MUMBAI VETERINARY COLLEGE, GOREGAON, MUMBAI

DEPARTMENT OF ANIMAL GENETICS AND BREEDING

B.V.Sc. & A.H.

PRACTICAL MANUAL
AGB-UNIT-I

(Biostatistics and Computer Applications)

MAHARASHTRA ANIMAL AND FISHERIES SCIENCES UNIVERSITY, NAGPUR

 Certificate
This is to certify that Mr. / Ms. ……………………………………………………………………..
Reg. No. ….Enroll. No……………… …studying in Second Year, B.V.Sc. & A. H. Mumbai
Veterinary College, Mumbai has successfully completed his/ her practical work for Unit-I of Animal
Genetics and Breeding during the academic year

Place: Mumbai

Date: Signature of Course Teacher

ANNUAL BOARD EXAMINATION

Evaluated the practical record submitted for Annual Board Practical Examination held on
at Mumbai Veterinary College, Mumbai.

Signature of External Examiner Signature of Internal Examiner

-2-
 INDEX

Sr. Page
Title Date Signature
No. No.

Frequency Distribution, Graphical and

1
Diagrammatical representation of data

Estimation of Measures of Central tendency

2 for simple and grouped data: Mean, Median,
Mode

Estimation of Measures of Dispersion for

simple and grouped data: Range, Mean
3
Deviation, Standard Deviation, Standard
Error, Variance, Coefficient of Variance.

4 Estimation of Correlation and Regression

5 Simple Probability problem

Normal distribution, test of significance:

6
Z test

Test of significance: student ‘t ’test for single

7
mean, difference of mean and paired ‘t’ test

8 Test of significance : chi – square test

9 Test of significance: ‘F’ test

10 Completely Randomized Design (C.R.D)

11 Randomized Block Design (R.B.D)
Computer basics and components of
12 computer, simple operations, entry of
biological data

‘Internet and e- mail, entering and saving

13 biological data through MS- office,
MS –Excel

-3-
Practical No : 1

Title : Frequency Distribution, Graphical and Diagrammatical

Representation of data

Objectives:

1. To classify the huge data into concise form

2. To make the understanding of the data easy

Steps :

1. Divide the whole range of data into various classes or class intervals

2. The number of class intervals depends upon volume of data to be classified

3. Allotment of observations is done to the respective class interval.

4. The frequency, relative frequency, percent relative frequency is calculated.

Frequency distribution is the classification of random variable into classes or class intervals
indicating number of times the representative of each class has been repeated into the data. The table
showing the distribution of frequency is called as frequency table.

A frequency distribution gives valuable insight into the behavior of random variable. In order to have
the right perspective of frequency counts, relative frequency and percent relative frequency are
included into frequency table which is now called as complete frequency distribution table.

Cumulative frequency:

A cumulative frequency distribution is a frequency distribution when successive frequencies

are added together so that each class includes all the classes below or above depending upon the end
from which the cumulative process begins.

Procedure for formation of Frequency table

Classification according to class intervals (Frequency distribution):

When the data is grouped into classes of appropriate intervals, showing the no. in each
class, we get frequency distribution.

 Class limits:

-4-
 o Class limits are the limits within with the class intervals lies. Thus each class intervals
has two limits, upper & lower.

 Class Interval:

o Class interval is a frequency distribution giving the range of value, the character
withthe lower and upper limit.

 Frequency:

o Frequency is total number of observations in that class.

 Width or length of class interval :

o Width of class interval is the difference between the upper limit & lower limit of the
same class.

 Formation of Frequency distribution:

Methods of tally marks – first we have to form the class interval.

Maximum value – Minimum value in the data set

Class interval =

No. of required classes (K)

This value should be rounded off for our convenience. The no. of required classes can
be calculated using the formula suggested by Struge’s rule.

K  1 + 3.322 log n

Where ‘K’ is the number of required classes & ‘n’ is total number of observations. After forming
the class interval each should be written one below the other & for each item in the collected data, a
stroke is marked against the class intervals in which it falls. Usually after every four stroke in class
interval the fifth item is indicated by striking the previous four strokes, thus making it easy to count.

Exercise No.-1:
Weight in kgs. of sheep at a particular age group in Mechary sheep farm is given below. Formulate
frequency distribution table:

-5-
17, 13, 11, 12, 17, 11, 14, 16, 18, 11, 23, 12, 21, 13, 23, 20, 18, 19, 14, 20, 24, 24, 11, 12, 24, 15, 16,
16, 18, 15, 23, 19, 21, 24, 23, 14, 18, 19, 16, 14, 14, 13, 11, 19, 16, 17, 19, 11, 13, 17, 18, 13, 21, 12,
23, 11, 19, 11, 24, 24, 20, 14, 16, 14, 16, 18, 23, 23, 19, 14, 14, 19, 16, 18, 16.

Exercise No.-2:
Egg yield of a layer from a particular time is given below formulate frequency distribution table:

200, 263, 265, 264, 260, 261, 267, 270, 280, 267, 264, 263, 261, 266, 260, 299, 292, 290, 280, 281,
288, 287, 289, 290, 270, 272, 273, 280, 281, 283, 285, 291, 260, 267, 290, 281, 296, 295, 292, 280,
270, 271, 300, 291, 298, 288, 289, 299, 266, 284, 285, 287, 289, 287, 291, 290, 300, 268, 261, 263,
292, 291, 290, 289.

Exercise No.-3:
Lactation length of buffaloes at LSR, Kattupakkam is given below formulate frequency distribution
table.
270, 280, 300, 305, 301, 290, 295, 297, 291, 294, 273, 286, 285, 283, 266, 283, 286, 290, 281, 296,
275, 277, 299, 277, 273, 304, 279, 271, 279, 286, 279, 280, 305, 277, 287, 285, 304, 285, 307, 281,
271, 274, 279, 272, 274, 276, 282, 286, 287, 289, 301, 304, 306, 300, 280, 291, 281, 284, 287, 290,
282, 281, 279, 274, 276, 271.

Exercise No-4:
The information on wool production in sheep in kg. is given below. Construct the discrete and
continuous type of frequency distribution table from the data.

4.5,3.2,2.8,2.1,1.9,3.1,2.7,2.6,2.2,4.5,2.8,4.1,3.2,1.9,1.9,2.7,3.9,3.6,2.6,2.8,4.1,2.7,1.8,
3.2,3.3,4.1,3.1,2.7,4.5,3.2,4.5,4.1,2.2,2.7,3.1,3.1,3.9,2.1,2.1,3.2,4.1,2.6,2.8,2.6,2.6,2.2,
1.9,1.9,4.1,2.2,4.5,2.7,4.5,3.2,1.9,3.9,3.1,2.6,3.2,2.7,3.0,3.0,1.9,3.0,2.8,4.1,3.0,2.7,3.3.

-6-
Graphical and Diagrammatical representation of data:
Objectives:

1. To create long lasting impression on mind of reader

2. Makes the comparison between the variables easy

3. To present the huge data in the form of Graph/picture

4. To understand the relationship between the variables
Diagrams always help the statistician to visualize the meaning or a numerical complex at a
single glance. A large number of diagrams are used in biostatistical analysis. The most useful and
effective way of presenting the data is through diagrams and graphs.

The important type of diagrams that are commonly used for presentation of data are Line
diagrams, Bar diagrams and Pie diagrams etc.

LINE DIAGRAMS:

It is the simplest type of a diagram. For diagrammatic representation of data, a line diagram
can present the frequencies of the discrete variable. The variable is taken on the X-axis, and the
frequencies of the observation on the Y-axis. The straight lines are drawn whose lengths are
proportional to the frequencies.

BAR DIAGRAMS:

Bar diagrams are commonly used in practice to represent the statistical data. They are also known as
one – dimensional diagrams because the length of bar is important, and not the width. Bar diagrams
can be of following types.

o Simple bar diagram:

A simple bar diagram is used to represent only one variable. As one bar represents only one
figure, there are as many bars as the number of figures.

Steps:

 Draw X axis and Y axis.

 Arrange the variables among X axis

-7-
  The respective frequencies are written on Y axis

 The bars are drawn against the frequency values on Y axis

Exercise No.-1:
Draw simple bar diagram & the line diagram for following data. Average income per farm.

Farm Income (in Lakhs )

A 7.0
B 11.3
C 12.5
D 9.0
E 10.0

Sub divided bar diagram:

In a divided bar diagram, the frequency is divided in to different components & such a
representation is called a divided bar diagram.

Steps:

 Draw X axis and Y axis.

 Arrange the variables among X axis

 The respective frequencies are written on Y axis

 The bars are drawn against the total frequency values on Y axis

 These bars are divided into various groups and properly shaded or coloured

 An index showing the different shades or colors should be written at top right corner of
each diagram.

