Normal Distribution

Dr. Tarek Tawfik

12/07/21 1
 Normal Or Gaussian Distribution
 One particular distribution, the term
normal is not meant to signify usual or
common.
 In fact, it might seem like a very artificial
construct that is anything but normal.
 However, it does play a central role in
statistical analysis and is the basis for most
of the procedures that follow.
 The normal distribution is the most
:important in statistics for three reasons
I. Although most distribution are not exactly
normal, most variables tend to have
approximately normal distributions.
II. Many inferential statistics assume that the
populations are distributed normally.
III. The normal curve is a probability distribution
and is used to answer questions about the
likelihood of getting various outcomes when
sampling from a population (Hypothesis
testing).
Characteristics Of the Normal Curve

Gauss
DM 20
Normal Distribution:
Many human traits, such as intelligence, personality, and attitudes,
also, the weight and height, are distributed among the populations in
.a fairly normal way
 Characteristics
.Smooth bell-shaped curve and symmetrical -1
The mean, mode, and median are coinciding and -2
. have the same value
”Uni-modal “entails only one mode -3
:The areas under the curve is divided as -4
Between ±1 S.D from the mean of normal - -
. distribution lies 68.3 % of the area under the curve
.% Between ± 2 S.D lies 95.4 - -
.% Between ± 3 S.D lies 99.7 - -
.I has two tails that extend to infinity -5 -
It describe the distribution of the numeric -6 -
. (Interval/Ratio) type of variables
Characteristics
Normal curve: one sample of 400 persons having a mean of
35 years and S.D of 13, how many of them lies between the
mean and one S.D.
273 (400X68.3/100) subjects (68.3 % of the 400 ) will have
.ages somewhere between (35-13) and (35+13)

.Average age and spread of three different populations

Group Mean age S.D Age range of middle

68.3 % of cases )years( )years(
(years)
48 -22 13 35 1 273
42 -28 7 35 2
56 -15 20 35 3

All three distributions have the same mean for age

but they are different in terms of the spread of ages
.around the mean
 .The Standard Normal Distribution
The process of stating the spread of cases in terms of
the number of S.D from the mean is called
Standardizing the distribution, and produces the
Standard Normal Distribution.
The standard normal distribution has a mean of zero
and a standard deviation of 1 .
This process allow us to measure all normal
distributions in terms of common units (standard
deviation units) regardless of the units in which the
variable is initially measured ‘ expressing the price of
various goods from different countries in terms of a
common currency’
Converting U.S $, Euro, and Pounds into GOLD.
 Standard Normal Curve

Mean = 0

SD 1 - S.D 1 +

12/07/21 10
Applications of the Normal Curve

Normal Curve

As an aid to As a tool for inferential

data description statistics
Location of individual values In the sampling distribution
 Data Description

Areas under the standard normal curve

Beyond both points Beyond one point
S.D from the mean Between both points )two tails( )one tail(

±1 0.683 0.317 0.158

±2 0.954 0.046 5
±3 0.997 0.003 0.023
0
0.001
Instead of expressing the area under the curve as a %, it is expressed as 5
.proportions: 68.3 %= 0.683
The values in the first two columns will always sum to 1 (0.683+0.317=1).
This because the two areas are equal the total area under the curve,
which is 100 % = 1
 The Normal curve and Probabilities

 The normal curve is a very specifically

defined polygon, that allow us to interpret the
proportions in the previous table as
probabilities.
 It is simply the chance that any given case
will have a certain value, or fall within a
certain range of values.
 Assume we pick up a person from the 400 subjects
at random with a mean age of 35 and S.D of 13
years, can we guess this person’s age?
 The common sense method: from the table of
areas under the curve we can conclude that the
probability that this person’s age is somewhere
between 22 (35-13) and 48 (35+13) years is 0.683
or about 68 %.
 Also, the probabilities of having an age of less than
22 years or over 48 years (one tailed ) is 0.1585
(about 16 %).
 For precise location and definition of values under
the normal curve we use the Z-Score.
 The Z-Scores
Z  i   / 
Z  i   / s
where :
i  the actual value measured in original units.
  is the mean of the population.
  Standard deviation of the population.
  is the mean of the sample
s  Standard deviation of the sample.

