Chapter 2:

Random Variables and Probability Distributions

Lesson 8: The Normal Distribution and Its Properties
TIME FRAME: 1 hour
LEARNING OUTCOME(S): At the end of the lesson, the learner should be able to:

describe a normal random variable and its characteristics

draw a normal curve
state the empirical rule
PRE-REQUISITE LESSONS: Random Variables, Probability Distribution of Continuous
Random Variables
KEY CONCEPTS: Mean, Variance of a Random Variable

LESSON OUTLINE:
1. Review of Continuous Random Variables
2. Motivation: Distribution of Weights of Babies
3. Main Lesson: The Normal Curve and Its Properties
4. Seatwork: Validating the Empirical Rule in the Weights Distribution
5. Enhancement : History regarding the Normal Curve
6. Further Enhancement: Distribution of Balls in The Quincunx

DEVELOPMENT OF THE LESSON

(A) Introduction: Review of Continuous Random Variables

Ask students to recall the definition of a continuous random variable. (It is a random
variable that can take any real value within a specified range. whereas a discrete random
variable takes some on a countable number of values). Students should also remember that a
continuous variable involves a measurement of something, such as the height of a randomly
selected student, the weight of a newborn baby, or the length of time that the battery of a
cellphone lasts.

(B) Motivation: Distribution of Weights of Babies follows a bell-shaped curve

Consider the following data pertaining to hospital weighs (in pounds) of all the 36 babies that
are born in the maternity ward of a certain hospital.

4.94 4.69 5.16 7.29 7.19 9.47 6.61 5.84 6.83

3.45 2.93 6.38 4.38 6.76 9.01 8.47 6.8 6.4
8.6 3.99 7.68 2.24 5.32 6.24 6.19 5.63 5.37
 5.26 7.35 6.11 7.34 5.87 6.56 6.18 7.35 4.21

with an average of 6.11 pounds and a standard deviation of 1.61 pounds.

Show students the histogram for these data set, or ask them to generate it; help them observe
that the histogram is approximately bell-shaped:
.3
.2
Density

.1
0

2 4 6 8 10
weight

Inform students that the many continuous random variables, such as IQ scores, heights of
people, or weights of M&Ms, have histograms that look bell shaped.

Tell them that the most important distribution in statistical science is a normal distribution,
which has a "bell-shaped" curve. Explain that there are many reasons why the normal
distribution is considered the most important curve in statistics.

(a) Many random variables are either normally distributed or at least, approximately
normally distributed. Heights, weights, examination scores, the log of the lifelength of
some equipment are among a few random variables that are approximately normally
distributed. Although the distributions are only approximately normal, the approximation
is usually quite close.

(b) It is easy for mathematical statisticians to work with the normal curve. A number of
hypothesis tests, and the regression model are based on the assumption that the
 underlying data have normal distributions. (Extra note: There are, however, other kinds
of continuous distributions, that are used in practice. For instance, the distribution that
has been found convenient for modeling the lifelength of equipment is the Weibull
distribution.)

Stress that the normal distribution is a continuous distribution just like the uniform and
triangular distribution. However, the left and right tails of the normal distribution extend
indefinitely but come infinitely close to the x-axis.

Draw a picture of the normal (bell-shaped) curve

Explain that the graph of the normal distribution depends on two factors - the meanand the
standard deviation . In fact, the mean and standard deviation characterize the whole
distribution. That is, we can get areas under the normal curve given information about the
mean and standard deviation.

Mention that the mean determines the location of the center of the bell shaped curve. Thus, a
change in the value of the mean shifts the graph of the normal curve to the right or to the left.

Ask students to recall what the mean, median and mode of a distribution represent. They
should say (a) the mean represents the balancing point of the graph of the distribution; (b)
the mode represents the high point of the probability density function (i.e. gthe graph of the
distribution), (b) the median represents the point where 50% of the area under the
distribution is to the left and 50% of the area under the distribution is to
the right.

For symmetric distributions with a single peak, such as the normal curve, assist to students to
think that the mean = median = mode.

Inform students that the standard deviation determines the shape of the graphs (particularly,
the height and width of the curve). When the standard deviation is large, the normal curve is
short and wide, while a small value for the standard deviation yields a skinnier and taller
graph.
Draw the curves on the board:

Mention to students that the curve above on the left is shorter and wider than the curve on the
right, because the curve on the left has a bigger standard deviation.

Help students notice that a normal curve is symmetric about its mean and is more
concentrated in the middle rather than in the tails, aside from observing that normal curves
differ in how spread out they are (and that the spread or variability is measured by the
standard deviation ).

Tell students that when a random variable has a normal distribution with mean and
variance 2, we denote this as X~N(,2).

