Biostatics

Statistics
The term statistics is derived from Latin word “ statisticus ” meaning study of facts and
figures. Statistics may be defined as the ‘’ branch of mathematics dealing with collection,
analysis, interpretation and presentation of numerical data.

It measures variability among different groups in mathematical terms.

Biostatics
Application of statistical methods in health sciences is called biostatistics

Now a day, biostatistics is considered to be the backbone of medical research. It provides

necessary tools to collect, analyze, interpret and present the data obtained in research in a
standard format. Thus, the data collected by different researchers becomes comparable with
each other. If we do not follow bio-statical protocols, our results of research may be erratic
and unacceptable to the medical research community.

In pharmacology, biostatical methods are mainly applied in evaluation of drug effects and
development of new drugs.

Biostatistics is also called Quantitative Medicine.

Branches of biostatics are:

 Descriptive biostatistics : Which only describes the data.
 Inferential biostatistics: In which we draw inferences (conclusions ) from the data.

Terms used in biostatistics

1) Population
It is an entire group of subjects to be studied. They may be humans,
experimental animals and likewise.
For example, if we wish to take height of adult males in Pakistan, all
males living in Pakistan will be the population.
2) Sample
This is the representative part of a population.
As it is impractical to measure the height of all males living in Pakistani , we take few
males who represent the adult male population in Pakistan. This set of males makes
the sample of the said population.
3) Data
It is the set of recorded values. It is raw material of biostatistics.
For example, if the height of six adult males of Pakistan are 170,172,172,165,168 and
175 cm, these values comprise the data.

4) Variable :
 A value that is measurable and mat vary (Take different values ) for different
individuals.
For example the value of height will vary from one adult Pakistani male to the other
variable is classified into
I. Numeric (quantitative) variable.
II. Categorical (Qualitative) variable

I. Numeric variable
It is further classified into
a. continuous variable
This variable accounts for only whole value in data e.g.1,2,2.5,3.5,4,10 etc.
b. discrete variable
This variable accounts for only whole value in a data which should not be a
decimal values e.g. 1,2,3,4,10,20,30 etc.
II. Categorical variable
It is further divided into
a. Nominal variable
This variable does not follow any order, e.g. in gender male & female.
b. Ordinal variable
This is the variable which follows some specific order, e.g. in age from youngest
to oldest.
5) Constant
There are the values that fixed and cannot change like pai will always be 3.14.

6) Attribute
It is the character or quality of an individual that may be used to categorize it.
For example Hypertensive subject, Tall boys etc.
7) Parameter
Parameter is a measurable characteristic of a population. These are the summary
values derived from population for its different characteristics like mean height, mean
weight etc.
8) Frequency
Each variable value in a collective set is called frequency.
9) Frequency distribution table
When we arrange the data it is presented as a table called frequency distribution.

Measure of central tendency

When data is arrange in an order , values in the middle are commonly the mostly occurring
and this portion of the data is called its center.

The statistical methods that measure how the values are distributed towards the center are
called Measures of central tendency.

These include Mean, Median and Mod.

 1. Mean
It is the average value in the data. It has three types.
 Arithmetic mean
 Harmonic mean
 Geometric mean

Arithmetic mean is the sum of all values divided by number of values. It is denoted by X.

X=X1+X2+X3………………..X0

For example mean of heights of Pakistani male will be

X=170+172+172+165+168+175

X=1022/6

=170

2. Median
This value divides the data into two halves. The data is arranged in ascending order. If
the values, the central value is the median. If the data has even number of figures, the
has even number of figures, then the middle two values are added and divided by2.
Heights of Pakistani males are 170,172,172,165,168,175
Arranged in ascending order as

165,168,170,172,172,175

Median= 170+172
2
Median =342/2
Median =171

3. Mode:
This is the moat repeated or occurring value in data.
For example ,172cm is the most occurring value in the given data for adult male
heights. This is the Mode of this data set.
4. Frequency of distribution curve.
The data which most of the values lie in the middle is called normally distributed
data. When this data is presented in a graph , it forms a typical curve called normal
distribution curve or Gaussian distribution curve. Values are plotted on X-axis and
numbers of individuals is plotted against y-axis. The hump of the curve denoted the
measure of central tendency i.e. mean , median or mode.
Fig

The curve on the top is shorter and wider than the curve on the bottom, because the
curve on the top has a bigger standard deviation.

Measures of dispersion
These statistics tell us how for data is scattered from its central values.

1. Range
This is the difference between minimum and maximum value.
2. Standard deviation
The term deviation means how widely individual values vary from their mean.
Standard deviation is its statistical calculation. If individual observations vary greyly
from the mean, standard deviation will be more and vice versa.
If the data is normally distributed ,68.3 % of the values lie within + 3 SD of mean.
Standard deviation is calculated by using formula as follows:

S.D= √∑D 2

(n-1)

3. Standard error
Research is conducted using samples from a population. If we measure all the
individuals in a population and take the population mean it is possible that it may be
different from the mean we have taken using a small group of people, i.e. sample.
Alternatively when two researchers measure same variable on two different samples
their means from the same variables may differ. This variation between sample mean
and population mean or means of two groups is measured by it is calculated by the
formula:
 S.E.= SD

√n

OR

S.E.= √∑D 2

n(n-1)

The larger the sample size, the less the S.E. will be.

Test of significance
When we have drawn the measures of central tendency and dispersion for data obtained from
different groups under research , we want to ascertain whether the difference noted is just by
chance or actually exists.

This analysis is made using tests of significance.

1. Probability or P-value
It is the value for probability of chance i.e. how much the noted difference may be due
to chcne.
In research typically the P- value is taken as 0.05 or less. P- value greater than 0.5
denotes the difference may be a result of chance.
2. CHI- square rest.
It is used to analyze qualitative or non- parametric variables. It is noted by X2.
3. Student t- test.
It is used to analyze difference between quantitative or parametric variables. It is
calculated by using formula
T= M1-M2
√SE1+SE2
If the value less than 2, it is note significant , and if more then 2, then its significant.