,A Z-score of +1 indicates one standard deviation above the mean

while a Z-score of – 1.5 S.D indicates 1.5 standard deviations
.below the mean
Z-score
A population of 400 with a mean age of 35 and S.D of
13, a member of them tells me that he is 61 years
old, how would you precisely calculate the
proportion in this population that are at or above
.this age (61 years)

Z= 61-35/13 = 2
it means that 61 is two standard deviations above the
average, by referring to the table of areas under
the curve we found that the proportion of people
.% that are 61 years or more is only 0.023 or 2.3
 .Areas under the standard normal curve
Z Area under curve between Beyond both points Beyond one point
both points (Around the mean) )two tails( )one tail(

±0.1 0.080 0.920 0.4600

±0.2 0.159 0.841 0.4205
±0.3 0.236 0.764 0.3820
±0.4 0.311 0.689 0.3445
±0.5 0.383 0.617 0.3085
±0.6 0.451 0.549 0.2745
±0.7 0.516 0.484 0.2420
±0.8 0.576 0.424 0.2120
±0.9 0.632 0.368 0.1840
±1 0.683 0.317 0.1585
±1.1 0.729 0.271 0.1355
±1.2 0.770 0.230 0.1150
±1.3 0.806 0.194 0.0970
±1.4 0.838 0.162 0.0810
±1.5 0.866 0.134 0.0670
±1.6 0.890 0.110 0.0550
±1.645 0.900 0.100 0.0500
±1.7 0.911 0.089 0.0445
±1.8 0.928 0.072 0.0360
±1.9 0.943 0.057 0.0290
1.96 0.950 0.050 0.0250
±2 0.954 0.046 0.0230
±2.1 0.964 0.036 0.0180
±2.2 0.972 0.028 0.0140
±2.3 0.979 0.021 0.0105
±2.4 0.984 0.010 0.0100
±3 0.996 0.004 0.0020
 The area between the mean and a point on
.the distribution
Assume that we have exam grades out of 100 for a sample of
.100 students with a mean of 60 and S.D of 10
?How many students are between the mean of 60 and 65

.Calculate the Z-score: Z = 65-60/10= 0.5 -1

Find the proportion that lies between this point (0.5) and the mean using -2
. the areas under the curve
of all cases will have a grade of 5 marks above or below )% 38.3( 0.383
.the mean
Since we are interested in only those students that are 5 marks above the -3
mean, we divide 0.383 / 2 = 0.1925 (19 % of the students received
. grades between 60 and 65 )
 .The area beyond a point on the distribution

How to find the percentage of cases that fall beyond a certain point
?on the distribution

What is the percentage of students who did exceptionally well in

?the exam and having the score of over 65
Z = 65-60/10 = 0.5 refer to area under the curve for area beyond
.the point defined by a z-score of 0.5 (only one tail)

?How many students have failed the exam

 Areas under the normal curve
Z Area under curve between Beyond both points Beyond one point
both points (Around the mean) )two tails( )one tail(
±0.1 0.080 0.920 0.4600
±0.2 0.159 0.841 0.4205
±0.3 0.236 0.764 0.3820
±0.4 0.311 0.689 0.3445
±0.5 0.383 0.617 0.3085
±0.6 0.451 0.549 0.2745
±0.7 0.516 0.484 0.2420
±0.8 0.576 0.424 0.2120
±0.9 0.632 0.368 0.1840
±1 0.683 0.317 0.1585
±1.1 0.729 0.271 0.1355
±1.2 0.770 0.230 0.1150
±1.3 0.806 0.194 0.0970
±1.4 0.838 0.162 0.0810
±1.5 0.866 0.134 0.0670
±1.6 0.890 0.110 0.0550
±1.645 0.900 0.100 0.0500
±1.7 0.911 0.089 0.0445
±1.8 0.928 0.072 0.0360
±1.9 0.943 0.057 0.0290
1.96 0.950 0.050 0.0250
±2 0.954 0.046 0.0230
±2.1 0.964 0.036 0.0180
±2.2 0.972 0.028 0.0140
±2.3 0.979 0.021 0.0105
±2.4 0.984 0.010 0.0100
±3 0.996 0.004 0.0020
Calculating Values from z-scores

What range of scores identify the middle 50 % of our

?students
.look at the areas under the curve , 0.5 = % 50
The closest value to 0.5 is 0.516 which is associated with a z-
score of ±0.7,then convert the score to grades using the
:formula
.Xi = Mean± Z (standard deviation)
Xi = 60 +0.7(10)=67 and Xi = 60-0.7(10)=53
Therefore the middle 50 % of students scored between 53 and 67 in the
exam. This also means that 25 % of students are below 53 and 25 %
. above 67
Thank You