Technical Note: The height of a normal curve at some value x is a formidable-looking

expression that depends on the mean and standard deviation :

1 1
f ( x) exp ( x )2 ,
2 2 2

You may notice that involves three famous numbers in the history of mathematics:

2 1.41421235652 ,
3.141592654 and
Eulers number e 2.7182818 .

Students need not be given this expression as they may feel threatened by it. Instead, you
should use the graphical form of the normal distribution by drawing the bell shaped curve.

Emphasize the following statements about the normal curve:

The total area under the normal curve is equal to 1.

The probability that a normal random variable X equals any particular value a ,
P(X=a) is 0 (since it is a continuous random variable).
The probability that X is less than a equals the area under the normal curve bounded
by a and minus infinity (as indicated by the shaded area in the figure below)
 The probability that X is greater than some value a equals the area under the normal
curve bounded by a and plus infinity (as indicated by the non-shaded area in the
figure above).
Since the normal curve is symmetric about the mean, the area under the curve to the
right of equals the area under the curve to the left of which equals , i.e. the
mean is the median.
The probability density function is maximized at , i.e. the mode is also the mean.
The normal curve has inflection points (i.e. point at which a change in the direction of
curvature occurs) at and at
As x increases without bound (gets larger and larger), the graph approaches but never
reaches, the horizontal axis. As x decreases without bound (gets larger and larger in
the negative direction), the graph approaches, but never reaches, the horizontal axis
Emphasize also to students that every normal curve (regardless of its mean or standard
deviation) conforms to the following "empirical rule" (also called the 68-95-99.7 rule):

About 68% of the area under the curve falls within 1 standard deviation of the mean.
About 95% of the area under the curve falls within 2 standard deviations of the mean.
Nearly the entire distribution (About 99.7% of the area under the curve) falls within 3
standard deviations of the mean.
 Explanatory Note: The empirical rule is actually a theoretical result based on an analysis of
the normal distribution. In the first chapter, it was pointed out that the importance of the
mean and standard deviation as summary measures is due to Chebychevs inequality, which
guarantees that the area under a distribution within two standard deviations from the mean is
at least 75%. For nearly all sets of data, the actual percentage of data may be much greater
than the bound specified by Chebychevs Inequality. In fact, for a normal curve, the area
within two standard deviations from the mean is about 95%. Also, about two thirds of the
distribution lie within one standard deviation from the mean and nearly the entire distribution
(99.7%) is within three standard deviations from the mean:

(C) Seatwork: Validating the Empirical Rule

Ask students to determine what is the frequency (and relative frequency) of babies weights
that are within:
 a) One standard deviation from the mean

ANSWER: 26 out of 36; or about 72% . Values within one from are in boldface:

4.94 4.69 5.16 7.29 7.19 9.47 6.61 5.84 6.83

3.45 2.93 6.38 4.38 6.76 9.01 8.47 6.8 6.4
8.6 3.99 7.68 2.24 5.32 6.24 6.19 5.63 5.37
5.26 7.35 6.11 7.34 5.87 6.56 6.18 7.35 4.21

b) Two standard deviations from the mean

ANSWER: 34 out of 36; or about 95% Values within two from are in boldface

4.94 4.69 5.16 7.29 7.19 9.47 6.61 5.84 6.83

3.45 2.93 6.38 4.38 6.76 9.01 8.47 6.8 6.4
8.6 3.99 7.68 2.24 5.32 6.24 6.19 5.63 5.37
5.26 7.35 6.11 7.34 5.87 6.56 6.18 7.35 4.21

c) Three standard deviations from the mean (ANSWER: 36 out of 36; 100%)

Remark:

In place of examining the distribution of weights from babies, you may want to examine
heights or weights of students obtained from the data collection activities in Lesson 1 of
Chapter 1.

(D) Enrichment

Historical Notes on the Normal Curve:

(i) The French-English mathematician Abraham de Moivre first described the use of the
normal distribution in 1733 when he was developing the mathematics of chance,
particularly for approximating the binomial distribution. Marquis de Laplace used the
normal distribution as a model of measurement errors. Adolphe Quetelet and Carl
Friedrick Gauss popularized its use. Quetelet used the normal curve to discuss the
average man with the idea of using the curve as some sort of an ideal histogram
while Gauss used the normal curve to analyze astronomical data in 1809.

