Biostatistics

SXEX1101 @ SIX1003
Prof. Dr. Normaniza Osman
03-79674185
[email protected]
 Introduction

Terminology

Statistics
Biostatistics???
 Raw data

http://
patentimages.storage.googleapis.com/ http://bluebuttontoolkit.healthit.gov/challenge/
EP0032134A2/imgb0045.png

https://encrypted-tbn3.gstatic.com/images?q=tbn:ANd9GcQLMrJqIJ6yzkupb4t-fFNTCpfbQRlXSjb1c8YN4M-1XkafcY4g2A
 Statistics
The discipline of statistics provides methods for
organizing and summarizing data
and for drawing inferences from the
data.
 Main division of Statistics

Descriptive statistics
• involve organizing, summarizing, and presenting
data in a clear format

• e.g. mean, median, variance, standard

deviation, and various types of plots for data
presentation
Main division of Statistics

Inferential statistics
• process of making inferences from a sample
to a larger population.

• involve estimation and hypothesis testing

• e.g. standard normal, Chi-squared, t and f

distributions.
Statistics
Objectives

• To learn how to organize, summarize,

and describe the data.

• To learn how to reach decisions about a

large body of data by examine only a
small part of the data.

• Able to analyze and interpret the data.

 Social Public e.g. identifying sources of
diseases and ways to treat
Sciences Health them

Molecular
Importances Biology Mechanical
of STATISTICS e.g. analysis
of microarray
Engineering
data

e.g. describing
quantitatively how
Marketing Ecology individuals in animal and
plant populations are
distributed.
Data
• Observations (e.g. measurements, genders, survey
responses) that have been collected.

Variable vs Observation
• An individual value of a variable e.g.
body weight is a variable and the value
60kg is an observation.
 Population
• A group of unit in the statistical studies.
consists of an entire set of objects,
observations, or scores that have
something in common.

Term Parameter
• A numerical measurement describing some
characteristics of a population

• A fixed constant (value usually unknown);

interest to estimate this value
Census - The collection of data
from every member of the
population.
Sample
• smaller subset of the
underlying population Term Statistic
• A numerical
• provides the actual
measurement
numerical information
describing some
used in making inferences
characteristics of
about the population
a sample
• sampling method must
• An estimate of
suitable e.g. unbiased
the population
parameter
Statistic/Parameter Sample Population
Mean (mu)
Standard deviation s (sigma)
Variance
Correlation r (rho)
Coefficient
Proportion p (pi)
 Continuous

Quantitative

Descriptive Discrete

Statistics
Qualitative

Inferential
 Descriptive
• to describe the data

• The presentation and interpretation of collected data

through the use of graphic, tabular or
numerical devices
• Summarize or describe the important characteristics
of a known set of population data.
 ve
ipti
Qualitative Quantitative

s cr
Data that can

De
Data that can be be measured
categorized or counted

e.g. colours,
feelings, left e.g. height,
or right, weight
gender.
Quantitative
• Data result from infinitely many possible values
that correspond to some continuous scale that
covers a range of values without gaps,
Cont interruptions, or jumps.
inuo • e.g. Height, weight, age, cholesterol level
us
• Data result when the number of possible values is
either a finite number or a ‘countable’ number of
possible values.
Disc • e.g. Numbers of siblings or number of kittens that a
female cat can give birth to.
rete
 Inferential

use sample data to make inferences

(or generalizations) about a
population
Or
use a small sample to say something
about a larger population
Types of data
measurements
First step in statistical analysis is to identify
the type of data you have

Important – to determine what kinds of

statistics you will be able to use

There are four scales of data

measurements
 Nominal

Ratio
Data Ordinal
Scales

Interval
 Nominal Scale
Characterized by data that consist of names,
labels, or categories only

The data cannot be arranged in an ordering

(unordered
scheme

categories)
e.g. gender, success or failure, eye colour
 Ordinal Scale
Data may be arranged
in some order, but
differences between
data values either
cannot be determined
or are meaningless
Arrange in Rank order
e.g. incomes, level of agreement,
attitude towards biodiversity
Interval Scale
Like ordinal, with the additional
property that the difference between
any two data values is meaningful.

However, there is no natural zero

starting point (where none of the quantity is
present)

Have negative value.

e.g. temperature
 Ratio Scale
The interval level modified to include the
natural zero starting point (zero is the
lowest value)

Do not have negative value

e.g. weight, age, food price, math score

 Statistical Notations
X (sometimes Y) is used to denote certain variable
e.g. X1, X2, X3 ……. (Xi)…….  Xn

Indicate one group observation with n sample

e.g. observation 1, 2, 3 …. and observation ith

In general, the ith observation is written as Xi, and

and the last one is denoted by XN.
 Symbol
å = additive notation

x is the variable, usually used to represent the

individual data values

n represents the number of values in a sample

N represents the number of values in a population

 Symbol
• The Greek letter S is used to indicate the sum
of these N observations.
𝑁

∑ 𝑋 𝑖 = 𝑋 1 + 𝑋 2+ 𝑋 3 +…+ 𝑋 𝑁
𝑖=1

• The symbol ∑ 𝑋 is read “summation of X sub i

𝑖=1
𝑖

(or sigma X sub i) where i assumes all integral

values from 1 to N inclusive.”
 One variable
5

X
i 1
i X1  X 2  X 3  X 4  X 5

X
2 2 2 2 2 2
i X1  X 2  X 3  X 4  X 5
i 1

5
( X i ) ( X 1  X 2  X 3  X 4  X 5 )
2 2

i 1
 Two variables
n n n

 (X
i 1
i  Yi )  X i   Yi
i 1 i 1

5 5 n

 (X
i 1
i  Yi )  X i (1  2  3  4  5)   Yi (1  2  3  4  5)
i 1 i 1
n

XY
i 1
i i  X 1Y1  ....  X 5Y5  ...  X nYn
n

 k nk  no  cummulative
i 1
n n

 kX
i 1
i k  X i
i 1
 1 2 3 4 5 
Xi 1 3 5 7 9 25
Yi 2 4 6 8 10 30

(Xi + 3 7 11 15 19 55
Yi)