Exercise No.-2:
Milk yield (litres) of three farms is given below construct subdivided bar diagrams

Farm Buffalo Cows Total Rs.

A 250 500 750

B 300 600 900

C 100 400 500

-8-
Exercise No.-3:
The wages given for the males and females are shown below. Construct subdivided bar diagram

Farm Male Female

A 500 400

B 600 450

C 650 500

o Multiple bar diagram :

Multiple bar diagrams are preferred whenever a comparison between two or more related
variable is to be made. The technique of simple bar diagram can be extended to represent two
or more sets of interrelated data.

Steps:

 Draw X axis and Y axis.

 Arrange the variables among X axis

 The respective frequencies are written on Y axis

 Multiple bars are drawn against the frequency values on Y axis

 An index showing the different shades or colors should be written at top right corner of
each diagram.

Exercise No.-4:
First three lactation lengths for 4 breed is given below. Construct component multiple
bar diagrams.

Breeds Lactation I Lactation II Lactation III

Sahiwal 280 260 310

Gir 220 250 280

Tharprkar 310 260 220

Red Sindhi 300 320 290

-9-
Exercise No- 5:

The year wise and faculty wise distribution of students from a college is given as below. Construct the
multiple bar diagram by using the information.

Year Number of students

Arts Commerce Science Others

2001 250 300 460 240

2002 320 280 300 150

2003 190 300 210 320

2004 300 210 190 300

2005 280 220 400 150

PIE DIAGRAM:

This is another way of presenting dictate data of qualitative characters such as blood groups,
age groups, total expenditure & total area under cultivation for different crops etc. In situations where
different components or frequencies are to be shown by means of sector of a circle , the angles of the
sectors are proportional to the respective measurements of the different components & such a diagram
is pre diagram.

Income for one source

Degree = X 360

Total income

Steps:
 The angle for sectors made in the center of the pie is calculated
 The sectors or components of a pie are drawn as per the angles they made in the
center of pie.
 Shade or color different sectors.
 An index is shown at top right corner of the pie diagram

- 10 -
Exercise No.-4:
Expenditure pattern of a family per month is given below. Draw a pie diagram.

Particulars Cost Rs. Angle ( °)

Food 1500
Rent 1500
Clothes 500
Recreation 500
Education 2000
Miscellaneous 1200
Total 7200

Exercise No.-5:
Income from various sources of mixed farming is given below. Construct a pie diagram.

Source Income Rs. Degree ( °)

Agriculture 4500

Animal Husbandry 7000

Bee keeping 3000

Poultry 5000

Other 2000

Total 21,500

Graphs:
A graph is a visual form of representation of statistical data. The frequency distribution
can be represented graphically in any one way of following types:

 Histogram

 Frequency polygon

 Frequency curve

 Cumulative frequency curve

- 11 -
Construction of graph:

It will be convenient in discussing frequency graphs to use the conventional mathematical

terms, ordinate and abscissa. In the construction of graphs the values of the variables are measured on
‘X’ axis and the corresponding frequencies in ‘Y’ axis. For the construction of graphs two lines are
drawn which cut each other at right angles. The horizontal line called abscissa and vertical line is
called as ordinate. The point where these lines cut each other is called as point of origin (0). Together
the ordinate and abscissa are called co-ordinates to the point.

HISTOGRAM:

In constructing the histogram, the variable, number of grains per spike should be taken on the
horizontal axis (X-axis) (abscissa) and the frequencies depending on it taken on vertical (Y-axis)
(ordinate). Each class represented by a distance, which is always proportional to its class interval.
When all the classes are of equal lengths, the size of the rectangles will be proportional to the
frequencies of the respective classes. In this way, there are no of rectangles each with a class interval
distance as its width, and frequency distance as its height. Histogram is two-dimensional where both
the length as well as the width are importance where as in a bar diagram is one dimensional.

Steps:

 Draw X axis and Y axis.

 Arrange the variables among X axis

 The respective frequencies are written on Y axis

 The bars of equal width are drawn against the frequency values on Y axis.

 The bars should be continued from point of origin upto the end without any break.

FREQUENCY POLYGON :

Frequency distribution can be portrayed graphically by means of a frequency polygon. To

construct a frequency polygon, we mark the frequencies on the vertical axis and values of the variable
on the horizontal axis as in case of histogram. A dot is placed above the mid-point of each class and
the height of a given dot corresponds to the frequency of the relevant class interval. By connecting
dots by straight lines the frequency polygon is prepared

Steps:

 Draw a Histogram

 Locate the mid point of each bar of histogram

 Join the points on each bar with smooth hand without formation of vertices

- 12 -
FREQUENCY CURVE:

The frequency polygon or histogram will approach more & more the form of a smooth curve.
Such a curve is obtained in normal distribution of individuals in a large sample or in a population. The
frequency curve is drawn freehand to eliminate as far as possible, the accidental variations that might
be present in the biological, agricultural and other data.

Steps:

 Draw a Histogram

 Locate the mid-point of each bar of histogram

 Join the points on each bar with smooth hand without formation of vertices

Cumulative frequency curve: (OGIVE)

A graph of cumulative frequency distribution is called Ogive (Oh-give). There are two
methods of constructing ogive, namely:

The “less than” method

The “more than” method

In the “less than” method, we start with the upper limit of the classes and go on adding the
frequencies. However, in case of “more than” method, we start with the lower limit of classes. The
first method gives a rising curve, whereas the second method shows a declining curve.

Steps:

 Calculate less than and more than frequency from simple frequency of each class.

 The variable values are written on X axis and respective frequencies on Y axis.

 The values of variables are plotted for less than and more than type separately against
frequencies on Y axis.

 The points are joined and less than and more than ogives are obtained

- 13 -
Exercise No.-1:
Construct a histogram, Frequency polygon, frequency curve, and cumulative frequency curve for the
following.

Class Frequency Lesser Greater

11-13 11 11 75
13-15 14 25 64
15-17 11 36 50
17-19 11 47 39

19-21 12 59 28
21-23 3 62 16
23-25 13 75 13

- 14 -
Practical No : 2
Title : Estimation of Measures of Central tendency/Averages for simple

and groupeddata: Mean, Median, Mode

Objectives:

 This makes the central theme of the data readily understandable.

 The averages are extremely useful for comparison purpose.

Types (forms) of averages:

1. Arithmetic mean (AM) or Mean of common mean

2. Geometric Mean (GM)

3. Harmonic Mean (HM)

4. Median

5. Mode

1. Arithmetic Mean (AM):

It is the value obtained by dividing the sum of the values of the given items (of a
variable) by the number of items. Thus,

Mean= Sum of the values of items in the series

Total number of items.

It is usually denoted by x

If we denote all the ‘n’ observations in a series by x1, x2, x3, ……, xn then arithmetic
mean or mean for that series will be given by

Arithmetic mean = x1+x2+…….+xn

In the case of frequency distribution if the different class marks of the ‘n’ classes are denoted
by x1, x2, x3, ….., xn and the corresponding frequencies by f1, f2, ….., fn then the mean of data is,

X= f1x1+ f2 x2+ f3x3+ .......... + fnxn

f1 + f2+ ……+ fn

- 15 -
 = ∑ fixi = ∑ fixi

∑ fi N

Steps:

 Add together the various values of the given variables.

 The variable is generally denoted as X and total ∑X is obtained.

 Divide the total obtained in above step by number of items

Combined mean:

If x1 is the mean of first group or n1 items, x2 is the mean of 2nd group of n2 items, then the combined
mean of the 2nd group is

n= n1x1 + n2x2

n1 + n2

x1 = (sum of the all the observations in the first group/n1)

n1 x1 = (sum of all the observations in the first group)

X2 = (sum of the observations in the second group/n2)

n2 x2 = (sum of all the observations in the second group.)

n1 x1+ n2 x2 = (sum of all the observations in the two groups of size n1+n2)

Therefore, X = (n1x1 + n2x2)

n1 + n2

Extending the above result if x1 is the mean of the ith group of n1 observations, then

X = ∑ nixi

∑n

- 16 -
 2. Geometric mean (GM)

Steps:

 Obtain mid points of various classes

 Obtain logarithms of mid points

 Multiply these logarithms by respective frequencies and obtain ∑ f log m

 Divide the ∑ f log m by total frequency and take antilog of value so obtained

The geometric mean is the nth root of product of ‘n’ items of a series. If x1, x2, x3, …, xn are the
‘n’ observations in a services in a series then GM is given by

GM = n√ (x1. x2... xn) = (x1.x2…..xn)1/ n

To simplify we take log on both sides,

Log GM = log (x1.x2…..xn)1/ n

= 1/n (log x1 + log x2 + ….. + log xn)

= ∑ log xi

Therefore, GM= antilog ∑ log xi

n

Thus, GM is the antilogarithm of the arithmetic mean of the logarithmic values. Logarithm of
geometric mean is the arithmetic mean of the logarithmic values.

