Normal Approximations to

Binomial Distributions
Jamie Frost
[Link]
@DrFrostMaths

Contact the resource team:

resources@[Link]
@DrFrostResource

Dr Frost Learning is a registered

Last modified:7th November 2024 charity in England and Wales (no
1194954)
Teacher Notes
Prerequisite
Knowledge
• 68-95-99.7 rule
• Probability of ranges using the normal
distribution.
• Inverse normal distribution.
• Standard normal distribution (unknown or )
• Binomial distribution

Throughout the slides, this symbol refers to a web link.

Unless
Key: otherwise specified, this will be to some functionality
within DF.
Key Points Solution step – All slides include
click to reveal pedagogical detail in the
! To be written ‘Notes’ section for each
in books Question/Discussion slide.
Dr Frost Learning is a registered
Prompt charity in England and Wales (no
How to use these slides
Though many slides in this resource will have titles specific to the topic, the slide titles in the
table below are used consistently within DFL resources for specific pedagogical purposes.
Any atypical use of a slide type, including any change of animation* or intended use, will be
outlined in the Teacher Notes for the slide.
Slide Title Explanation Default Animations*
To be used as a prior knowledge check or to review
Recap prerequisite knowledge. Can be used as a starter or as part of Green click-to-reveal boxes.
the main lesson.
To be used to highlight key concepts or theorems. This could
Usually in sequence with
The Big include the ‘why’ of the topic - including “real-life” contextual
some green click-to-reveal
Idea scenarios, or putting into context of other mathematical
boxes.
concepts (past and future).
Solution animates in
Example To be modelled by the teacher.
sequence.
Green click-to-reveal boxes.
Test Your
To be completed by students and used for Assessment for For multi-step answers,
Understandi
Learning, primarily using mini-whiteboards. reveal in parts or click final
ng
answer to reveal full solution.
To be used as ‘Example’ &‘Test Your Understanding’ above, Example animates in
To be used as ‘Example’ &‘Test Your Understanding’ above,
Example within the same slide to provide scaffold via visible modelled sequence,
Examplefollowed
animates byinTYU
Example within the same slide to provide scaffold via visible modelled
Problem solution. question with
sequence. Clickgreen click-to-
the header to
Problem solution.
Pair TYU column is blank initially, to focus attention on example. reveal boxes for solution
reveal TYU question, then
Pair TYU column is blank initially, to focus attention on example.
Reveal question by clicking ‘Test Your Understanding’ steps.
green click-to-reveal boxes.
banner.
To be used as fluency practice. Multiple questions in rapid Green click-to-reveal boxes.
Quickfire succession,
To be used for calculations
as fluency that can
practice. be completed
Multiple questions mentally.
in rapid For multi-step
Green answers,
click-to-reveal boxes.
Questions
Quickfire Often used forfor
succession, shorter questions/
calculations that formulae or to isolate
can be completed a small
mentally. reveal in parts oranswers,
For multi-step click final
Questions Often used for shorterpart of the method.
questions/ formulae or to isolate a small line toin
reveal reveal
partsfull solution.
or click final
part of the method. line to reveal full solution.
To be used as a diagnostic question. Multiple choice questions,
Multi-choice with
To be plausible
used distractors,
as a diagnostic to allow
question. teachers
Multiple
Dr Frost to diagnose
choice
Learning is questions, Arrowinpoints
a registered charity Englandto answer,
and Wales on (no
Recap: Normal Distribution
IQ (‘Intelligence Quotient’) is a measure of intelligence, with
distribution .

a Determine the probability that a randomly chosen person has

an IQ above .
b Determine the IQ for which of people have an IQ above.

Use your calculator’s

distribution functionality,
a s.f. ? using as the lower
𝑓 (𝑥 ) boundary.
b
? Draw a diagram. To use the
inverse normal
distribution on your
calculator, you need the area
0.6 𝟎.𝟒 up to a particular value.
100𝑘
Recap: Binomial Distribution

At a restaurant, an average of out of every

customers order salmon for the main course.
A random sample of customers is selected.

a State a suitable distribution to model the number

of customers that will choose salmon, .
b Find the probability that exactly people order
salmon.
a Recall that a binomial
?
distribution is suitable when
we want to find the
b probability of a given
3 s.f. ? number of successes out of
a fixed number of trials.

is a binomial distribution
You can work this out where is the number of trials
directly using a calculator, and is the probability of
rather than needing to use success on each trial.
the formula.
Approximating a Binomial Distribution as a
Normal
What things do you notice about
𝑝 (𝑥 ) this binomial distribution?

