Normal Distribution
Normal Distribution
Normal Distribution
This lesson includes an overview of the subject, instructor notes, and example exercises using
Minitab.
Normal Distribution
Lesson Overview
The normal distribution is probably the most widely used of all probability models. Its
popularity is partly due to the Central Limit Theorem, which states that as the sample size
increases, the sampling distribution of the mean can be approximated by a normal distribution.
Some variables typically modeled by the normal distribution include height, weight, IQ, memory,
reading ability, etc.
The normal distribution is completely described by two population parameters – the mean
and the standard deviation σ. The mean of a normal distribution is its center, and the standard
deviation measures the spread of the distribution about the mean.
In this lesson, the normal distribution will be defined, and we’ll use it to determine probabilities
for variables that can be modeled by it. We’ll compute these probabilities using a normal table
and Minitab.
Prerequisites
None. Although integration of the normal probability density function is included, covering it is
optional.
Learning Targets
• View, recognize, and graph a normal distribution with mean μ and standard deviation σ.
The 68–95–99.7 rule, also known as the 3σ rule or empirical rule, will be discussed.
• Determine probabilities associated with a normal distribution using a normal probability
table and Minitab.
WWW.MINITAB.COM/ACADEMIC
• Standardize values from any normal distribution. In other words, convert x values
associated with a normal distribution with mean μ and standard deviation σ to z-scores.
A z-score will be defined in the lesson.
• Work backwards when given a probability from a normal distribution to determine its
corresponding x value.
Time Required
It will take the instructor 60 minutes in class to introduce the normal distribution and work
through examples for determining probabilities and corresponding x values. We recommend
starting the activity sheet in class so that students can ask the instructor questions while working
on it. The exercises on the activity sheet will take 60 minutes, and they can be used as
homework or quiz problems.
Materials Required
Assessment
The activity sheet contains exercises for students to assess their understanding of the learning
targets for this lesson.
Possible Extensions
The instructor may want to extend this to the Binomial Distribution and/or Other
Distributions after this lesson so that students can have exposure to additional distributions.
References
Graphing the probability density function – which describes probabilities over intervals –
associated with a normal distribution reveals a bell-shaped curve with a center at mean μ and a
spread of standard deviation , where is always positive. The normal distribution is
completely described by these two population parameters.
Below are graphs of normal distributions with various means and standard deviations. Notice
that the bell of the curve is located at the mean μ. The shape of the curve is dependent on the
standard deviation , where a larger corresponds to a wider curve.
1 (𝑥−𝜇)2
−
𝑓(𝑥) = 𝑒 2𝜎 2 for all real number 𝑥 ′ s
𝜎√2𝜋
The function f(x) is called a probability density function and graphing it for any μ and positive
yields a bell curve centered at μ with spread . The function f(x) is a legitimate probability
density function since f(x) > 0 for all real numbers x and the area under the curve of f(x) is 1.
x2
1
• Mean μ=0 and standard deviation =1 is written as: 𝑓(𝑥) = e −
2 for all real number
√2π
x’s.
(x−3) 2
1
Mean μ=3 and standard deviation =2 is written as: 𝑓(𝑥) =
−
• 2√2π
e 2∙22 for all real number
x’s.
• Approximately 68% of the data points lie within 1 standard deviation (1) of the mean
μ. See the yellow area in the picture below.
• Approximately 95% of the data points lie within 2 standard deviations (2) of the
mean μ. See the pink and yellow areas combined below.
• Approximately 99.7% of the data points lie within 3 standard deviations (3) of the
mean μ. See the blue, pink, and yellow areas combined below.
The empirical rule is useful in estimating probabilities for data that is modeled by a normal
distribution.
Minitab 19
Minitab Express
Distribution Plot
Normal, Mean=1 00, StDev=1 0
0.04
0.03
Density
0.02
0.01
0.00
70 80 90 1 00 110 1 20 1 30
X
The curve is centered at μ = 100 and the spread is = 10. We can find the desired probabilities
using the empirical rule.
(a) The probability that a randomly chosen person has an IQ less than 90 is approximately 16%.