In case of frequency distribution (grouped data), GM is given by

GM = (x1f1, x2f2... xnfn) 1/n

where N= Total frequency = ∑ f1 log x1 is the mid-point of the class with frequency f1

Simplify by taking logarithm on both sides,

Log GM = 1/N log (x1f1, x2f2, .... , xnfn)

= 1/N (f1log x1+ f2log x2+……+ fnlog xn)

- 17 -
 = ∑ fnlog xn
N

antilog(∑filog xi)
Therefore, GM =
N
Where x1 is the mid value of the class whose frequency in f2

3. Harmonic mean (H.M.)

Steps:

 Calculate the mean values

 Obtain the reciprocal of mean values

 Multiply these reciprocals with respective frequencies and take the total ∑ (f ×1/ x)

 Divide the frequency by the total obtained. It will give the value of harmonic mean

Harmonic mean is the total no. of items of a variable divided by the sum of the
reciprocals of the items. If x1, x2… xn are the ‘n’ observations & HM represents the harmonic mean,
then,

HM = n = n

(1/ x1+ 1/ x2+ ……+ 1/ xn) ∑ 1/ xi

= 1 = n / (∑1/ xi ) 1

(∑1/ xi)/ n) A.M. of reciprocals

Harmonic mean is the reciprocal of arithmetic mean of reciprocal values. In the case of a
frequency distribution, HM is obtained by using the formula,

HM = f1+ f2+ ……+ fn = N

F1/x1+ f2/x2+ …+fn/xn ∑ fi / xi

Where xi is the mid value of the class whose frequency is fi. N is total frequency.

- 18 -
 4. Median:

It is the value which has got equal number of observations on either side when the items
arearranged in the ascending or descending order of magnitude.

Median divides the data into two equal parts, one part will consist of all variables less than median &
other part greater than median.

For an ungrouped (row) data:

Case A:

When n is odd then median = size of (n+1/2)th item after arranging the data in ascending or
descending order of magnitude.

Case B:

When n is even then median = average of (n/2)th and (n/2) +1th item after arranging the data in
ascending or descending order.

In the case of frequency distribution, Median is the value which has got equal no. of
frequencies on either side i.e. which correspondent to the cumulative frequency of N/2. It is obtained
by

Median= L+ ((N/2-m) x C/f

Where,

L- is the lower boundary of the medial class,

N- is the total frequency,

m -is cumulative frequency up to the medial class,

C -is width of the class interval &

f -is the frequency in the medial class.

Note:

1) Medial class is the class corresponding to the cumulative frequency equal to or just greater
than N/2.

2) Median can be computed using ogive. It is the x co-ordinate of the point of intersection of the
less than and the greater than cumulative frequency curve.
- 19 -
Steps:

 First arrange the data in ascending or descending order of magnitude

 Then ascertain the middle item by applying the formula

Median= N+1/2 in an array.

 This is true for odd number of items

 If the number of observation are even then median is the value mid way between the
two middle values

5. Mode:

It is the size of the most frequent item in a large set of data. Thus, mode is the value of
that variable which occurs most frequently or repeats itself the greatest no. of times.

In the case of grouped data, mode can be calculated by,

Mode = L+ (Cf2 /f1+f2)

Where

L- is the lower boundary of the modal class,

f1, f2 -are the frequencies in the preceding & succeeding modal class

C-is the class interval.

Note:

1. Mode can be computed from histogram. Left in the X co-ordinate or the point intersection or the
two diagonal from the top corners of the modal class to the pre & post class top corners.

2. As a first approximation mid –point or the modal class will be taken as the value or mode which is
called ‘crude mode’.

3. In a moderately asymmetrical distribution mean mode =3 (mean- median) (approx.),

Mode = 3 median - 2 mean (approx.) this an empirical mode.

4. A distribution can have more than one mode. If it has got two modes ,it is called uni-mode
distribution ; if it is has got two modes , it is called bi-modal distribution ;if it has got three mode, it is
called as tri-modal distribution ; it has got more than three mode ,it is x called as multi modal or poly
modal distribution .

- 20 -
Exercise 1:

The data pertaining to distance covered in kilometers by each person with bicycle is given below.
Compute Arithmetic mean.

Distance(Km) No of persons

20 12

25 11

30 10

35 04

40 03

Exercise 2:
From the following data of wedges obtained by the labors in a factory, calculate the arithmetic mean
by using direct method and step deviation method.

Wedges No of Labors

0-50 05

50-100 02

100-150 03

150-200 01

200-250 02

- 21 -
Exercise 3:
The data on the length (mm) of 20 types or wool are given below find the Arithmetic mean, Geometric
mean, Harmonic mean, Median and Mode.
132, 132, 138, 138, 140, 142, 144, 145, 146, 146,

147, 147, 149, 150, 152, 154, 161, 164, 168, 176

Exercise 4:
Find A.M., G.M, H.M, Median and Mode for the following data

Lactation length (days) No. of animals (F)

120-130 7

130-140 11

140-150 21

150-160 11

160-170 7

170-180 3

- 22 -
Practical No : 3
Title : Estimation of Measures of Dispersion for simple and grouped

data: Range, Mean deviation, Standard deviation, Standard error,

Variance, Coefficient of variance.

Objectives:

The measures of central tendency are used to study only one aspect of the data i.e the central
position however it is not sufficient to describe the total data. Along with the central aspects it is also
necessary to describe the variation present in the data. This presence of variation in the data can be
studied by measures of dispersion. The measures of dispersion are also called as measures of spread or
measures of scatter.

Absolute & relative dispersion :

When dispersion is expressed in terms of original units of series for eg. Weight in kg, income
in rupees etc. It is called absolute dispersion. If dispersion is expressed in terms of a pure number, free
from units of measurements then the dispersion is relative dispersion.

A relative measure of dispersion is an absolute measure of dispersion divided by an average.

Different measures of dispersion:

 Range

 Quartile deviation

 Mean deviation

 Standard deviation

Range:

It is the difference between the highest and lowest values in the raw data series for the grouped
data, the range is the difference between the lower limit of the first class and upper limit of the last
class. It is a very simple measure of dispersion. It is useful for the study of variation in money rate &
rarest of exchange, weather forecast etc.

Relative measure of dispersion for range is the ratio (R.R) given by

(H- L)
R.R. =
(H+L)

- 23 -
Steps :

 Locate the maximum and minimum values from the array

 The difference between this maximum and minimum value is the range

2. Quartile déviation:

Also known as the semi-quartile range. It is based on quantities, which are points that divided
the data series into four equal parts. The lower or First Quartile (Q1) divided lower half of the
distribution into 2 equal parts i.e. it is the value below which 25 % of the observation lie and above
which 75 % of the observations i.e. Similarly the upper or third quartile (Q 3) divided the upper half of
the distribution into 2 equal parts i.e. it is the value below which 75 % of the observation lie and above
which 25 % of observations lie.

The different Q3 – Q1 is called inter quintile range and QD is given by

Formula to compute QD = (Q3-Q1) / 2.

In the case of raw data, agree arranging the data in ascending order

Q1 = Size of the {n+1/4}th term.

Q3 = Size of the 3 {n+1/4}th term

Then, QD = Q3 – Q1/ 2

In case of frequency distribution or grouped data,

Q1 = L1 + (n /4 – m1) X C

F1

Where L1 is the lower boundary of the quartile class F1 is the cumulative frequency in the first
quartile class & C is the width or the class interval.

Q3 = L3 + (3 n / 4 – m3) X C
F3

Where L3 is the lower boundary of the third quartile class ; m3 is the cumulative frequency in the third
quartile class and C is the width of the class interval.

- 24 -
Then

QD = (Q3-Q1) / 2

Relative measure of QD is known as the quartile coefficient of dispersion (QC)

QC of dispersion = (Q3- Q1)

(Q3+Q1)

3. Mean deviation:

Mean deviation or average deviation in a series is the AM of the deviations of the various
items from an average (mean, median as mode) of the series taking all the deviations as positive.

For raw data, MD about mean = ∑ [Xi -X]

MD about an average A = [X1-A]

Where A is mean or median or mode Xi is mid-point of the I th class with frequency f1

The relative measure of mean duration is known as mean coefficient of dispersion or

coefficient of mean deviation and is obtained by dividing the MD by the average from which it is
computed.