𝑋 𝐵 ( 8 ,0.5 )

It’s symmetrical: because the The shape

chance of getting success vs failure resembles a
is 50-50, the probability of getting normal
6 successes is exactly the same as distribution.
Approximating a Binomial Distribution as a
Normal
What things do you notice about
𝑝 (𝑥 ) this binomial distribution?

It’s asymmetrical: we
expect successes in 8 𝑋 𝐵 ( 8 ,𝟎 . 𝟖 )
trials, so there is a
positive skew towards
a greater number of
successes

𝑥
Up to 7 successes, the shape of the distribution is still
normal-like. The problem is that the outcome of this
binomial is capped at 8 successes, whereas a normal
distribution is unbounded. The lack of symmetry of the
binomial, combined with the limited range of its outcomes,
Motivation
Suppose the number
of trials in the
Binomial distribution
is large.
Suppose also that we’re trying
to find the probability of a
range.

This calculation involves lots of factorials. e.g.

We would
need to add
Large factorials are notoriously time-
the
intensive to calculate in mathematics. We can
probabilities of
get approximations for them (see Stirling
each
approximations), but there is no quick way to
individual
calculate them exactly.
outcome in
Motivation

𝑝 (𝑥 ) 𝑓 (𝑥 )

𝑷 ( 𝑿 ≤ 𝟑)

𝑥 𝑷 (𝒀 ≤? ) 𝑥
𝑋 𝐵 ( 8 ,0.5 ) 𝑌 𝑁 (? ,? )
Question 1: What mean But if we somehow approximate
and standard deviation the binomial distribution as a
should we use for the normal distribution, finding an
normal distribution? area under the graph can be
Question 2: What range do done very efficiently by a
we find, equivalent to the ? calculator.
The Approximating Distribution
Notation Watch: We typically
use for the original distribution
and for the approximating
distribution.

𝑿 𝐵 ( 8 , 0.5 ) 𝒀 𝑁 ( 𝜇 , 𝜎2)
What would be a sensible to use for ?

The mean of the normal distribution should

match that of the binomial distribution.
Recall the mean value (known as the ‘expected
value’) of a binomial is
What would be a sensible to use for ?

Again, this should match the binomial

distribution.
The variance is

! If is large and close to , then the binomial

distribution can be approximated by the normal
distribution where
Quickfire Approximating Distributions
1
Mental trick:
Original Distribution Approximating multiply this
Distribution number by

a ?
b ?
c ?

d ?

e ?

f ?
Continuity Corrections
Suppose we wanted to use to estimate . What
problem would we face if we tried to calculate ?

𝑝 (𝑥 ) 𝑓 (𝑥 )
𝑷 ( 𝑿 =𝟑 )
𝑷 ( 𝒀 =𝟑 )

𝑥 𝑥
𝑋 𝐵 ( 8 ,0.5 ) 𝑌 𝑁 ( 4, √ 2 )
2

Because normal distributions are continuous, it

makes no sense to calculate , because the
probability is effectively 0. For a normal
distribution, we’d typically calculate the probability
of range of values.
Continuity Corrections

The solution is to divide the normal

distribution up into regions of width 1.
What region would we find to
estimate ?
𝑝 (𝑥 ) 𝑓 ( 𝑥 ) 𝑃 ( 𝑋 =3 ) ≈ 𝑷 ( 𝟐 . 𝟓 ≤𝒀 ≤ 𝟑. 𝟓 )
𝑷 ( 𝑿 =𝟑 )

𝑥 𝟐 . 𝟓𝟑 . 𝟓 𝑥
𝑋 𝐵 ( 8 ,0.5 ) 𝑌 𝑁 ( 4, √ 2 )
2
Examples
! A continuity correction is
approximating a discrete range using a
continuous one.
1. If or , convert to first.
2. Enlarge the range by at each end.
Determine the continuity correction for the following original ranges
of a binomial distribution.

a𝑃 ( 𝑋 =5 ) ≈ 𝑷 ( 𝟒 . 𝟓 ≤ 𝒀 ≤ 𝟓 .𝟓 )
b𝑃 ( 𝑋 ≥ 4 ) ≈ 𝑷 ( 𝒀 ≥𝟑 . 𝟓 )
c𝑃 ( 𝑋 < 2 ) ¿ 𝑃 ( 𝑋 ≤ 1)
≈ 𝑷 ( 𝒀 ≤𝟏 . 𝟓 )
d𝑃 ( 5 ≤ 𝑋 ≤ 7 )≈ 𝑷 ( 𝟒 . 𝟓 ≤ 𝒀 ≤ 𝟕 .𝟓 )
Quickfire Continuity Corrections
! A continuity correction is
approximating a discrete range using a
continuous one. We are making
1. If or , convert to first. the range 0.5
‘longer’.
2. Enlarge the range by at each end.
2
Original Range Range for
a ?
b ?
c ?
d ?
e ?
f ?
g ?
h ?
The Full Process
Let .
a Use a normal approximation, estimate
the value of , to 5 significant figures.
b Calculate the percentage error from the
true value of
Step 1: Write the
a approximating distribution .

