BIOSTATISTICS -

DR. PRATHA AKOLU

MDS PART – I
CONTENTS
 Introduction
 Collection of Data
 Sources & Presentation of Data
 Measures of Central Tendency
 Measures of Dispersion
 Normal curve
 Tests of significance
 Sampling
 Reference
INTRODUCTION
 It is just a tool for analysis and interpretation.

 Statistics is the science of compiling, classifying and tabulating numerical

data and expressing the results in mathematical or graphical form.

 Biostatistics is that branch of statistics concerned with mathematical facts

and data related to biological events.

 It is the term used when the tools of statistics are applied to data that is

derived from biological sciences such as medicine or dentistry.

COLLECTION OF DATA
 DATA – Comprises of details of population size, geographic distribution, ethnic
groups, socio-economic factors & their trends over time.
 Depending on the nature of the variable,

Data

Qualitative Quantitative

Collected
Collected on
through
the basis of Eg :- gender Eg :- indices
measureme
qualities
nt
 when the variable under
observation takes only
fixed values like whole
numbers
Discrete

Eg :- DMF teeth

Quantitative

When variables can take

any value in range

Continuous

Eg :- Arch length
METHOD OF DATA COLLECTION

Primary Source

• Data is obtained by the investigator himself (1st hand

information)
• Eg :- Personal Interviews

Secondary Source

• Already recorded data is utilized

• Eg :- OPD records
PRIMARY DATA

DIRECT
ORAL HEALTH QUESTIONNAIR
PERSONAL
EXAMINATION E METHOD
INTERVIEWS

A list of question
Face to face contact
pertaining to the
with person from Information on oral
survey is prepared,
whom the health, conducted by
informants are asked
information is dentists
to supply the
obtained
information
PRESENTATION OF DATA
 Data collected and compiled from experimental work, surveys,
registers/ records.
 Such data is not helpful for easy understanding
 Thus, data needs to be sorted and classified.
 The objective being to make the data –
 Simple
 Concise
 Meaningful
 Interesting
 Helpful in further analysis
 DATA PRESENTATION

TABULATION
CHARTS AND LINE STATISTICAL
DIAGRAMS DIAGRAMS MAPS

FREQUENCY
SIMPLE DISTRIBUTION
TABLES TABLES

HISTOGRAMS PIE CHARTS PICTOGRAM

BAR CHARTS

FREQUENCY FREQUENCY
COMPONENT POLYGON CURVE
MULTIPLE BAR
SIMPLE BAR CHART BAR CHART
CHART
TABULATION
 Tables are simple devices used for presentation of statistical
data.
 Data must be presented according to size or importance,
chronologically or alphabetically.
 Tables should be self explanatory, codes abbreviation or symbols
should be explained in detail in footnote.
 Each row and each column should be labelled concisely and
clearly.
 Every table should contain a title as to what is depicted in the
table. The title is commonly separated from the body of the
table by lines or spaces.
 Master table
 Contains all the data from a study.

Sr no. Name Age Sex Teeth Caries

present

 Simple Table
 Presents only one characteristic of data

Age Height

 Frequency Polygon
 Depicts frequency of occurrence

Marks Range Tally Frequency

0-25
25-50
CHARTS AND DIAGRAMS
 Purpose of diagrammatic and Graphic presentation:
 They are self explanatory
 They are simple and consistent.
 They give a bird eye view of the entire data.
 They are attractive to the eye and lasting impression.
 They have greater memorizing effect.
 They facilitate the comparison of data relating to different
periods of time.
 MULTIPLE BAR CHART

BAR CHART  Represents more than one variable

 Represents data along the

bar
 Can be
 Vertical
 Horizontal Category 1

COMPONENT BAR CHART

SIMPLE BAR CHART  Individual bars are divided into two or
 Represents only one more parts
28

variable
21

19

Category 1 Category 2
PIE CHART
 Represents total frequency
 Total angle – 360 degree
 Divided into different sectors corresponding to frequencies of
variables
Caries

0-10 years 10-20 yrs 20-30 yrs 30-40 years

LINE DIAGRAM
 Study values of variables over time i.e hours, days, weeks etc

0
Category 1 Category 2 Category 3 Category 4

Series 1 Series 2 Column1

HISTOGRAM
 Pictorial diagram of frequency of distribution
FREQUENCY POLYGON
 It is also a pictorial diagram of frequency distribution.
 Mid point on histogram bars are connected by line.
CARTOGRAM/SPOT
MAP/SHADED MAP
 These maps are used to show geographical distribution of frequencies of a
characteristic.
 If shades are used it is called shaded map.
SCATTER DIAGRAM
 It is diagram which shows the relationship between two variables.
 Relationship between two variables.