(MD about an average A)

Coefficient of MD about an average A = (A)

MD about an average A X 100

Coefficient of MD about an average A in % = A

NOTE: In actual practice, MD is elucidated either from mean or median, but mode is not used is value
is indiscriminate. However, median is preferred to mean because mean deviation from the median is
minimum.

- 25 -
Steps:

 An average is computed for the given data

 Difference between the value of each item and the average is calculated

 The arithmetic mean of these deviations gives the mean deviation

4. Standard deviation (SD)

It is the most perfect & widely used measure of dispersion. It is an improved method over MD. It
is the root mean square of the deviation measured from the mean in other words; SD is the Square root
of the AM of the square of the deviation of items taken from AM of the series.

It is denoted by ‘σ ’ (Sigma)

For raw data

σ = √∑ [Xi -X]2

For grouped data where X1 is the mid value of the class interval whose frequency is f1 & N is the total
frequency.

σ = √∑ fi [Xi -X]2
N

To simplify above,

For raw data

σ = √∑ Xi2 - ∑Xi]2/ n
N

For grouped data

σ = √∑ fiXi2 - ∑fiXi]2/ n
N

- 26 -
Steps:

 Calculate the sum total of the variable (∑X)

 Divide ∑X with total no of observations and calculate mean value

 Calculte the deviation of each item from mean value

 Square the values of deviation of each item from mean value

 Calculate standard deviation by using the formula

SD = √(X-X)2 /n

Shepherd’s Correction :

In computing the standard deviation, sometimes, grouping error may occurs n account of
grouping of data into different classes for statistical adjustment of this grouping error , shepherd has
suggested a correction value to be deducted from the variance of the grouped data , given by

C² Square of width of class interval

12 l2

σ = √∑ fiXi2 - ∑fi Xi]2/ n - C²

N 12

Coefficient of variation:

Relative measure of standard deviation is known as the coefficient of variation (CV or Cov)
and is defined as SD / Mean.

σ X 100

Therefore, CV = X

Thus, CV is the % variation from the mean, with so being treated as the total variation. Higher cv
indicates greater variability & less CV implies better consistency of data.

- 27 -
Variance:
Square of standard variation is called as variance. It is the mean square deviation. It is sum of
the squared deviation of individual observations from the mean divided by No. of observation. It is
denoted by ‘σ 2 ’

Standard error:

The mean of a random sample may be taken as a representation of the population mean. The
difference between the sample mean & the population mean is due to the sampling & it is called the
sampling error or the standard errors.

It is defined as the SD of the mean of different samples, taken from population.

If we study only one sample then

SE (X) = SD

Where n is the size of the sample and SD is √n that of the sample.

Exercise No.-1:
Find Range, Quartile Deviation, Mean Deviation , Standard Deviation and their relative measure for
thefollowing data.
1.3, 1.1, 1.0, 2.0, 1.7, 2.0, 1.9, 1.8, 1.6, 1.5

Exercise 2:
Compute the Mean Deviation and Standard Deviation for the following data on weights
of fishes in grams
10, 20, 30, 50, 70, 80, 90

- 28 -
Exercise 3:
Compute the Standard deviation and coefficient of variation from given data
Sr. No Class interval Frequency
1 3.25-3.55 02
2 3.55-3.85 05
3 3.85-4.15 11
4 4.15-4.45 05
5 4.45-4.75 02

Exercise 4:
Calculate range, QD, MD, Mean, Median , Mode, Standard deviation and their relative measures for
following data

Lactation period Number of cows

120-130 7

130-140 11

140-150 21

150-160 11

160-170 7

170-180 3

- 29 -
Practical No : 4
Title : Estimation of Correlation and Regression

Objective: The correlation is calculated to find out the association between the two or more variables

Correlation of the statistical technique which measures and analyses the degree or extent to
which two or more variables fluctuate with reference to one another. Correlation thus denotes inter
dependence among variables. The degrees are expressed by coefficient that ranges between -1 & +1.
The direction of change is indicated by + or – sings, the former, refers to the sympathetic movement in
the same direction and the latter, in opposite direction. An absence of correlation is indicated by zero.

Types of correlation

There are four broad types of correlation.

o Positive and Negative

o Simple and Multiple

o Partial and Total

o Linear and Nonlinear

Methods:

Various methods classified broadly as:

Graphic

Through graph through sector diagram

Algebraic

Product movement/ Rank method concurrent

Covariance method deviation method

Algebraic methods:

These methods are based on mathematical formulae. The relationship is expressed by coefficient in
range of ±1. The following are main algebraic methods:

- 30 -
Karl Pearson’s method or covariance method (product movement):

This measure known as `Pearsonian correlation coefficient between two variables x & y,
usually denoted by ‘r’ and is defined as the ratio of the covariance between x & y [written as Cov(x,
y)] to the product of standard deviations of x & y.

Symbolically,

Steps:

 Calculate the sum total of both the variables i.e ∑X and ∑Y

 Then from above values (∑X)2 and (∑Y)2 is calculated

 Make the squares of each variable and add them to calculate ∑X2 and ∑Y2

 Calculate ∑XY by multiplying the values of both the variables X and Y

 Calculate Correlation by using following formula

r = ∑XY - ∑X × ∑Y/n

√(∑X2 - (∑X)2 /n) (∑Y2 - (∑Y)2 /n )

Example No- 1.
Find the coefficient of correlation between x and y from the following data, using co-variance
method.

X 5 10 5 11 12 4 3 2 7 1

Y 1 6 2 8 5 1 4 6 5 2

2] Rank correlation method (spearman’s correlation):

Some times we may not know the actual values, but their ranking may be known. In such
occasions, this method would be used. Even when the actual values are available, we can rank them by
using rank correlation method.
- 31 -
When the no. of paired observations exceeds 30, it is very difficult to rank them and hence unless rank
is given, it is better to avoid this method. This method is also called as spearman’s rank correlat ion
coefficient.

Exercise 2:

Obtain the coefficient of correlation for the following data on length (X in cm ) and weight (Y in
gram) of the frogs.

Length (X in Cm) Weight (Y in gms)

5 8

7 9

3 5

1 4

9 9

12 13

8 7

3 9

- 32 -
Regression:

The meaning of regression is the act of returning of going back in a study where data on age
and weight of animal are available, age could be considered as the independent variable, while,
weight, as the dependant variable. It means that the weight regresses on age.

In the regression, line Y = a+bx, a and b are calculated by

bxy = Cov (xy)

σ x2

∑xy - ∑x . ∑y

= n

∑ x2 – (∑ x)2

a =y - b x

Objectives:

To estimate the unknown values of a variable from known values of another correlated variable

Exercise 1:
Obtain the two regression equations length (X) on weight (Y) and weight (Y) on length
(X) from the following data on length and weight of fish.
X 5 7 3 1 9 12 8 3
Y 8 9 5 4 9 13 7 9

- 33 -
Exercise No.-2: From a farm, 36 sheep were selected at random. The weight of sheep (X) and
length of wool fibre (Y) of the selected animal were recorded. The results are given in table. Fit a
regression line and y and x Estimate the length of wool for sheep weighing 30 Kgs.

Y X Y X Y X

95 22.4 143 24.5 112 22.9

109 23.3 127 23.6 131 23.9

133 24.1 92 21.1 147 24.8

132 24.3 88 21.4 90 21.2

136 23.5 99 23.4 110 22.2

116 22.3 129 23.4 106 22.7

126 23.9 91 21.6 127 23.0

124 24.0 103 21.4 145 24.0

137 24.9 114 23.3 85 20.6

90 20.0 124 24.4 94 21.0

107 19.8 143 24.4 142 24.0

108 22.0 108 22.5 111 23.1

- 34 -
Practical No : 5
Title : Simple Probability problem

Objective:

The main objective of probability is that it indicates the desire for an event to occur.

The theory or mathematical probability has its origin in the 17th century. There are three diff.
approaches of measuring probabilities. They are classical probability, relative frequency of occurrence
and automatic probability.

The outcomes or random experiment are termed as event. The probability for the occurrence of
an event ‘A’ is defined as the ratio between the no. of favorable outcomes for the occurrence of the
event and the total number of possible outcomes i.e.

Probability of an event = Number of favorable outcomes

Total no. of Outcomes

In this definition, we have to assure that the outcomes are equally likely to occur.

Example:

Suppose a coin is tossed, there are two possible outcomes head and tail. Both are equally likely
events.