• By the empirical rule, approximately 68% of IQ’s are within 1 of the mean μ = 100.
Therefore approximately 68% of IQ’s are between 90 and 110 with approximately 32%
outside 90 and 110.
Minitab 19
Minitab 19 (Mac)
Minitab Express
(b) The probability that a randomly chosen person has an IQ greater than 120 is approximately
2.5%.
• By the empirical rule, approximately 95% of IQ’s are within 2 of the mean μ = 100.
Therefore approximately 95% of IQ’s are between 80 and 120 with approximately 5%
outside 80 and 120.
• By symmetry, approximately 2.5% of IQ’s are greater than 120 (and approximately 2.5%
are less than 80).
We can calculate this probability in Minitab as we did for part (a). The only difference is that we
want to view the shaded area corresponding to the Right Tail for a specific x value—in this case,
our x value is 120, since we want to determine the probability of an IQ being greater than 120.
The graph below on the left shows the desired probability (2.3%).
• By the empirical rule, approximately 95% of IQ’s are within 2 of the mean μ = 100.
Therefore approximately 95% of IQ’s are between 80 and 120.
• By symmetry, approximately 47.5% of IQ’s are between 100 and 120.
We can calculate this probability in Minitab as we did in (a) and (b). Now, we need to view the
shaded area corresponding to the Middle for two x values—in this case, X value 1 is 100 and X
value 2 is 120, since we want to determine the probability that a person’s IQ is between 100 and
120. The graph above on the right shows the desired probability (47.7%).
Computing z-Scores
Considering the empirical rule and its application to all normal distributions, it is useful to talk
about a value from a normal distribution in terms of the number of standard deviations it is
from the mean. That is, we can measure the distance between an x value and the center at mean
μ in units of size . Converting a value to these distance units is called standardizing, or
standardization, where a standardized value is called a z-score.
Suppose the data value x is from a normal distribution with mean μ and standard deviation . To
standardize x, we
𝑥−𝜇
𝑧=
𝜎
A z-score measures the number of standard deviations between an x value and the mean μ. If a
z-score is positive, then the x value is to the right of the mean μ. If a z-score is negative, then the
x value is to the left of the mean μ.
90 − 100
𝑧= = −1
10
This z-score indicates that the value 90 is 1 standard deviation to the left of the mean μ = 100.
120 − 100
𝑧= =2
10
This z-score indicates that the value 120 is 2 standard deviations to the right of the mean μ =
100.
If X is a normal random variable with mean μ and standard deviation σ, then the random variable:
X −
Z=
Since any normal random variable can be converted to a standard normal random variable, a
“normal table” appears in statistics textbooks to determine the area under the standard normal
curve. This area corresponds to the proportion of data values or the probability of being beyond
some z value on the standard normal curve. You can find a normal table at the end of this lesson.
Example 3
Assume Z is a standard normal random variable. Determine the following probabilities using the
standard normal table provided at the end of this lesson. Shade the area(s) of interest on the
graphs.
P(Z 1.25) is the area to the left of 1.25 on the standard normal curve. This is the type of area,
less than a designated value, which is directly given in the normal table.
• The first two digits of the value, 1.2, are contained on the left hand side of the table.
• The last decimal place, 5, is shown along the top of the table.
• The corresponding area (probability) is shown in the body of the table. It is
approximately 0.89435.
Minitab 19
Minitab 19 (Mac)
Minitab 19
Minitab 19 (Mac)
Minitab Express
P(Z > 2.07) is the area to the right of 2.07 on the standard normal curve.
• The first two digits of the value, 2.0, are contained on the left hand side of the table.
• The last decimal place, 7, is shown along the top of the table.
• The area (probability) in the body of the table is the area to the left of 2.07. Since the
total area under the curve is 1, the area to the right of 2.07 is approximately 1 – 0.98077
= 0.01923.
Minitab 19 (Mac)
Minitab Express
• The first two digits of the value, 1.7, are contained on the left hand side of the table.
• The last decimal place, 8, is shown along the top of the table.