4.5

3.5

2.5

1.5

0.5

0
Category 1 Category 2 Category 3 Category 4
MEASURES OF CENTRAL
TENDENCY
 The main objective of measures of central tendency is to condense the
entire mass of data and facilitate comparison.
 The most common measures of central tendency that are used in dental
sciences are –
1. Arithmetic mean – mathematical estimate.
2. Median – positional estimate.
3. Mode – based on frequency.
ARITHMETIC MEAN
 It is the simplest measure of central tendency.
 It is obtained by adding the individual observations and then divided by the
total number of observations.
 Advantages – 1. Easy to calculate and understand.

2. It is most useful of all the averages.

3. Based on all items of the given data
MEDIAN
 It is the middle most value in a distribution arranged either in an ascending
or descending order.
 When the number of observations is odd – M = the middle value.
 When the number of observations is even then M = The sum of n/2 th and
the next value, divided by two.
 Advantages
1. Simple to understand and easy to compute.
2. Can be calculated for open-ended classes.
3. Not affected by extreme values.
4. Can be graphically determined.
MODE
 The mode is the observation that has been repeated the most number of
times. In a grouped data it is the class interval having the highest
frequency.
 A set of observations can be described as either unimodal, bimodal or
multimodal depending on the number of modes present in that distribution.
 Advantages
1. Easy to compute.
2. Is an average used in day-to-day life.
3. No need of arranging data.
MEASURES OF DISPERSION
 The different measures of dispersion are:
 A] Range.
 B] Mean deviation.
 C] Standard deviation.
 Objectives:

- determine reliable average.

- Serve as a basis for the control of variability.
- Enable other statistical analysis.
- Compare with other sets of data.
RANGE
 Range is the difference between the highest and the lowest values found in
the set of observations.
 Uses of range:

- quality control.
- Variation in status.
- Weather reports.
MEAN DEVIATION
 It is the mean of the deviations of all the different observations from the
calculated mean.
STANDARD DEVIATION
 It was introduced by Karl Pearson. It is a useful measure of variability.
 It overcomes the zero sum concept of the mean.
 Calculating the standard deviation:
 SD = √∑[x – x bar]2/ n.
 Where ,
 SD = Standard deviation,
 x bar is the calculated mean,
 x = the different observations,
 n = the number of observations.
THE NORMAL DISTRIBUTION –
NORMAL CURVE
 The most important probability density function used in
statistical analysis is the normal or gaussian distribution. The
features of this distribution are:
 1. Shape: The normal distribution is represented by a smooth
bell shaped curve that is symmetric about the population
mean . The exact shape of a particular normal curve
depends on the,
 Location of its center, which is the mean .
 The degree of spread of the other observations around
the center.
 2. Area under the curve: The relative frequencies of the values around the
 in a normal distribution are (as given in fig)
 Fifty percent of the observations lie above the mean and the
remaining 50% lie below the mean.
 Approximately 68% (68.26% to be precise) lie within one standard
deviation of the mean.
 Approximately 95% lie within two standard deviations of the mean.
 Approximately 99% (99.02% to be precise) lie within three
standard deviations of the mean.
TEST OF SIGNIFICANCE
 When different samples are drawn from the same population, the estimates
might differ
 This difference in the estimates – sampling variability
 Test of significance – deals with technique to know how far the estimates
differ due to sampling variability.
1. Standard of Error of mean = S.D/√n
2. Standard of Error of proportion = √pq/n
3. Standard of Error of difference between 2 means
4. Standard of Error of difference between 2 proportions
PARAMETRIC AND NON
PARAMETRIC TESTS
Parametric tests Non Parametric tests
• Based on specific • Not based on any
distribution such as particular parameter such
Gaussian as mean
• Student’s t- test • Do not require that the
• Z test means follow a particular
distribution such as
Gaussian.
• Chi-square test
CHI SQUARE TEST
 Developed by Karl Pearson
 Data is measured in terms of attributes or qualities and is intended to test
whether the difference in the distribution of attributes in different groups is
due to sampling variation or not – Chi Square Test is applied.
 Used

1. To test significance of difference between two proportions

2. Can be used when there are more than two groups to be compared.
STEPS
1. Test the null hypothesis
 Eg: To test if there is an association between oral hygiene instructions received
and the occurrence of new cavities
2. The x2 – statistic is calculated