Probability of getting head is ½. i.e.50%

Example: Suppose a bag contains 4 white and 6 red balls and a ball is drawn at random. These are

10 equally likely events,

The probability that the ball drawn is white = 4 = 2

10 5

Additional Theorem:

In the case of a set of mutually exclusive events, the probability (p) for the
occurrence of all these events is equal to the sum of their separate probabilities P1 , P2 , ......... Pn.

i.e. P = P1 + P2 +P3 + ......... +Pn

- 35 -
Multiplication Theorem:

In the case of a set of independent events, The probability (p) for the occurrence of one
all these events is equal to the product of their separate probabilities P1 , P2 , ......... Pn.

i.e. P = P1 x P2 x P3 x ..........x Pn

Exercise No.-1: An animal scientist works with 3 different cattle breeds to improve theirs
efficiency of beef production. Suppose he has 10 steers. 5 of which are Angus, 3 are Brahman and 2
are Hereford. He selected 3 animals at random,what is the probability that all 3 are Agnus?

Exercise No.-2: An animal scientist wishes to choose 3 animals for his research from 8 sheeps, 3 of
which are Mechari. What is the probability that a randomly selected group will have 2 Mechari and 1
Suffolk?

Exercise No.-3: Suppose a bag contains 4 white and 6 red balls. What is the probability that A)
Both are white balls B) Both are red C) One white and one red (two balls are drawn at random)

Exercise No.-4: If we assure that male & female calves are equally likely, what is the probability of
exactly 3 out 10 calves born will be male?

- 36 -
Practical No : 6
Title : Normal distribution, test of significance: Z test

Objective:

The ‘Z ’ test of significance is used where the sample size is large. The large size population in
statistics can be any population with 30 or more number of observations. We made the decision about
the Ho by comparing the calculated ‘Z’ with the minimum ‘Z’ values required to reject the Ho. It was
pointed out that these minimum Z values required to reject the null hypothesis are valid only for cases
with large samples.

Steps:

 Calculate the population mean and sample mean

 Calculate difference between the population mean and sample mean
 Calculate the standard deviations for both population and sample as per discussed in
practical no 3
 Calculate Z value by using following formula

X1- X2
Z= √ (s12/n1+s22/n2)

1. To test the significant difference between sample mean & population mean:

Step1: Ho: No significant diff. bet. Sample & population mean

Step2: Test statistics or z statistics given by
Z= Sample mean- population mean
SE of difference
= X-M where X = mean of sample
SE (X-M) M = mean of population
Z= X-M where σ is SD of population, n is size of sample
σ √n
When σ is not known, replace it by‘s’ which is S.D. of sample. Then,
Z= X-M
Step 3: Conclusion
i. È z È < 1.96, we say z is not significant & Ho accepted. we denote it by z=( )NS

- 37 -
 ii. È z È > 1.96, z is significant & Ho is rejected. Denoted by z = ( )*
iii. È z È > 2.58, z is lightly significant & Ho is rejected Denoted by z = ( )**

3. To test significant diff. bet. Sample proportion with population proportion

If ‘n’ is no of trials, ‘p’ is the proportion of success, then p follows normal distribution
when ‘n’ is large.
Ho: there is significant diff. between p & P

Test statistic, Z = p-P = p-P where Q=1-P

SE (p-P) PQ
n

Conclusion: As in previous test.

4. To test significant difference between 2 proportions:

Ho: There is no significant diff. between 2 proportions

Let P1, P2 be observed proportions of success out of n1, n2 ,trials respectively

Z = P1 - P2 P1 - P2 where, q1= 1-p1, q2= 1-p2

SE (p-P) √ (p1q1/n1)+p2q2/n2

= P1 - P2 where P = n1 p1 + n2 p2 & Q=1-P

√PQ (1/n1+1/n2) n1+n2

2. (A) To test the significant diff. bet. Two sample means, when taken from diff. population:
Ho: There is no significant difference between 2 sample means.

Z= x1- x2 where x1 & x2 are sample means with sizes n1 & n2 respectively
SE (x1 - x2)

= x1 - x2 б1, б2 - SD of population
√ (б12/n1+б22/n2)
When б1, б2 are not available use S1 & S2

- 38 -
 = x1 - x2 x1 , x2 – mean of samples of sizes n1 & n2
√ (s12/n1+s22/n2) With SD of S1 & S2 respectively

Conclusion: as in previous Test

2.(B) To test the significant diff. bet. The sample means when taken from same population
Ho: there is no significant difference between 2 sample means
Z = x1 – x2
σ √1/n1+1/n2)

Where б (SD) of population is not known Replace б2 = n1 s12 + n2s22

n1 + n2
= x1 –2 x2 2
√ (s /n +s /n )
1 2 2 1

Exercise 1:

A sample of 300 broilers is taken from a farm and their mean weight was found to be 2.23 kg with
standard deviation 0.24 kg. Verify whether the sample could have been taken from a population with
mean 1.9 kg.

Exercise 2:

The mean milk yield during lactation in 2 farms is estimated as 2176 kg and 2425 kg with standard
deviation 56.2 kg and 3.2 kg by taking a sample of size 50 and 60 respectively. Test whether the milk
yield of the 2 farms differs significantly.

Exercise 3:

Two samples of sizes 35 and 48 are taken from a broiler farm with mean weight in kg as 1.8 and 2.0
respectively with standard deviation 0.72 and 0.56. Test the significance difference in the samples.

- 39 -
Practical No : 7
Title :Test of significance: student ‘t ’test for single mean, difference of mean and
paired ‘t’ test

Objective:

The main objective of ‘t’ test is to study whether the difference between the population mean and
sample mean or between the two sample mean is significant or not when sample size is small i.e. less
than 30

The ‘t’ test of significance is given by W. S. Goset which enables us to make reasonably valid
inferences about the population using statistics from small samples (Less than 30 observations). It is
commonly referred as ‘t’ distribution and the test of significance based on it is called ‘t’ test.

Steps:

 Calculate the population mean and sample mean

 Calculate difference between the population mean and sample mean
 Calculate the standard deviations for both population and sample as per discussed in
practical no 3
 Calculate t value by using following formula

X1- µ
Z= SD/√ n

1. To test the significant of the sample mean from population mean

Step 1: Ho: There is no significant difference bet. Sample mean & population mean
Step 2: Test statistics is given by
t = difference in mean of sample & population
SE of difference
= x- m = x-m with d. f.= (n - 1)
SE(x-m) s/√n
x = mean of sample of size ‘n’ with SD ‘s’ with d. f. = (n - 1)
m = population mean

- 40 -
Step3. Conclusion
i. If È t È < table value of t for (n - 1) d. f. at 5% level‘t‘ is non-significant Ho:
is accepted. we denote it as t = ( )NS
ii. If È t È > table value of t for (n - 1) d. f. at 5% level ‘t’ is significant Ho is
rejectedDenoted as t = ( )*
iii. If È t È > table value of t for (n - 1) d. f. at 1% level ‘t’ is highly significant
Ho is rejected Denoted as t = ( )**

2. To test the significant difference bet. Two sample means when sizes are less than 30 & they
are dependent (Paired‘t’ test)

Two samples are dependent when they have some common factors linking the observations in
the two samples
Step 1. Ho: there is no significant difference bet. 2 sample means
Step 2. Test statistics is given by

t = d–o = d with d.f. = n-1

SE of d s/√n
d = Difference in observation of 2 samples
S= SD of d

Conclusion: As in previous test

3. To test the significant difference bet.2 samples when samples are independent (Non paired or
Unpaired‘t’ test)

By independence of the 2 samples, we mean that there is no relationship bet.2 individuals

contributing the sample, this sample drawn from different populations diff. parts of some population
will be independent.

Step 1: Ho: There is no significant difference between the sample means

Step2: Test statistics‘t’ is given by,

t = Difference in the means of sample

SE of difference
= x1 - x2 with d.f. = n1+n2-2 where s2= (n1-1) s12+(n2-1) s22
√ s2 (1/ n1 +1/ n2) n1+n2-2

x1, x2 are means of 2 samples with SD s1, s2 of steps n1, n2 respectively.

Conclusion: as in previous test (with respect to d.f.)

- 41 -
4. To test the significance of an observed correlation coefficient
Step1: H0: There is no significant correlation
Step2: Test statistics is given by
T= r-0 SE
of (r-0)
= r
2
(1-r ) /(n-2)
= r √(n-2) with d.f.= n-2
√ (1-r )
2

Conclusion: as in previous test (with respect to d.f.)