Step 2: Use the continuity

to 5 s.f. correction on the range.

Step 3: Use your

b ¿ 0.16313 calculator’s normal
distribution mode.
Find the actual value using the
binomial distribution given, using
Percentage error the binomial distribution mode on
your
Side Notes: Notice that calculator.
even though is not close to , the normal
approximation is still extremely accurate. This is because,
although the binomial distribution is not symmetrical, as we saw
earlier for , we still get a normal shape but one of the tails
distorted due to the ‘capping’. However, as the left tail, far away
from the
mean of , represents such a tiny amount of probability,
Test Your Understanding
3 Let .
Use a normal approximation, estimate the
value of , to 5 significant figures.

to 5sf ?

4 Let .
Use a normal approximation, estimate the
value of , to 5 significant figures.

to 5sf ?
In a Modelled Context
5 [Edexcel A Level Jun 2022 P3 Stats Q1b]
Each time George throws the ball, the probability of the ball
hitting the target is .
The random variable represents the number of times George
hits the target in throws.
George now throws the ball at the target times.
Use a normal approximation to calculate the probability that
he will hit the target more than times.
(3 marks)

The only additional step is that we

must work out the distribution
ourselves using the given
information.
?
Finding the in
It may be that we are given the probability of a range, but what this
range is, is unknown.

[Edexcel S2 June 2018 Q5e Edited]

The proportion of customers who buy organic fruit from Tesson
supermarket is . During a particular day, a random sample of
customers from Tesson supermarket is taken. Using a suitable
approximation, the probability that fewer than of these customers
bought organic fruit is correct to decimal places.
Find the value of .
(6 marks) Solve the
equation.
Find the approximating must be a whole
distribution as usual. number. The
small discrepancy
We are told is because was a
“probability of fewer rounded value.
than ”
Be careful with the
Because is unknown, use the
continuity
distribution to find the value of
correction.
corresponding to
Then use
See skill 643.
Test Your Understanding
6 [OCR A2 June 2019 P2 Q13]
It is known that of adults in the UK use a certain app. A researcher
selects a random sample of adults in the UK. The random
variable is defined as the number of adults in the sample who use
the app.
Given that , calculate the largest value of .

?
Unknown Number of Trials
A fair coin is thrown times.
Using a normal approximation, the probability it comes down
heads at most times is correct to 4 significant figures.
Determine the value of .

Use and as Let

usual.
or

Continuity
correction
.
We don’t know or , so
use , and the inverse
0.02461 normal mode on your
420.5 calculator with the area
as
This is a
quadratic in
terms of
Test Your Understanding
7 [Edexcel S2 June 2016 Q5]
In a large school, of students own a touch screen laptop.
A random sample of students is chosen from the school.
Using a normal approximation, the probability that more than of
these students own a touch screen laptop is correct to 3
significant figures.
Find the value of .
(8 marks)

Use and as Let

usual.
or
This is a
Continuity quadratic
correction. in terms of

?
We don’t know
or , so use , and
0.9599 the inverse normal
0.0401 mode on your
55.5
calculator with the
area as
Appendix: Why Do We Get the Normal
Distribution?
Why does the binomial distribution resemble the normal
distribution when is large?

𝑋 𝐵𝑒𝑟𝑛𝑜𝑢𝑙𝑙𝑖 ( 0.6 ) Suppose we considered just 1 trial.

We output 1 on success and 0
otherwise.
This special case of the binomial
distribution, where , is known as a
Bernoulli distribution (see skill
𝑋1 837).
We can add discrete random
variables together to represent the
𝑋2 sum of the outcomes. For example, if
and are each distributed using ,
then represents their sum. Since
each can be 0 or 1, the outcome of
can be 0, 1 or 2.
Appendix: Why Do We Get the Normal
Distribution?
We could get a binomial distribution of trials by
simply adding Bernoulli distributions together.

Each Bernoulli gives 1 for success and 0

for failure, so by adding of them together,
we get the count of successes out of
trials.

𝑋 1+ 𝑋 2+ 𝑋 3 + …+ 𝑋 𝑛

The Central Limit Theorem (skills 850,868) states that

if we find the sum or mean of identically
distributed variables (i.e. copies of the same
random variable), then as the number of variables
becomes large, the resulting distribution is
approximately normally distributed.

Since the binomial distribution is a sum of identical

Bernoulli distributions, then by the CLT, we get an
approximate normal distribution as a result.