 O – observed frequency
 E – Expected frequency

3. Applying the test



4. Finding degree of freedom – d.f (depends on no. of columns and rows)

 d.f = (column – 1 ) (row – 1)

5. Probability table
 Null hypothesis is verified
STUDENT’S T TEST
 Designed by W.S. Gossett
 Applied to find the difference between two means
 Criteria for applying t test

1. Random samples
2. Quantitative data
3. Sample size < 30
4. Variable normally distributed

Unpaired t test
 When sample in two groups give individual value,
 To test for the difference in between the groups
UNPAIRED T TEST
 When each individual gives a pair of observations, to test for
the difference in the pair of values, t test is utilized.
 Steps

1. Null hypothesis
2. The difference in each set is obtained. D=x1-x2
3. Calculate mean of difference
4. Calculate standard deviation
5. Test statistic is calculated t = d /SD/√n
6. Find degree of freedom
7. Compare calculated value.
Z TEST
 Used to test the significance of difference in means for large
samples (>30)
 Pre-requisites are:

1. Sample must be randomly selected

2. Quantitative data
3. Sample size > 30
4. Variable normally distributed
5. Observation-mean = x-x
Standard deviation SD
SAMPLING
 Sample is the part of a population.
 Sampling is the process or technique of selecting a sample of
appropriate characteristics and adequate size.
 Sampling frame is the total of the elements of the survey
population.
 Advantages
1. Reduces cost of investigation, time required, no. of personnel
involved.
2. Allows thorough investigation.
3. Provides adequate and in-depth coverage of sample units.
 Ideal Requirements
1. Efficiency – Ability to yield the desired information
2. Representativeness – Should represent parent population
3. Measurability – the investigator should be able to estimate the
extent to which findings from sample are likely to differ from
parent population.
4. Size – Large enough to minimize sample variability.
5. Coverage – Adequate coverage is essential
6. Goal Orientation – sample should be oriented towards study
objectives and research design.
7. Feasibility – Simple enough to be carried out in practice.
8. Cost-effectiveness – Desired info with appreciable savings in
time.
SAMPLE SELECTION
Purposive selection

• Represents the population as whole.

• Purposively select individuals who seem to represent the
population.
• Easy to carry
• Does not need preparation of samples
Random Selection

• Sample is selected in such a way that all characteristics of

population are reflected in sample.
SAMPLING METHODS
(DEPENDING UPON TYPE AND NATURE OF POPULATION OF
POPULATION)
Non-probability Sampling
(Not truly representative and are therefore less desirable than probability samples )
• Quota Sampling
• Purposive Sampling
• Convenience Sampling
Probability Sampling
(Each individual unit in the total population has a known probability of being
selected)
• Simple Random Sampling
• Systematic Sampling
• Stratified Sampling
• Cluster Sampling
Non Probability Sampling

Quota Sampling

• General composition of the sample is decided in advance.

• Right no. of people be somehow found to fill these quotas.
• Done to insure inclusion of particular segment.
• Eg- Minority samples

Purposive Sampling

• Constructed to serve a very specific need.

• It is snowball sample (chain referral sampling)
• Eg- Inclusion of business execs

Convenience Sampling

• Is a matter of taking what is available.

• Accidental sample.
• Eg- volunteers select
Probability Sampling

Simple Random Sampling

• Selection of unit is determined by chance only.
• Procedure – sampling frame, Size of sample, required no. of units.
• Methods
• Lottery method – Blindfold Selection.
• Table of random nos. – random arrangement of digits in rows and
columns
Systematic Sampling
• Obtained by selecting one unit at random.
• Then selecting additional units at evenly spaced interval.
• Adopted as long as no periodicity of occurrence of any particular event
in population.
Stratified Sampling

• Population is divided into subgroups or strata

according common characteristics.
• Types
• Stratified Random Sampling - used when the
population is heterogeneous with regard to
characteristic under study.
• Stratified Systemic Sampling – eliminates sampling
variation.
Cluster Sampling

• Used when population forms natural groups or

clusters.
• Sampling units are clusters.
• They are less expensive.
OTHER SAMPLING METHODS
Multiphase Sampling

• Part of information is collected from whole sample and part from sub – sample.

Multistage Sampling

• 1st stage is to select the groups.

• Then subsamples are taken in as many subsequent stages as necessary to obtain desired sample size.
REFERENCES
 Essentials of public health dentistry - Soben Peter –
5th Editiion
THANK YOU !