(ii) In some disciplines, such as engineering, the Normal distribution is also called the
Gaussian distribution (in honor of Gauss who did not first propose it!). The first
unambiguous use of the term Normal distribution is attributed to Sir Francis Galton
in 1889 allthough Karl Pearson's consistent and exclusive use of this term in his
prolific writings led to its eventual adoption throughout the statistical community.
(E) Further Enrichment : Distribution of Balls in a Quincunx

If students have access to internet, ask them to go to this website

http://www.mathsisfun.com/data/quincunx.html

In this webpage, they are shown a quincunx or "Galton Board" (named after Sir Francis
Galton). This is a triangular array of pegs. Balls are dropped onto the top peg, and then
subsequently they bounce their way down to the bottom where they are collected in little
bins. Each time a ball hits one of the pegs, it bounces either to the left or right with equal
probability, and consequently the number of pegs collecting in the bins form a "bell-shaped"
curve (especially as the number of rows (and bins) as well as the number of balls increases).

Tell students to reset defaults with the simulator and use 6 rows, and drop about 50 balls.

then 12 rows with about 100 balls:

 The chance of falling on the kth peg given n rows and a probability p of bouncing left is
given by the so-called binomial probability distribution,

!
( ) (1 )
! ( )!

In particular, for 12 rows (n=12) and a probability of bouncing left of 0.5 (p=0.5), we can
calculate the probability of being in the 5th bin from the right (k=5) as follows:

12!
(5!7!) 0.55 (0.5)7 = 0.193

In fact we can build the entire probability distribution for rows=12 and probability=0.5 like
this:
Bin number from 12 11 10 9 8 7 6
right
Probability 0.000244 0.00293 0.016113 0.053711 0.12085 0.193359 0.225586
Bin number from 5 4 3 2 1 0
right
Probability 0.193359 0.12085 0.053711 0.016113 0.00293 0.000244
KEY POINTS

The normal distribution, a special continuous distribution, is extremely important in

statistics because many random variables that occur in real applications have normal
distributions (or approximately normal distributions).
The normal distribution, characterized by its mean and its standard deviation ., has
a graph that is bell-shaped; it is also symmetric about the mean so that in
consequence, the mean is the median, and is also the mode (since the curve is highest
at the mean).
The normal curve satisfies the Empirical Rule: (a) approximately 68% of the area
under the normal curve is within one standard deviation from the mean;
approximately 95% of the area under the normal curve is within two standard
deviations from the mean; and nearly everything (approximately 99.7% of the area
under the normal curve) is within three standard deviations from the mean.
REFERENCES

Albert, J. R. G. (2008).Basic Statistics for the Tertiary Level (ed. Roberto Padua, Welfredo
Patungan, Nelia Marquez), published by Rex Bookstore.
De Veau, R. D., Velleman, P. F., and Bock, D. E. (2006). Intro Stats. Pearson Ed. Inc.
Workbooks in Statistics 1: 11th Edition, Institute of Statistics, UP Los Banos, College Laguna
4031

http://www.amsi.org.au/ESA_Senior_Years/PDF/ExpoNormDist4f.pdf

http://cnx.org/contents/228deca0-5532-488c-8422-5878022132d6@9/Normal-Distribution:-
Teacher's

https://www.opened.com/video/normal-distribution-explained-simply-part-1/43505

https://www.opened.com/video/introduction-to-the-normal-distribution/109297

https://www.youtube.com/watch?v=yTGEMoaWDCQ

http://people.wku.edu/david.neal/183/Unit1/Normal.pdf

http://www.learner.org/courses/againstallodds/unitpages/unit07.html
 ASSESSMENT

1. The data below and the accompanying histogram give the weights, to the nearest
hundredth of a gram, of a sample of 100 coins (each with a value of P10). The mean
weight is 8.69 grams and the standard deviation is approximately 0.055 gram.

8
6
Density

4
2
0

8.55 8.6 8.65 8.7 8.75 8.8

weightcoin

a. Compare the mean and median.

b. What percentage of the data are within one standard deviation of the mean? within
two standard deviations? within three standard deviations?
c. Suppose you were to randomly select a coin from this collection. What is the chance
that its weight would be within one standard from the mean? two standard deviations?
three standard deviations?
 d. What percentage of the data is below the mean?
e. Suppose you were to randomly select a coin from this collection. What is the chance
that its weight would be below the mean?
Soln:
a. Very close, the median is 8.7 grams.
b. 67%, 95%, 100%.
c. According to the empirical rule, the chances are 68%, 95% and 99.7%.
d. 48%
e. 50%

2. Fifty students were asked to run a 100-meter dash. The data below represents the time it
took to finish the dash, and the histogram. The mean time for the 50 students is 15.8
seconds, and the standard deviation s is approximately 3.29 seconds.
16 14 14 16 21 14 17 15 16 21
14 10 9 20 12 12 19 11 15 14
18 18 13 18 23 8 20 13 16 23
16 17 15 18 17 16 13 15 18 19
12 12 15 17 14 16 17 16 16 21

a. Calculate the median, and compare it with mean.

b. What percentage of the data are within one standard deviation of the mean? within
two standard deviations? within three standard deviations?
c. Suppose you were to randomly select a student from these fifty students. What is the
chance that the time it took for him/her to run the 100-meter dash would be within
one standard from the mean? two standard deviations? three standard deviations?

