while the values across the top are the values of z to the
second decimal place. The '4 d.p.' values are 'lower tail' probabilities corresponding to the z-
values. So, P(Z < 0.40) = 0.6554, P(Z < 0.41) = 0.6591, and P(Z < 0.42) = 0.6628.

1 2 3 4 5 6 7 8 9
z 0.00 0.01 0.02 0.03 ADD (× 0.0001)
0.0 0.5000 0.5040 0.5080 0.5120 4 8 12 16 20 24 28 32 36
0.1 0.5398 0.5438 0.5478 0.5517 4 8 12 16 20 24 28 32 36
0.2 0.5793 0.5832 0.5871 0.5910 4 8 12 15 19 23 27 31 35
0.3 0.6179 0.6217 0.6255 0.6293 4 7 11 15 19 22 26 30 34
0.4 0.6554 0.6591 0.6628 0.6664 4 7 11 14 18 22 25 29 32
0.5 0.6915 0.6950 0.6985 0.7019 3 7 10 14 17 20 24 27 31
The values across the top of the right hand side of the table correspond to the third decimal
place of z. The remaining values on the right must be (multiplied by 0.0001 and then) added
to the probabilities from the left hand side of the table. So:

S1 5 The Normal Distribution 1 14/08/2025

 P(Z < 0.411) = 0.6591 + 0.0004 = 0.6595 P(Z < 0.418) = 0.6591 + 0.0029 =
0.6620

We can calculate the corresponding upper tail probabilities using the fact that the total
probability must be equal to 1, so P(Z > 0.411) = 1 – 0.6595 = 0.3405
P(Z > 0.418) = 1 – 0.6620 = 0.3380 etc.

The diagram on the left represents P(Z < k) for k > 0, the diagram on the right represents P(Z
> k) for k > 0.

Unfortunately, standard normal tables only give probabilities for Z < k when k ≥ 0. If k < 0
we need to make use of the symmetry of the distribution. So, for example:

P(Z > –0.82) = P(Z < +0.82) = 0.7939 P(Z < –0.82) = P(Z>0.82) = 1 –
P(Z<0.82) = 1 – 0.7939 = 0.2061

Finally, to find the probability that Z

lies between two values:

P(k1 < Z < k2) = P(Z < k2) – P(Z

< k1 )

Examples
Q1. Use standard normal tables to calculate the following:
(a) P(Z < 0.52) (b) P(Z > 0.52) (c) P(Z > –0.332) (d)
P(Z < –0.136)

Q2. Calculate the following probabilities for the standard normal variable Z ~ N(0, 1):
(a) P(0.039 < Z < 0.439) (b) P(–0.204 < Z < 0.204) (c) P(–1.5 < Z <
1.2)

Q3. Given that Z ~ N(0, 1), find the values of k such that:
(a) P(Z < k) = 0.6591 (b) P(Z > k) = 0.6591 (c) P(Z < k) =
0.4114

See also Dobbs & Miller Statistics 1p.142 Exercise 9A Q1 – 4

S1 5 The Normal Distribution 2 14/08/2025

 S1 5.2 General Normal Random Variables
The Standard Normal Variable Z ~ N(0, 1)
We saw previously the standard normal variable, Z ~ N(0, 1), had mean 0 and standard
deviation 1. We saw also how to use standard normal tables to calculate probabilities for the
standard normal variable:

P(a < Z < b) = P(Z < b) – P(Z < a)

For k > 0: P(Z < k) = value from normal tables P(Z > k) = 1 – P(Z <
k)
For k < 0: P(Z < k) = P(Z > –k) = 1 – P(Z < –k) P(Z > k) = P(Z < –k)

Notice that changes always occur in pairs. Perform two of:

 reverse the inequality,
 change from P(Z) to 1 – P(Z),
 change from z-value from k to –k.

General Normal Variable X ~ N(μ, σ2)

We’re going to need probabilities for normal distributions whose means are not 0 and
standard deviations are not 1. If a continuous random variable X represents a normal
distribution with mean μ and standard deviation σ then we write: X ~ N(μ,
σ 2)

Standardising a General Normal Variable

The value σ2 is known as the variance of X and is equal to the standard deviation squared, or
(standard deviation)2 . In order to calculate probabilities for a general normal variable we use
standard normal tables. To do this we must change our general X values to standard Z values.
The process of doing so is known as standardisation.

The formula to change from a general to a standard normal variable is:

***LEARN***

To change back from a standard normal variable to the general normal variable, or to
destandardise, we us

the reverse formula:

Example 1
The normal random variable X satisfies X ~ N(100, 52). Calculate:
(a) P(X < 110) (b) P(X > 112) (c) P(X > 95) (d) P(95
< X < 110)

Solution 1
(a) μ = 100, σ = 5, so

(b)

S1 5 The Normal Distribution 3 14/08/2025

(c)

(d)

= -

Example 2
Given that X ~ N(80, 10), (a) calculate P(X < 77)
(b) find the value of k if: (i) P(X < k) = 0.6 (ii) P(X < k) =
0.25

Solution 2

(a) μ = 80, σ = √10 so

= = 1-

(b) (i) From tables: P(Z < 0.253) = 0.5987 + 0.0012 = 0.5999

(ii) From tables: P(Z < 0.674) = 0.7486 + 0.0013 = 0.7449

So P(Z < - 0.674) = 0.2501

Hence

Example 3
Given that X ~ N(10, 4), calculate:
(a) P(X < 12) (b) P(X > 14.5) (c) P(X > 8) (d) P(X < 8.5)
(e) P(9.5 < X < 11.5) (f) the value of k if: (i) P(X < k) = 0.85 (ii) P(X > k) =
0.72

See also Dobbs & Miller Statistics 1p.145 Exercise 9B Q1 - 8

S1 5 The Normal Distribution 4 14/08/2025

 S1 NORMAL DISTRIBUTION 5.3 – Finding μ or σ or
both
Finding mu and sigma
For a general normal distribution, X ~ N(μ, σ 2), a single probability value is enough to
calculate one of μ or σ if the other is known. Given two probability values, both μ and σ can be
found.