1. To test the significant of the sample mean from population mean

Step 1: Ho: There is no significant difference bet. Sample mean & population mean
Step 2: Test statistics is given by
t = difference in mean of sample & population
SE of difference
= x- m = x-m with d. f. = (n - 1)
SE (x-m) s/√n
x = mean of sample of size ‘n’ with SD‘s’ with d. f. = (n - 1)
m = population mean

Step3. Conclusion
iv. If È t È < table value of t for (n - 1) d. f. at 5% level ‘t’ is non-significant
Ho: is accepted. we denote it as t = ( )NS
v. If È t È > table value of t for (n - 1) d. f. at 5% level ‘t’ is significant Ho is
rejectedDenoted as t = ( )*
vi. If È t È > table value of t for (n - 1) d. f. at 1% level ‘t’ is highly significant
Ho is rejected Denoted as t = ( )**

2. To test the significant difference bet. Two sample means when sizes are less than 30 & they
are dependent (paired ‘t’ test)

Two samples are dependent when they have some common factors linking the observations in
the two samples
Step 1. Ho: there is no significant difference between two sample mean
Step 2. Test statistics is given by

- 42 -
 t = d–o = d with d.f. = n-1
SE of d s/√n
d = Difference in observation of two samples
S= SD of d

Conclusion: As in previous test

3. To test the significant difference between two samples when samples are independent (Non
paired orUnpaired‘t’ test)

By independence of the 2 samples, we mean that there is no relationship between two

individuals contributing the sample, this sample drawn from different populations diff. parts of some
population will be independent.

Step 1: Ho: There is no significant difference between the sample means

Step2: Test statistics ‘t’ is given by

t = Difference in the means of sample

SE of difference
= x1 - x2 with d.f.= n1+n2-2 where s2= (n1-1) s12+(n2-1) s22
√ s2 (1/ n1 +1/ n2) n1+n2-2

x1, x2 are means of 2 samples with SD s1, s2 of steps n1, n2 respectively.

Conclusion: as in previous test (with respect to d.f.)

4. To test the significance of an observed correlation coefficient

Step1: H0: There is no significant correlation
Step2: Test statistics is given by
T= r-0 SE
of (r-0)
= r
2
(1-r ) / (n-2)

= r √ (n-2) with d.f. = n-2

√ (1-r2)

- 43 -
Exercise 1:
The heights of the 10 males of a given locality are found to be 70,67,62,68,61,68,70,64,64,66 inches.
It is reasonable to believe that the average height is greater than 64 inches. (Table value for 5 %
significance- 1.833 and 1 % significance -2.821)

Exercise 2:

A random sample of 10 boys had following IQs : 70,120,110,101,88,83,95,98,107,100 do this data

supports the assumption of a population mean IQ of 100. (Table value for 5 % significance- 2.262 and
1 % significance -3.250)

Exercise 3: The first clip wool yield of daughters of ram and their dams (in 100 gms) are as follows.
Test the significance difference between the daughter and the dam. (Table value for 5 % significance-
2.262 and 1 % significance -3.250)

Daughter 19 14 21 18 21 17 17 15 8 20
(x)

Dam (y) 9 17 14 16 15 18 9 8 10 11

Exercise 4: Below given the gain in the weight (lbs) of pigs fed on two diets A and B. Test if two
diets differ significantly as regards their effect on increase in weight (Table value for 5 % significance-
2.074 and 1 % significance -2.819)

Diet 25 32 30 34 24 14 32 24 30 31 35 25
(A)

Diet 44 34 22 10 47 31 40 30 32 35 20 23
(B)

- 44 -
Practical No : 8
Title :Test of significance: Chi Square Test

Objective:

This test is used in case of observed frequencies to be tested for their with expected or
theoretical frequencies or to test whether two factors of classification of a set of individuals
present in the form of two-way table are independent or not

Steps:

 The expected frequency for each observed frequency is calculated based on apriori
hypothesis or on the basis of null hypothesis
 The observed frequency data and respective expected frequency data are organized in a
chi square table(2×2 etc) or contingency table
 The degree of freedom is obtained as (r-1)(c-1) where r is number of rows and c is
number of columns in a table.
 (O-E), (O-E)2, ( O-E)2 /E and ∑ ( O-E)2 /E are computed and the calculated chi square
is obtained by using formula
X2 =  (O- E)2
E

1) Chi- test of goodness of fit

This test is performed to test whether the deviation of observed frequencies in a given
data from the expected frequencies are due to real causes or due to chances

The test statistic for goodness of fit is given by

X2 =  (O- E)2
E
O- Observed frequency
E- Expected frequency
d.f. = Total no. of class- 1

In this case
H0: The fit is good or there is no significant diff. between observed and expected frequency

- 45 -
Conclusion:
1. If calculated x2 is less than table x2 for respective degrees of freedom at 5% level x2 is not
significant which is denoted by x2= ( )NS i.e. H0 is accepted. Fit is good.
2. If calculated x2 is > table x2 for the respective d.f. at 5% level x2 is significant & denoted by x2
= ( )*. H0 is rejected. The fit is not good or theoretical frequencies are not according to the
theory.
3. If calculated x2 is > table x2 for the respective d.f. at 1% level x2 is highly significant and
denoted by x2= ( )**. H0 is rejected. This fit is not good or theoretical are not according to the
theory.

Chi- square test of independence

This is performed when the data is presented is form of contingency table. A table giving the
simultaneous classification of body of data in two different ways is called a “contingency table”. If
there are ‘r’ rows & ‘c’ columns, the table is said to be an ‘r’x’c’ contingency table x 2 test is applied to
test whether the factors classified are in dependent or not i.e. the 2 factors are associated or not. The
d.f. for ‘r’x’c’ table is (r-1) (c-1)

Application of x2 statistics in a 2x2, X2 contingency table

Level 1 Level 2 Total

Level1 A b a+b
Level 2 C d c+d
Total a+c b+d a+b+c+d=n

Step 1: H0:The factors are independent

Step 2: x2 is given by
X2 = (ad-bc)2 x n with d.f. = (2-1) (2-1) = 1
(a+b) (c+d) (a+c) (b+d)

Conclusion: As in case of x2 test of goodness of fit (with resp. d.f.)

The above formula is applicable when a, b,c,d are > 5. If one of more is less than 5, Yate’s
correction of continuities is to be applied as follows:
X2 = (|ad-bc|- n/2)2 x n
(a+b) (c+d) (a+c) (b+d)
In a (rxc) contingency table the expected value(E) in ith row & jth column is calculated by
(Ri* Cj)/N
Where, Ri = Sum of all values in ith row
Cj = Sum of all values in jth column

- 46 -
 N = Grand total i.e. the sum of all values in given contingency table
Then,
x2=  (D-E)2
E
Conclusion:
As in previous test (with respective d.f.)

Exercise 1: The following table gives the classification of 400 plants according to the nature of leaves
and flower color. Test whether the frequencies are in the ratio 9:3:3:1(Table value for 5 %
significance- 7.815 and 1 % significance -11.345)

Blue flower and flat Leaves 234

Blue flower and crimpled Leaves 63

White flower and flat Leaves 76

White flower and crimpled Leaves 27

Exercise 2:

In the farm of 250 calves were born during a particular time. The number of male calves is 135. Test
whether the sexes are equally born (Table value for 5 % significance- 3.841 and 1 % significance -
6.635)

- 47 -
Practical No 9
Title : Test of significance: ‘F’ test

Objective:

This test is mainly used to test the significance of difference between the variances of the two samples

i. H0: There is no significant difference between variables of 2 samples

ii. The ‘F’ statistics is given by

F= Estimated variance of first sample

Estimated variance of second sample

= (S1)2 with d.f. = (n1-1), (n2-1)

(S2)2

(S1)2 & (S2)2 are the variance of the sample of size n1, n2 respectively.
F has 2 degrees of freedom, one for highest variance & another for smaller variances.

NB: We have to put the greatest of variance (S12, S22) in the numerator

Conclusion:
i. If cal |F| < tab F for d.f. = (n1-1), (n2-1) at 5% level F is not significant. H0 is accepted.
ii. If cal |F| > tab F for d.f. = (n1-1) , (n2-1) at 5% level
F is significant F = ( )* H0 is rejected.
iii. cal |F| > tab F for d.f. = (n1-1), (n2-1) at 5% level
F is highly significant F = ( )** H0 is rejected.

- 48 -
Exercise 1:

Test whether the variances are significantly different following samples (5 % significance) (Table
value for 5 % significance is 4.147)

Sample I 45 46 49 25 17 18 13 56 58

Sample II 47 49 43 27 29 38 37 - -

Exercise 2:

The random samples were drawn from two normal populations and their values are given below. Test
whether the two populations have same variance at the 5 % level of significance. (Table value for 5 %
significance is 3.072)

Sample I 66 67 75 82 84 88 90 92 - - -

Sample II 64 66 74 78 82 85 87 92 93 95 97

- 49 -
Practical No : 10
Title : Completely Randomized Design (CRD)

Objective:

This design is used for comparison between two or more groups. More specifically this design is
described as one way classification with fixed effects.