Soln:
a. Very close, the median is 8.7 grams.
b. 70%, 92%, 100%.
c. According to the empirical rule, the chances are 68%, 95% and 99.7%.

3. Toss a fair coin twice and let X be the number of heads obtained. Generate the histogram
for the distribution. Consider tossing the fair coin three times, five times, ten times,
fifteen times; generate the histogram for the number of heads also for these cases.
As the number of tosses increases, what curve can be used to approximate the histogram?

(a) Probability Distribution for number of heads in two tosses of a fair coin

X 0 1 2
P(X=x) 0.25 0.5 0.25
 0.6

0.5

0.4

0.3

0.2

0.1

0
1 2 3

(b) Probability Distribution for number of heads in three tosses of a fair coin

x 0 1 2 3
P(X=x) 0.125 0.375 0.375 0.125

0.4

0.35

0.3

0.25

0.2

0.15

0.1

0.05

0
0 1 2 3

(c) Probability Distribution for number of heads in five tosses of a fair coin

x 0 1 2 3 4 5
P(X=x) 0.03125 0.15625 0.3125 0.3125 0.15625 0.03125
 0.35

0.3

0.25

0.2

0.15

0.1

0.05

0
0 1 2 3 4 5

(d) Probability Distribution for number of heads in ten tosses of a fair coin

x 0 1 2 3 4 5
P(X=x) 0.000976563 0.009766 0.043945 0.117188 0.205078 0.246094

x 6 7 8 9 10
P(X=x) 0.205078 0.117188 0.043945 0.009766 0.000977

0.3

0.25

0.2

0.15

0.1

0.05

0
0 1 2 3 4 5 6 7 8 9 10

(e) Probability Distribution for number of heads in fifteen tosses of a fair coin
 x 0 1 2 3 4 5
P(X=x) 3.05176E-05 0.000458 0.003204 0.013885 0.041656 0.091644

x 6 7 8 9 10
P(X=x) 0.15274 0.196381 0.196381 0.15274 0.091644

x 11 12 13 14 15
P(X=x) 0.041656 0.013885 0.003204 0.000458 3.05E-05

0.3

0.25

0.2

0.15

0.1

0.05

0
0 1 2 3 4 5 6 7 8 9 10

(f) As the number n of tosses increases, the normal curve can be used to approximate the
histogram of the number of heads in n tosses of a fair coin (a binomial probability
distribution)?

4. Suppose that the weights of Filipino grade 11 students are normally distributed with a
mean of 52 kilograms and a standard deviation of 1 kilogram. Explain what this means in
terms of the properties of a normal distribution
Solution. Let be all grade 11 Filipino students, and let X denote their weight (in kg).
Then X ~ N(52, 1) kg. That is,
(i) the average weight, the most likely weight, and the median weight are all 52 kg.
(ii) weights of the grade 11 Filipino students as a whole are symmetric about the weight
52 kg.

(iii) Around 68% of weights are from 51 to 53 kg ( ); around 95% of weights are
from 50 to 54 kg ( 2); and around 99.7% of weights are from 49 to 55 kg ( 3 ).
(iv) A histogram of weights of Filipino grade 11 students creates a Bell-Shaped Curve
with the percentages of high and low weights dropping off exponentially
 The following data pertaining to the points scored of the high school basketball team in 28
games;

66 75 41 57 54 82 67
42 60 37 49 87 101 78
60 66 48 43 42 61 64
67 37 51 63 68 77 13

The average of the points scored is 59.14286, while the standard deviation is 17.856 and the
histogram is provided below.
.03
.02
Density

.01
0

0 20 40 60 80 100
basketball

Can we approximate the distribution of scores with a normal curve, and what does this mean
in terms of the properties of a normal curve?

Solution. Yes, we can approximate the basketball scores distribution with a normal curve.
Let X denote their basketball score. Then X ~ N(59.14, 17.962). That is,
(i) the average score, the most likely score, and the median score are all 59 points.
(ii) scores in the basketball games as a whole are symmetric about the 59 points

(iii) Around 68% of scores are from 41 to 77 points ( ); around 95% of scores are from
23 to 95 points ( 2); and around 99.7% of weights are from 6 to 113 points ( 3 ).
(iv) A histogram of the basketball scores creates an approximate Bell-Shaped Curve with
the percentages of high and low scores dropping off exponentially