Example 1
A continuous random variable X follows a normal distribution with mean μ and variance 9.
Given that P(X < 30) = 0.9525 find the value of μ.

• From standard normal tables, P(Z < 1.67) = 0.9525

• Hence:

Example 2
Given that X ~ N(–10, σ2) and P(X > 0) = 0.0778, find the value of σ.

• P(X > 0) = 0.0778 Þ P(X < 0) = 1 – 0.0778 = 0.9222

• From standard normal tables, P(Z < 1.42) = 0.9222

• Hence

Example 3
A continuous random variable X follows a normal distribution with mean μ and variance σ2.
Given that P(X < 225) = 0.8212 and P(X < 200) = 0.4052, find the values of μ and σ.

• From standard normal tables, P(Z < 0.92) = 0.8212, so:

• The value 0.4052 does not appear in normal tables, but 1 – 0.4052 = 0.5948 does, and
P(Z < 0.24) = 0.5948

• Further, P(Z < –0.24) = 1 – 0.5948 = 0.4052. Hence:

• equation (1) – equation (2) Þ

See also: Dobbs & Miller Statistics 1p.145 Exercise 9B Q9 – 12

p.148 Exercise 9C Q7 – 10

S1 5 The Normal Distribution 5 14/08/2025

 S1 5.4 - The Normal Approximation to the Binomial
The Binomial Distribution
Recall that a binomial distribution refers to:
 n independent trials of an experiment which can have two possible outcomes, success
and failure, where p = P(success) is constant
 If X ~ Bin (n, p) then E(X) = μ = np and Var (X) = σ2 = npq where q = 1 – p

In the diagrams below the vertical lines represent binomial probabilities, while the curve
represents a general normal random variable, X ~ N(np, npq). In both cases, n = 20, while p
= 0.5 on the left and p = 0.8 on the right.

n = 20, p = 0.5 n = 20, p = 0.8

In the diagram on the left, the normal curve matches the binomial probabilities very closely,
although on the right, since p is a long way from 0.5, the match is less good. For values of p
close to 0.5, the normal distribution X ~ N(np, npq) is a good approximation to X ~ Bin(n, p)
even for relatively small values of n. The approximation breaks down, however, as p moves
away from 0.5

Continuity Correction
There is another problem with using the normal distribution to approximate binomial
probabilities. General normal random variables are continuous random variables, while the
binomial distribution is a discrete distribution.

n = 20, p = 0.4 n = 30, p = 0.65

These diagrams illustrate how the area under the normal curve approximates to the binomial
probabilities. For example, the binomial probability P(X B = 12) can be approximated by the
normal probability
P(11.5 < XN < 12.5). This transformation from a single value of the discrete variable to a
range of values of the continuous variable is known as the continuity correction.

S1 5 The Normal Distribution 6 14/08/2025

 n = 50, p = 0.7 n = 150, p = 0.2

These last two diagrams illustrate how, as n increases the approximation remains valid even
for values of p a long way from 0.5.

The Normal Approximation to the Binomial

The binomial random variable XB ~ Bin(n, p)

can be approximated by the normal random variable XN ~ N(np, npq) where q

=1–p

provided that np > 5 and nq > 5. A continuity correction must be employed when
calculating probabilities, so:
P(XB = k) ≈ P(k – 0.5 < XN < k + 0.5)

Example 1 Given that X ~ Bin (70, 0.3), by using a suitable approximation, calculate
approximate values for: (a) P(10 ≤ X ≤ 20) (b) P(22 < X < 30) (c)
P(X > 30)
[Ans: 0.4469, 0.3345, 0.0066]

 n = 70, p = 0.3, so X ~ N(21, 14.7) approximately

 Employing a continuity correction: P(10 ≤ XB ≤ 20)

P(9.5 < XN < 20.5)

 (both 10 and 20 need to be included, so we start at 9.5 and finish at 20.5)




 From tables:

 Hence, P(10 ≤ XB ≤ 20) =

approximately

Example 2
A manufacturer of biscuits produces 3 times as many cream ones as chocolate ones. A large
box of randomly chosen biscuits contains 120 biscuits. Using a suitable approximation, find
the probability that it contains fewer than 35 chocolate biscuits. [Ans: 0.829]

Example 3
Kamal has 30 hens. The probability that any hen lays an egg on any day is 0.7. Hens do not
lay more than one egg per day, and the days on which a hen lays an egg are independent. Use
a suitable approximation to calculate the probability that Kamal’s hens lay fewer than 20 eggs
on a particular day. [Ans: 0.275]

S1 5 The Normal Distribution 7 14/08/2025

See also: Dobbs & Miller Statistics 1p.157 Exercise 9D All questions

S1 5 The Normal Distribution 8 14/08/2025