23

NORMAL
DISTRIBUTION

KSU COLLEGE OF MEDICINE

2019 - 2020

ACKNOWLEDGMENTS
DONE BY
AFNAN ALMUSTAFA
ASEEL BADUKHON

Special thanks to SARAH ALANZI & 436 TEAM

 Original, refrased. Our notes. Doctors notes. Golden notes.

NORMAL DISTRIBUTION PAGE 02

TABLE OF CONTENTS

NORMAL
DISTRIBUTIO
N

LECTURE OBJECTIVES

By the end of this lecture, I am able to understand:

Able to understand the concept of Normal distribution.
Able to calculate the z-score for quantitative variable.
Able to apply the concept in the interpretation of a clinical data.
 NORMAL DISTRIBUTION PAGE 03

OVERVIEW
Introduction

- Problem: Assume that among diabetics the fasting blood level of glucose is
approximately normally distributed with a mean of 105 mg per 100ml and an SD of
9 mg per 100 ml. What proportion of diabetics having fasting blood glucose levels
between 90 and 125 mg per 100 ml ?
- The Normal or Gaussian distribution is the most important continuous probability
distribution in statistics.
- The term “Gaussian” refers to ‘Carl Freidrich Gauss’ who develop this distribution.
- The word ‘normal’ here does not mean ‘ordinary’ or ‘common’ nor does it mean
‘disease-free’.
- It simply means that the distribution confirms to a certain formula and shape.

Gaussian Distribution

- Many biologic variables follow this pattern:

Hemoglobin, Cholesterol, Serum Electrolytes, Blood pressures, age, weight, height.
- One can use this information to define what is normal and what is extreme.
- In clinical medicine 95% or 2 Standard deviations around the mean is normal.
- Clinically, 5% of “normal” individuals are labeled as extreme/abnormal.
- We just accept this and move on.
NORMAL DISTRIBUTION PAGE 04

Normal Distribution

Symmetrical about mean, μ. (mean will be

in the center)

Mean, median, and mode are equal.

Characteristics

Total area under the curve above the

x-axis is one square unit

1 standard deviation on both sides of

the mean includes approximately
68% of the total area.

3 standard
2 standard
deviations includes
deviations includes
approximately 99%
approximately 95%

● Uses
○ It’s application goes beyond describing distributions.
○ It is used by researchers.
○ The major use of normal distribution is the role it plays in statistical
inference. Most of the statically theory based on this concept
○ It helps managers to make decisions.

● What’s so Great about the Normal Distribution?

○ If you know two things ( Mean & Standard deviation ) you know everything
about the distribution. You know the probability of any value arising.
NORMAL DISTRIBUTION PAGE 05

EXAMPLE
Standardised Scores

- My diastolic blood pressure is 100. So what ?

- Normal is 90 (for my age and sex). Mine is high, But how much high?
- Express it in standardised score. How many SDs above the mean is that?

Mean = 90, SD = 4 (my age and sex)

This is a standardised score, or z-score

Look z tables (or computer), See how often this high (or higher) score occur.
NORMAL DISTRIBUTION PAGE 06

Measures of Position

● z Score (or standard score)

○ The number of standard deviations that a given value x is above or below
the mean. meaning that my BP is 100 and the mean is 90, the SD is 4, how many SDs do i need
to reach my BP from the mean? 10 is the difference, 10 divided by 4 = 2.5 , so i need 2.5 times
○ //(90+4)=1 SD,, (90+8)= 2 SD,, (90+10)=2.5 SD
○ The Z score makes it possible, under some circumstances, to compare
scores that originally had different units of measurement.

● Suppose you scored a 60 on a numerical test and a 30 on a verbal test. On

which test did you perform better?
○ First, we need to know how other people did on the same tests.
■ Suppose that the mean score on the numerical test was 50 and
the mean score on the verbal test was 20.
■ You scored 10 points above the mean on each test. Can you
conclude that you did equally well on both tests?
■ You do not know, because you do not know if 10 points on the
numerical test is the same as 10 points on the verbal test.