XiYi 2 12 30 56 90 190
Examples
1 2 3 4 5 
Xi 1 3 5 7 9 25
Yi 2 4 6 8 10 30
(Xi + Yi) 3 7 11 15 19 55
XiYi 2 12 30 56 90 190

Find:
5 5 5

 (X
i 1
i  Yi )  X i   Yi
i 1 i 1

XY
i 1
i i 
5

XY
i 1
i i  X 1Y1  X 2Y2  X 3Y3  X 4Y4  X 5Y5
 Exercise
• 1) If X1 = 3, X2 = 9, X3 = 11,
3

∑ 𝑋𝑖
Find 𝑖 =1

• 2) If X1=9, X2=6, X3=5, X4=8, X5=12, find:

5

∑ ( 𝑋 ¿¿𝑖−5)2¿
a) 𝑖=1

(
b)
 TWO-Way Table
i
Row Column
1 2 3 4 5
1
2
3
4
j 5

∑ 𝑋 2 𝑗 = 𝑋 21+ 𝑋 22 + 𝑋 23 + 𝑋 24 + 𝑋 25
𝑗=1
5 5 5
∑ 𝑋 𝑖 3=? ∑ ∑ 𝑋 𝑖𝑗 =?
𝑖=1 𝑖=1 𝑗=1
Examples
Student Course
1 2 3 4 5 Σ
1 55 56 73 75 77 336
2 27 69 69 77 71 313
3 72 47 46 78 37 280
4 49 53 32 80 90 304
5 73 55 59 97 80 364
Σ 276 280 279 407 355 1597

What is the mark of student “3”?

What is the total mark of student “3”?

What is the mark of students 2, 3 and 4?

What is the total mark of all students?

 ADDITIVE NOTATIONS
4 5

∑ ∑ 𝑋 𝑖𝑗 =( 𝑋 24 + 𝑋 25 ) + ( 𝑋 34 + 𝑋 35 ) +( 𝑋 44 + 𝑋 45 )
𝑖= 2 𝑗= 4

4 5

∑∑ 𝑋 𝑖𝑗2 =( 𝑋 242 + 𝑋 252 ) + ( 𝑋 342 + 𝑋 352 ) + ( 𝑋 442 + 𝑋 452 )

𝑖= 2 𝑗= 4

[ ]
4 5 2 4

∑ ∑ 𝑋 𝑖𝑗 =∑ ( 𝑋 𝑖 4 + 𝑋 𝑖5 ) 2

𝑖= 2 𝑗=4 𝑖=2

2 2 2
¿ ( 𝑋 24 + 𝑋 25 ) + ( 𝑋 34 + 𝑋 35 ) + ( 𝑋 44 + 𝑋 45 )
 i=1 2 3 4  Xi
j=1 7 9 2 10 28
2 12 11 8 6 37
3 4 8 1 5 18
4 13 6 8 11 38
5 17 5 7 13 42
 Xj 53 39 26 45 163
4 5

∑ ∑ 𝑋 𝑖𝑗 =163
4 4

𝑖 =1 𝑗=1 ∑ ∑ 𝑋 𝑖𝑗 =?
4 5 𝑖= 2 𝑗=3

∑ ∑ 𝑋 𝑖𝑗 =110
( )
4 4 2
𝑖= 2 𝑗=1
4 3 ∑ ∑ 𝑋 𝑖𝑗 =14 2+92 +16 2=?
∑ ∑ 𝑋 𝑖𝑗 =9+2+10+…+5=60 𝑖= 2 𝑗=3

∑ (∑ )
𝑖= 2 𝑗=1 4 4 2
4 3
𝑋 𝑖𝑗 2 =1002 +65 2 +146 2=?
∑ ∑ 𝑋 𝑖𝑗 =9 + 2 +10 +…+5 =496
2 2 2 2 2
𝑖= 2 𝑗=3
𝑖= 2 𝑗=1
 Examples

( )
4 5 2 4

∑ ∑ 𝑋 𝑖𝑗 =∑ ( 𝑋 𝑖4 + 𝑋𝑖 5 ) 2

𝑖= 2 𝑗=4 𝑖=2

¿ ( 6 +5 )2 + ( 8+7 )2+ ( 11+13 )2=922

Common mistake !!

( )
4 5 2 𝟒

∑ ∑ 𝑖𝑗 ∑ ( 𝟐 𝒋 𝟑 𝒋 𝟒 𝒋 )
𝑋 = 𝑿 + 𝑿 + 𝑿 𝟐

𝑖= 2 𝑗=4 𝒊=𝟐
𝟐 𝟐
¿ ( 𝑿 𝟐𝟒 + 𝑿 𝟑𝟒 + 𝑿 𝟒𝟒 ) + ( 𝑿 𝟐𝟓 + 𝑿 𝟑𝟓 + 𝑿 𝟒𝟓 )
𝟐 𝟐
¿ ( 𝟔+𝟖 +𝟏𝟏 ) + ( 𝟓+𝟕+ 𝟏𝟑 ) =𝟏𝟓𝟐𝟓