This is simplest of all experimental designs. This is the design in which the treatments are
assigned completely or random to the experimental units or vice-a-versa i.e. it imposes no restrictions
on the allocation of treatments to the experimental units. CRD is preferred when all the experimental
units considered for the experiments are known to be homogeneous. Any number of experimental
units and treatments can be utilized in this design. As the design is highly flexible and simple, the
CRD is widely used. Analysis is simple, even if certain values are missing. The experimental units
will be allotted to the different treatments by using random number table.

Collection and analysis of data:

After having randomized the experimental units over different treatments, initial recordings (if any) of
the experimental units are noted against each experimental unit. e.g. Incase of weight gain study,
initial weights are to be recorded. Then the experimental units are subjected to respective treatments
and after the experimental period, the response values will be observed.

Data of response values:

Tr1 Tr2 ………… Trt

Y11 Y12 ….. Y1t

Y2t Y2t …. Y2t

. . …..

. . …..

Yn1t Yn2t Yntt

T1 T2 Tt

Let us have ‘t’ treatments each having replications n1, n2, ....... nt then

n1+n2+ ..... +nt = N.

- 50 -
Stepwise procedure:

H0 : there is no significant difference in the treatments.

Step 1: a. Calculation of treatment total T1, T2 , …… Tt

b. Calculation of grand total (G.T.) = T1 + T2+…+ Tt

c. Calculation of correction factor (C.F.) = (G.T.)²/N

Step 2: Calculation sum of squares

a. Total sum of squares (TSS) = y11 ² + y12² +…..+ynt ² – CF

b.Treatment sum of squares (TrSS) = T1²/ n1+ T2²/n 2 + …+ Tt ²/ nt–CF

Error sum of squares (ESS) = TSS – TrSS

Step 3: Formation of analysis of variance table

ANOVA table or AOV table:

Source of Degree of Sum of squares MeanSquares F

variation freedom (d.f.) (M.S.)
(S.V.)
SS/d.f.

Treatments (t-1) TrSS TrMS =TrSS

(t-1)
TrMS
EMS=ESS
Error (N-t) ESS (N-t) EMS

Total (N-1) TSS

Error D.F. = Total D.F. – Treatment D.F.

= (N-1) – (t-1) = N-1 – t-1

= N-t

Step 4: Initial decision:

If calculated F< table value of F for t-1, N-t d.f. at 5% level, F is not significant. All treatments are
alike.

- 51 -
If calculated F>tab F for (t-1) (N-t) d.f. at 5% level, F is significant. F= ( )°. H0 is rejected.

If calculated F>tab F for (t-1) (N-t) d.f. at 1% level is highly significant. F= ( )°°. H0 is rejected.

Step 5:

Critical difference= √EMS (1/ n1+1/ n2) x t error d.f. at 5% (1%)

Treatment sum of squares = [(T1²+ T2² + ….+ Tt²) /n] – CF

C.D. between any 2 means at 5% or 1% level = √2Ems/n x table t for error d.f. at 5 % or
1% level treatment

Example No.1
The 4 different kinds of diets are fed to 5 chickens each and weight gain measured is given below.
Analyze the data (Table value at 5 %- 3.24 and 1% - 5.29)

Feed 1 Feed 2 Feed 3 Feed 4

11 8 12 6

12 9 10 5

13 7 9 4

15 15 14 7

14 16 11 8

Example No.2
A set of data involving 4 feed stuffs A.B.C.D tried on 20 chicks is given below .all 20 chicks are
treated alike in all respects except feeding treatments is given .Analyze the data (Table value at 5 %-
3.24 and 1% - 5.29)

Feed 1 Feed 2 Feed 3 Feed 4

55 61 42 169

49 112 97 131

42 30 81 169

21 89 95 85

52 63 92 154

- 52 -
Practical No : 11
Title : Randomized Block Design (RBD)

Objective:

This design can be made use of when the experimenter finds that the available experimental units
can be grouped into different homogenous blocks each containing units as many as the numbers of
treatments or as a multiple of the number of treatment. The blocking has to be done on the basis of any
observable character, which is likely to have influence over the factors under study. e.g. in the study of
comparative effects on weight gain in chicks, the chicks may be grouped according to strains .If we
are to conduct a randomized block design to try ‘t’ treatments, we need ‘bt ’no. of experimental units,
where b= no. of blocks and t=no. of treatment. Each block is a replication by it self and no. of
replication =no. of blocks =b. This is restriction in the case of RBD. RBD is an equi-replicated design.

In this design, we make use of principle of local control with the idea of reducing the
variability due to experimental error. First we arrange the experimental units in to homogenous blocks
such that within the blocks the experimental units are as far as possible homogeneous and between the
blocks there is variability. RBD is a better design than CRD, as this will have greater precision of the
estimates and greater efficiency of designs.

Randomization:

Having formed the no. of blocks, experimental units to the different treatment will be
allotted independently for each blocks randomly and response value will be tabulated as follows;

Treatment Tr1 Tr2 …………. Trt Total

1 Y11 Y12 Y1t B1
2 Y21 Y22 Y2t B2
……
…..
B Yb1 Yb2 Ybt Bt
Total T1 T2 Tt GT

Stepwise procedure:

H0 = Treatment are those which do not differ significantly

Step 1: a: Calculation of treatment total i.e. T1, T2…Tt

b: Calculation of block total i.e. B1, B2,….Bt

- 53 -
 c: Calculation of grand total (GT)= T1+T2+……+Tt

=B1+B2+…..+Bt

= sum of the treatment total

= sum of the block total

d: Calculation of correction factors (C.F.)= (GT)2 /n ( as N= bt)

Step 2 : calculation of sum of squares

a: Total sum of square (T.S.S.) =∑∑yij2 – C.F. I.e. sum of square of all the b x t values –C.F.

b: Treatment sum of squares (TrSS)= T12+T22+….+Tt2 - C.F.

c: block sum of squares ( B.S.S)= B12+ B22+……Bt2 – C.F.

d: error sum of squares (ESS) = TSS – (TrSS + BSS)

Step 3: ANOVA table or AOV – Analysis of variance.

1. Source of variation

2. Degree of Freedom

3. Sum of Squares

4. Mean Square

5. F value

Source of Degree of Sum of squares Mean square F

variation freedom
Treatment (t-1) TrSS TrMS=TrSS TrMS
(t-1) EMS
Block ( b-1 ) BSS BMS=BSS BMS
(b -1) EMS
Error (b-1) (t-1) ESS EMS= ESS
(b -1)(t-1)
Total (bt – 1 ) TSS

- 54 -
Step 4: Initial decision

1. If calculated F < table value of F for t-1, N-t d.f .at 5% level , F is not significant. H0 is
accepted. All the treatments are alike.

2. If calculated F > tab F for (t-1), (N-t) d.f. at 5% level. F is significant. H0 is rejected.

3. If calculated F > tab F for (t-1), (N-t) d.f. at 1% level. F is highly significant. H0 is rejected.

In the last two cases we have to calculate critical difference between any two treatment
means at 5 % and (1%) level

=√2EMS/b x table value of ‘t’ for error d.f. at 5% (1%) level

Bar chart representation will be done to show the significance between treatments.

Example No.1. 4 diets were fed to 5 chickens each and weight gained by each is given below.
Test the differences of diets.( Table value at 5 % = 3.24, 1 %=5.29)

Feed 1 Feed 2 Feed 3 Feed 4 Total

1 11 8 12 06
2 12 9 10 5
3 13 7 9 4
4 15 15 14 7
5 14 16 11 8
Total

Example No.Effect of protein supplement on average dairy milk of cow is given below. Cows
have been put into 6 blocks. Process the data and give its significance.
(Table value at 5 %- 5.18 and 1% - 7.56)

Feed A Feed B Feed C

1 10.4 12.6 9
2 10.5 12 9.7
3 5.9 11 12
4 6.7 8.8 9
5 8.0 9 8.9
6 6.7 12 10

- 55 -
Practical No: 12
Title: Computer basics and components of computer, simple operations, entry of biological data

‘Computer’ in simple terms can be defined as “any electronic data processing machine to
which we give data (e.g.: name, age, salary details, etc) as well as instructions (e.g.: add, multiply,
calculate, etc) on which it processes and gives us information called output” (in the desired form on
the output devices as per the instructions coded in the program.

Man has invented many electronic devices but the computer is the most powerful and versatile
device. It is affecting our lives in many ways. Earlier computers were used for scientific and
engineering computation. But 75% of work done by the computers is non-computational type. Today
computers are used in many different ways. Modern computers possess many interesting, important
and useful characters

Characteristics of computers:

 Speed: The computer is a fast working machine. It is speedy and can perform millions of
operations in one nanosecond. A powerful computer is capable of doing millions of operations
per second.

