BIOSTATISTICS
INTRODUCTION:
Introduction:
Statistics is defined as the science & art
which deals with the study of systemic
collection, compilation, analysis &
interpretation of numerical data affected by
multiple causes.
Biostatistics:
Defined as the science & art which deals
with the study of systemic collection,
compilation, analysis & interpretation of
numerical data related to biological aspect or
events affected by multiple causes.
Medical statistics:
Science & art which deals with the study of
systemic collection, compilation, analysis &
interpretation of numerical data related to
biological aspect of human being affected by
multiple causes.
Vital statistics:
Science & art which deals with the study of
systemic collection, compilation, analysis &
interpretation of numerical data related to
vital events of human life i.e. birth, death,
sickness, marriage etc. affected by multiple
causes.
Biostatistics comprises:
1. Biostatistics proper –
Mortality
Morbidity
Fertility etc.
2. Elementary statistics –
Central tendency,
Frequency distribution,
Dispersion,
Normal distribution curve,
Graphic presentation,
Sampling etc.
Use/Functions Of Statistics:
1. Collection of information.
2. Simplification of huge & complex sets of
data.
3. Clarification of ideas.
4. Community health diagnosis.
5. Measurement of association between 2 or
more variables.
6. Designing experimental studies.
7. Analyzing data & drawing conclusion.
8. Providing information support for planning,
monitoring, evaluation & managing health
service programs.
Use of Biostatistics:
1. To describe the level of health in a
community.
2. To diagnose the community ill.
3. To identify the met & unmet health
needs.
4. To determine the priorities of health
program.
5. to determine the success or failure of
specific health program.
6. Demand fund for health works.
SCALES OF
MEASUREMENTS:
Definition –
All numerical values used to
differentiate between events have
logical relationship with each other , the
composite of this relationship is called a
scale.
Types – 4 Basic types;
1. Nominal
2. Ordinal
3. Interval
4. Ratio
1. Nominal scale;
Qualitative variable, can’t arranged in
definite order,
Code number is used to identify the
categories( Code # have no quantitative
significance),
e.g. – Marital status;
Single
Married
Widow
Divorced.
Sex; 1 = Male
2 = Female.
2. Ordinal scale;
Qualitative variable which can be ranked by
this scale.
e.g. Anemia –;
Mild – Moderate – Severe.
Social class;
Lower – Middle – Upper class etc.
3. Interval scale;
Equal difference between any pair of
numbers in the scale.
e.g. Income; 100 tk – 500 tk,
501 tk – 1000 tk. Etc.
4. Ratio scale;
Relation in size between two quantities e.g.
Dr. patient ratio.
DATA:
Definition:
Collected information through careful
observation of attributes or events or
measurements. “ Datum “ – Singular word.
Classifications:
A. On the basis of source –
1. Primary data; e.g. Subject,
Respondents, Group of individuals Object /
Agents, Animate / Inanimate , Places etc
2. Secondary data; e.g. Reports,
Hosp.records, Govt. Documents etc.
3. Tertiary data; Collected from Text book,
Literature review, Historical review etc.
B . On the basis of Nature –
1. Quantitative data ; Expressed as
numerically.
- It has 2 types;
a. Continuous – Has fraction value
e.g. Height, Weight etc.
b. Discrete – No fraction value e.g.
Pulse rate, Family members
etc.
2. Qualitative; Expressed in narrative form
e.g. Age, Sex, Rich/ Poor, Vaccinated/
Non – Vaccinated etc.
C. On the basis of processing:
1. Un – Grouped data; When data is
processed on individual basis. E.g. 1 year,
2years, 3 years etc.
2. Grouped data; When processed in group
wise e.g. Age 1 – 5 years, 6 – 10 years etc.
Objectives of Data:
1. To know the extent of problem.
2. It helps in future planning &
recommendation.
Data collection methods:
1. Documents review – Data taken from
others records, reports etc.
2. Observations – Systematically
selecting,
watching, recording behavior &
chacteristics of living being, objects
or phenomena.
Types: a. Participant observation –
The observer takes part in the
situation.
b. Non – Participant
observation – Observer watches the
situation openly or concealed , but
doesn’t participate.
e.g. Children washing hand before
3. Interview: i.e. Oral questioning.
4. Written questionnaires: Answered by the
respondents in written form.
e.g. Mail questionnaire, Structured
questionnaire.
5. Measurement / Experimental: e.g. Study on
weight & height of children by using weight &
height measuring scale.
6. Combined methods:
e. g. Interviewed + experimental,
Written questionnaire + experimental etc
.
Data collection/Research
instrument:
A document by which the data are
collected called research instrument.
e.g. Questionnaire, check list,
measuring tape etc.
Sources of data:
Census reports, Annual vital statistical
reports ( Number of still birth, Live
birth Death etc), Medical text book &
journal, Hospital records, Survey
reports etc.
Data presentation methods:
A. Table; 2 types –
1. On the basis of variable:
a. Univariate,
b. Bi - variates,
c. Multi – variates.
2. On the basis of nature:
a. Simple table,
b. Frequency table.
B. Diagram & Chart; 8 types –
1. Bar diagram :
- Simple bar diagram,
- Multiple bar diagram,
- Component bar diagram.
2. Histogram,
3. Frequency polygon,
4. Line diagram,
5. Pictogram,
6. Map diagram / Spot map,
7. Scatter / dot diagram
AND
8. Pie chart.
Multivariate
Bi -variate
Uni - variate
Year CBR CDR MMR
2000 9 6 4
2005 7 5 3
2010 6 4 2
Tables.
Frequency table: With Talley mark.