● Suppose you scored a 60 on a numerical test and a 30 on a verbal test. On

which test did you perform better?
○ Suppose also that the standard deviation on the numerical test was 15
and the standard deviation on the verbal test was 5.
■ Now can you determine on which test you did better? Verbal is better
because you almost 2 standard deviations
Standard Deviation is 15 that the reason .

Close to the average

Why flat?

Away from the average

● To find out how many standard deviations away from the mean a particular score is,
use the Z formula: important to remember that the standard deviation is the unit of measurement

Population: Sample:
NORMAL DISTRIBUTION PAGE 07

Cont’

Properties of Z-score:
● Allows you to describe a particular score in terms of where it fits into the overall
group of scores.
○ Whether it is above or below the average and how much it is above or below
the average.
● A standard score that states the position of a score in relation to the mean of the
distribution, using the standard deviation as the unit of measurement.
○ The number of standard deviations a score is above or below a mean.

Interpreting Z Scores

Unusual values Ordinary values Unusual values

-3 -2 -1 0 1 2 3
Z
NORMAL DISTRIBUTION PAGE 08

INTERPRETATION
The Standard Normal Table

● Using the standard normal table, you can find the area under the curve that
corresponds with certain scores.
● The area under the curve is proportional to the frequency of scores.
● The area under the curve gives the probability of that score occurring.

The Tables: (Please read the heading of the table before you find the values; notice the differences)

Reading the Z table: (Pay attention to the color code)

Finding the proportion of Finding the proportion of Finding the proportion of

observations between the observations between a
observations above a
mean and a score when score and the mean when
Z=1.80 score when Z=1.80 Z=-2.10

Finding the proportion of

observations below a
score when Z=-2.10
NORMAL DISTRIBUTION PAGE 09

INTERPRETATION
Z scores & the Normal Distribution

Can answer a wide variety of questions about

any normal distribution with a known mean
and standard deviation.

Will address how to solve two main types of

normal curve problems:
Finding a proportion given a
score.

Finding a score given a

proportion.

EXAMPLE
Assuming the normal heart rate (H.R) in normal healthy individuals is normally
distributed with:

Mean = 70 and Standard Deviation =10 beats/min

Then: (next slide for the graphs and dr’s notes)

2) What area of 3) What area of 5) What area of

4) What area of
1) What area the curve is the curve is the curve is
the curve is
under the curve above 90 between below 40 beats
above 100
is above 80 beats/min? 50-90 per min or
beats/min?
beats/min? beats/min? above 100 beats
Ans: 0.025 per min?
Ans: 0.0015
Ans: 0.16 (16%) (2.5%) Ans: 0.95 Ans: 0.0015 for
(0.15%)
(95%) each tail or 0.3%
NORMAL DISTRIBUTION PAGE 10

EXAMPLE

1-How much percentage of patients are their heart rate is 80 b\m and above? Total are 100%.//1 standard
deviation cover 68%, half of 68%(we did this step because the SD covers 10 above and below 70 and in this
case we only want above so we take the half) ,34% Subtract from 50 = 16%(we did this step because 34% are
above the mean (70) with a range of 10(70+10=80) but what about those above it?(and our case we only want
above 80) we subtract from 50 because this 50% represents those above the mean and 34% are within (70--80)
and we want above 80 which leaves us with the other part of the 50% (50-34=16%))..who have a beat above 80
beats/min
2-=2.5% Because we are asking 2 standard deviations. 2 standard deviations (90) , mean is 70 , Probability .025
or 2.5%
3-0.95 or 95% because of 2SD.
4-0.0015.(3 standard deviations). so small area in the extreme right side.
5-Extreme 3 standard deviations - extreme +3 standard deviations on upper side =because it crosses 3
standard Deviations

Assume that among diabetics

the fasting blood level of
glucose is approximately
normally distributed with a mean
of 105 mg per 100ml and an SD
of 9 mg per 100 ml. What
proportion of diabetics having
fasting blood glucose levels
between 90 and 125 mg per 100
ml ?
We have to get 2 z score, it is not a whole
number to apply the rule it is in fraction.
90 - 105 \ 9 = -1.6 this is one z score
125 - 105 \ 9 = 2.2 this is the second z score. then
you continue solving as in the picture below.