• The area (probability) in the body of the table is the area to the left of 1.78, which is
approximately 0.96246.
• The first two digits of the value, 0.5, are contained on the left hand side of the table.
• The last decimal place, 6, is shown along the top of the table.
• The area (probability) in the body of the table is the area to the left of 0.56, which is
approximately 0.71226.
Subtract P(Z < 0.56) from P(Z < 1.78), which yields 0.96246 – 0.71226 = 0.2502.
We can check this in Minitab as we did for the previous two parts. Instead of selecting Left Tail
or Right Tail, we need to select Middle in the Shaded Area menu and type 0.56 for X value 1
and 1.78 for X value 2.
Example 4
The time it takes a cell to divide (called mitosis) is normally distributed with an average of one
hour and a standard deviation of 5 minutes. What proportion of dividing times will take less
than 65 minutes?
If we let X represent the time for a cell to divide, then X has a normal distribution with mean μ =
60 minutes and standard deviation = 5 minutes. The z-score for the time 65 minutes is:
65 − 60
𝑧= =1
5
The probability that the cell division time is less than 65 minutes can be found using the normal
table.
Let X represent Bella’s jumping height. Then X has a normal distribution with mean μ = 3 feet
and standard deviation = 0.8 feet. The z-score for the height 2.4 feet is:
2.4 − 3
𝑧= = −0.75
0.8
The probability that her jumping height is greater than 2.4 feet can be found by using the
normal table.
If we let X represent the reaction time, then X has a normal distribution with mean μ = 1.25
seconds and standard deviation = 0.46 seconds. The z-scores for the times 1.00 second and
1.75 seconds are:
The probability that the reaction time is between 1 and 1.75 seconds can be found using the
normal table.
𝑃(1 < 𝑋 < 1.75) = 𝑃(𝑋 < 1.75) − 𝑃(𝑋 < 1) = 𝑃(𝑍 < 1.09) − 𝑃(𝑍 < −0.54) ≅ 0.86214 − 0.29460
= 𝟎. 𝟓𝟔𝟕𝟓𝟒
Computing this probability in Minitab, we get a slightly more accurate answer than provided by
the normal table.
Example 7
What z-score corresponds to the proportion 0.1 in the left tail of a standard normal distribution?
Drawing a picture of the normal distribution with the corresponding proportion is a good way to
start.
Example 8
What z-score corresponds to the proportion 0.15 in the right tail of a standard normal
distribution?
Since we are determining a right tail proportion, determine the z-score in the normal table
corresponding to 1 – 0.15 = 0.85. Looking in the body of the normal table for the value closest
to 0.85 gives us z = 1.04.
Example 9
Mensa (from the Latin word “mind”) is an international society devoted to intellectual pursuits.
Any person who has an IQ in the upper 2% of the general population is eligible to join. Assume
that IQs are normally distributed with = 100 and σ = 15. What is the lowest IQ that will qualify
a person for Mensa?
Let X represent a person’s IQ score. We are trying to determine the x value shown in the graph
below.
• Since we are using a right-tail proportion, determine the z-score in the normal table
corresponding to 1 – 0.02 = 0.98. Looking in the body of the normal table for the value
closest to 0.98 gives us z = 2.06.
• Now we need to “unstandardize” the z-score back to the original x (IQ) scale. We know
that x must satisfy the following equation:
𝑥 − 100
= 2.06
15
𝑥 − 100 = 2.06 ∙ 15
𝑥 = 100 + 2.06 ∙ 15
𝑥 = 130.9
Minitab 19
Minitab 19 (Mac)
Minitab Express
• Since we are determining a right tail proportion, determine the z-score in the normal
table corresponding to 1 – 0.90 = 0.10. Looking in the body of the normal table for the
value closest to 0.10 gives us z = -1.28.
• Now we need to “unstandardize” the z-score back to the original x (lifetime of bulb)
scale. We know that x must satisfy the following equation:
𝑥 − 1000
= −1.28
100
Using Minitab and the instructions for Example 9, we obtain x = 871.8 hours.