 Accuracy: A computer is always very accurate. The accuracy of a computer is consistently

high and the degree of accuracy depends on its design. The computer can work accurately if

1. The source of output data is authentic

2. The data entry is correct

3. The instructions are accurate

The computer commits a mistake only if the given data is wrong. Thus, there is no doubt that the
computer is very efficient and reliable. However, its efficiency and reliability depend upon a number
of external factors. Sometimes errors may occur in the results only because of human.

 Diligence: Unlike human, the computer is free from tiredness, lack of concentration and it can
work for hours together without getting tired and making mistakes. In repetitive kind of work,
computer scores over man. It reproduces the same result for the same job irrespective of how
many times it is executed.

 Versatility: Versatility is one of the most wonderful things about the computer. One moment it
is preparing the result of a particular exam then the next moment it is busy in preparing the
electricity bill.

- 56 -
  Power of remembering: Computer has the capacity to store a large amount of data and
information and supply the stored information when asked for. Every piece of information can
be retained as long desired by the user and can be recalled as and when required. A computer
can forget or use certain information only when it is asked to do so. So it is entirely up to user
to make a computer retain or forget particular information.

 Automation: Computer is an automatic machine. Once we give appropriate information to a

computer it performs all the necessary tasks until it has instructions to stop.

 No IQ (Intelligence Quotient): Computer is not a magical device. It possesses no intelligence

of its own. Its IQ is zero. Till today it has to be told what to do and how to do and in what
sequence. Hence, only the user will determine what task the computer will perform.

 Efficiency: The computer can work for a long time without getting tired with the same
efficiency. It does not have feelings to express.

 Power of making logics: The computer can take logical decisions if the required data is
provided to it. This work is carried out by ALU ( Arithmetic Logic Unit)

Organization of computer:

All computer systems perform five basic operations as follows:

I. Input: The process of entering the data and instructions into the computer is called inputting.

II. Store: Saving the data and instructions into the computer memory and recall the saved data
when we want to make use of it.

III. Process: Performing arithmetic and logical operations or results for the data given on the data
used for our instructions in order to convert the data into useful information.

IV. Output: The process of producing useful information or results for the data given by the user in
the form of printed copy or visual display.

V. Control: Directing the manner and sequence in which all the above operations are performed.

Classification of Computer:

Depending upon the size and cost of the computers, they are classified as microcomputer,
mainframe computers and semi computers

 Microcomputers/ Micros:

Microcomputers are a typically single user system, which means that only one user can
use it at a time. Microcomputers have too low data storage capacities (500 MB to 2 GB). Their
processing power is also limited in terms of the number of instructions that they can process/

- 57 -
 second. Therefore they are not suitable for the applications that require large storage capacities.
Microcomputers have revolutionized the computer industry itself due to their size and cost.
They have limited input and output capacities, little software and little storage capacity. On the
other hand, they are cheap and easy to use. The most common applications of the
microcomputers are word processing, spreadsheet, calculation and database management. e.g.:
1 PMPC, Apples, Mackintosh and HCL

 Minicomputers/Minis:

These machines were introduced in 1960. These are larger in size, are much more
expensive than micros. Normally they are designed to support more than one user at a time.
They possess larger storage capacities and operate at higher speeds. These computers are used
when the volume of processing is larger. e.g.: Data processing for medium sized organization.
These are also used in server in Local Area Network. e.g.: digital equipment PDP, 11/45 and
VAX11.

 Mainframe computers:

Mainframes are more powerful than minis. They operate at very high speeds, have very
large capacities and operate at higher speed and can support hundreds of users.

 Super computer:

Super computers are the fastest and most expensive machine used for commercial
purposes. There are less than 500 conventional supercomputers in the world. Supercomputers
are used for weather forecasting, biochemical research, in medical field, to study the structure
of viruses, etc.

PC Family:

The PC family was introduced by IBM (International Business Machine). The most successful models
are PC, PC-XT and PC-AT

 PC (Personal Computer): Designed using 8088 microprocessor or chip memory capacity is 640
KB (RAM). No hard disk but consists of 1 or 2 floppy device drives. Speed is 8 MHz, runs on
DC.

 PC-XT ( Personal Computer with External technology): Has stored hard disk. IBM in 1983
used 8088 microprocessor or introduced chip. PC-XT memory capacity is 640 KB, speed 10 to
12 MHz. It is also single used O.P. which uses DOS.

 PC-AT( Personal computer with Advanced Technology): 1984, uses 80256, is 10 bit microchip
memory

- 58 -
  PC-AT(386): It was introduced in 1985. It uses 80386 microprocessor unit chip. Supports
multiprogramming and multitasking facilities. Its memory capacity is 4 MB and plus it
includes 1 or 2 floppy drives. Its computing speed is 66 MHZ or higher.

 PC-Pentium: Recent version of PC family. Uses 64 bit microprocessor chip. Its memory
capacity is 8 MB and it includes 1 or 2 floppy drives. Computing speed is about 100 MHz and
plus.

Block Diagram of Computer:

Computers perform arithmetic and logical operations on input data produce output which constitutes
the hardware of the computer

Input devices Output Devices

Memory unit
Mouse, key board, Monitor, printer,
scanner plotter
Control Unit

Arithmetic Unit

Input Devices: The input devices convert the information into suitable binary form and send it to the
processing unit. e.g.: Optical Mark Reader(OMR), Optical Character Reader(OCR), Magnetic Ink
Character Reader (MICR), Touch screen, Joystick.

Keyboard: Consists of

 Alphanumeric keys- letters and number keys

 Numeric keypad- Contains number and cursor control key, num lock key

 Cursor control keys are marked with arrows

 Function key from F1 to F12

 Some special keys like print screen, scroll lock, insert, delete, home, end, control after key are
used in combination with keys for specific purposes.

 Mouse: A mouse in addition to the keyboard

- 59 -
CPU: Brain of the computer system (microprocessor) the controller or the control unit. It requires data
instructions stores them temporarily and processes the data as per instructions. CPU is formed of three
units- memory unit, ALU, Control unit

 Memory Unit: There are two types of computer memory

 Primary/ Internal memory: RAM, ROM

 Secondary/ External memory:

 Magnetic tapes and cassettes

 Magnetic disc, floppy disc and hard disk

 Optical disk

Output Devices: Monitor, Printer

Printer:

 Impact printer: Dot matrix, Daisy Wheel, Line printers

 Non impact printer: Inkjet printer, Laser printer, plotters

Kinds of Softwares:

 Pre-written software:

 Application Software

 System Software

 Custom built software: O.S. is a set of software programs that controls overall operations of
the computer system.

Functions:

 Input/ output device management

 Memory management

 CPU management

 Disk management

 User interface

- 60 -
Exercise: Give the importance of the computers in Veterinary Science.

- 61 -
Practical No : 13
Title : ‘Internet and e- mail, entering and saving biological data through

MS- office, MS – Excel

email (or e-mail) is defined as the transmission of messages over communications networks.
Typically the messages are notes entered from the keyboard or electronic files stored on disk. Most
mainframes, minicomputers, and computer networks have an email system.

Electronic mail (email or e-mail) is a method of exchanging messages between people using
electronics. Email first entered substantial use in the 1960s and by the mid-1970s had taken the form
now recognized as email. Email operates across computer networks, which today is primarily
the Internet.

Email consists of messages which are sent and received using the Internet. There are many
different email services available that allow you to create an email account and send and
receive email and attachments, many of which are free.The general format of an email address is local-
part@domain, and a specific example is jsmith@example.com. An address consists of two parts. The
part before the @ symbol (local-part) identifies the name of a mailbox. This is often the username of
the recipient, e.g., jsmith. Webmail's primary difference from client-based email is how you access
it.Webmail is accessed on the Internet through a Web browser while client-basedemail is accessed
through a desktop program. ... Webmail can be accessed from any computer, making it convenient
when you don't have your computer on-hand

Entering the data into excel sheet:

1. Insert a new Excel worksheet, and name it Sports.
2. Browse to the folder that contains the downloaded sample data files, and
open OlympicSports.xlsx.
3. Select and copy the data in Sheet1. ...
4. On the Sports worksheet, place your cursor in cell A1 and paste the data.

How to Sort in Excel

1. Highlight the rows and/or columns you want sorted.
2. Navigate to "Data" along the top and select "Sort."
3. If sorting by column, select the column you want to order your sheet by.
4. If sorting by row, click "Options" and select "Sort left to right."
5. Choose what you'd like sorted.
6. Choose how you'd like to order your sheet.

- 62 -
Follow these steps to decide how to organize your data into tables:
1. Name your database. ...
2. Identify the objects. ...
3. Define and name a table for each object. ...
4. Identify the attributes for each object. ...
5. Define and name columns for each separate attribute that you identify in
Step 4. ...
6. Identify the primary key.

- 63 -