Height of Marking Frequency of
group in cm each group
160 – 161 IIII IIII 10
162 – 163 IIII IIII II 12
164 – 165 IIII IIII IIII 15
Criteria’s of a table:
1. Table should be clear, simple &
numbered i.e. Table 1, Table 2 etc.
2. A brief & self explanatory title must be
given & no table should be too large.
3. The heading of columns & row should
be clear & concise.
4. Data must be presented according to
importance.
5. Vertical arrangement of data must be
preferred than a horizontal one.
6. Foot note should be given where
necessary.
7. Sources should be mentioned when
the data is taken from secondary source.
Simple bar diagram:
Gap between 2 Bar. BAR DIAGRAM
1. Values are placed
by bar.
2. Bars are placed in
2 axis i.e. X & Y.
3. May be
horizontal/
vertical.
4. The width of the
bar should be
equal.
5 MULTIPLE
4.5 BAR
4 DIAGRAM
3.5 •2 or 3 simple bar
3 are drawn side by
2.5
CBR side without
CDR leaving any gap.
2 IMR
1.5 •Each of them is
1 differently colored.
0.5
•It’s drawn when 2
0 different factors are
2010 2011 2012
compared to the
same period of time.
140
COMPONENT
120
BAR DIAGRAM:
100
Variouscomponent
80 are placed in one bar
Amoebic
60 watery which constitute the
40 Bacillary total.
20 Component parts are
0 differently colored.
00
01
02
03
20
20
20
20
Each component
represent the
proportion to the
magnitude of the
particular item.
HISTOGRAM:
Special form of
bar diagram.
The bars are
adjacent to each
other on the x
axis. (No gap
between bars).
It represents
classes of
continuous data &
the class intervals
are plotted on X
axis.
HISTOGRAM CONT…..
The width of the bar represents the class
interval i.e.
Family size in Frequency
member.
1-3 08 frequency
3- 5 20
5- 7 35
7- 9 25
1- 3– 5– 7 - 9
CLASS INTERVAL
FREQUENCY
Click icon to add picture POLYGON:
Use for frequency
distribution of
continuous data.
Obtained by
joining the mid
point of class
intervals & the mid
points are
connected by
straight line.
LINE DIAGRAM:
20
It’s used to show
18 the trends of a
16 disease or event
14 over a period of
12 time i.e. rising,
10
falling or showing
fluctuation of event.
8
0
1995 1996 1997 1998
Pictogram:
Usually used for illiterate.
Small picture or symbols are used
to represent the data e.g. picture of
infant to indicate IMR.
Map Diagram:
Diseases or any other events
are expressed by spot / symbol on
the map.
Number of spot indicate the
frequency.
Usually used for epidemic
SCATTER /
Y-Values DOT
3.5 DIAGRAM:
3
It shows the
2.5
2 Y-Values
association or
1.5 correlation
1 between two
0.5 variables.
0
0.5 1 1.5 2 2.5 3
PIE CHART
percentage
Data is
presented in a
circle subdivided
1
. 1.2
Adult into parts to reveal
4 Ado- the components.
les- Circle has 360◦
cents
3.2 8.2 Children at the centre which
Infant is divided into
sectors in
proportion to yhe
components of the
data.
The sectors are
colored differently.
VARIABLES:
Definition:
Any characteristics of a person or object or
phenomena that can take different values for
different individuals , place or time.
OR
Variables are the characteristics or attributes
that may vary from person to person , place to
place or from time to time.
Classification:
A. On the basis of criteria;
1.Qualitative – It takes narrative
value e.g.
Sex, Religion etc.
- It has 2 types;
a. Nominal – Not maintained order e.g.
Woman is Married / Widow / Divorced.
b. Ordinal – Maintain order according to
priority e.g. Mild / Moderate / Severe.
2. Quantitative – It takes numerical values
e.g. Height, Age, weight etc.
- It is also 2 types;
a. Discrete – No fraction value, only absolute
value e.g. pulse rate, family size etc.
b. Continuous – Any value within a meaningful
range. May have fraction value e.g. Height,
Income etc. Continuous variable again 2 types –
Ratio variable – No existence or meaning of
‘0’e.g. Serum billirubin level, weight etc.
Interval variable – Meaningful ‘0’ e.g.
Temperature 0◦c, IQ test, Air pressure etc.
B. On the basis of Mutual
relationship / Function – 4 Types;
1. Independent Variable,
2. Dependent Variable,
3. Confounding Variable,
4. Intervening Variable.
1. Independent variable – Variable that
influences or regulates the other variables
called so. These variable describes or
measures the study factors.
2. Dependent variable – Variable that are
influenced by other variables. These may be
the results, effect or outcome.
Example;
Effects of blood sugar on Diabetes.
Independent Dependent
Variable Variable.
3. Confounding variable – Variable that
influence both the independent & dependent
variables & modify the effects (Strengthen /
weaken).
It influences the problem & the associated
factors .
In short – ‘ Hidden Variable’.
Example – Independent Dependent
Effect of blood sugar on Diabetes.
Dietary habit/ Genetic factor/ Pancreatic
disorder etc (Confounding variables) all the
factors influences the both dependent &
independent variables.
4. Intervening variable – Variable that
modifies (enhance / diminishes / blocks) the
influence of independent variable on the
dependent variable.
Example;
Pregnancy leads to Anemia.
Independent Dependent Variable.
Other factors like pregnancy order, Age of the woman, food
during pregnancy etc may modify the effect of pregnancy
are confounding variable, But Iron tab. Taking / not directly
affect on Hb in pregnancy. These are called Intervening
variable.
Difference between-
Intervening & Confounding variable
1. Modify independent 1. Modify both
variable & can be measured. Dependent &
Independent
variable &
may/ may not
measured .
