Role of Statistics in Field Sciences

CHAPTER –III
DATA PRESENTATION
Introduction
After collection and editing of data, still found very voluminous and unwieldy, and needs to be
organized and presented in meaningful and readily comprehensible form in order to not only draw
inference from the data but to facilitate further statistical analysis.
There are three ways of presenting data - Textual, Tabular and Diagrammatic/Graphic.
A. Textual Presentation: In textual presentation, data is presented along with the text. For
comprehension, one has to read through and scan the whole text in order to grasp its meaning and
implications, e.g. the population census of Pakistan 1998 revealed that male dominate the population
of Pakistan.
B. Tabular Presentation: In tabular representation, data is arranged in a systematic way in rows and
columns. The huge and unwieldy raw data can be condensed in a table form through the process of
classification. “Classification is the process of arranging the data into homogenous groups.” According
to Connor, “Classification is the process of arranging things (actually/ notionally) in classes according
to their resemblances and affinities, and gives expression to the unity of attributes that may subsist
amongst a diversity of individuals.” Thus, through the process of classification, we arrange the
heterogeneous elements of the collected data into homogenous sub-groups i.e. the elements of the
same characteristics in one class. When the whole data is divided into a number of classes, it does not
mean that now they are fit for the purpose of comparison, interpretation or drawing inference from
them at a glance.
Aims of Classification:
1. To reduce the large sets of data to an easily understood summary
2. Displaying the points of similarity and dissimilarity.
3. To save mental strain by eliminating unnecessary details,
4. To reflect the important aspects of the data.
5. To prepare the ground for comparison and inference.
Basic Principles of Classification: while classifying large sets of data, the following points
should be kept in mind:
1. The classes, in which the data are to be divided, should be mutually exclusive and no
overlap should exist between successive classes.
2. The classes should be all inclusive.
3. Conventional classification procedure should be adopted.
4. The classification procedure should not be so elaborative to lead to trivial classes.
Characteristics: Classification of the data largely depends upon the nature, scope and purpose of
the investigation. These are:
1. Classification should be Unambiguous: There should be no ambiguity in classification, if it
is so the purpose would die because it means for filtering down the ambiguity. Classes should
be defined in such a way that there should be no confusion in the mind of the reader, e.g. in
census, the population is divided into two classes, e.g. literate and illiterate in 1961.
2. Classification Should Be Stable: An ideal classification should have the advantage of being
stable for the purpose of comparison or otherwise a serious problem.
3. Classification should not be rigid: Classification should be flexible and not rigid. It should
have the capacity of adjusting itself to the new situations and circumstances. No
classification remains stable forever because with the passage of time some classes have to
be dropped because they will become obsolete while some fresh classes have to be added. An

1
 Role of Statistics in Field Sciences

ideal classification is one which is capable of adjusting itself to these changes and sustains its
stability. For this, the data is first divided into some major classes, and then each class is sub-
divided and changes may be made in the sub-classes and not in the major classes.
Types of Classification: Classification of the statistical data are made on the basis of the
characteristics possessed by individuals in different groups of units of a universe, and the types of
classification depends upon the characteristics of the statistical data- descriptive/ numerical.
I. Descriptive Characteristics: When the data are classified on the basis of the descriptive
characteristics according to attribute or qualitative variable, e.g. poverty, sex etc. Descriptive
variable cannot be measured quantitatively but their presence or absence can be observed.
II. Numerical Data: When the data are classified on the basis of numerical characteristics
according to class intervals or quantitative variable or simply a variable. A variable is a
variant or a measurable quantity which varies from one value to another. A variable which
can assume any of a prescribed set of values is called the domain of a variable. If the variable
can assume only one value, then it is called constant, e.g. age, income, height.
III. Continuous and Discrete Variable: A variable can either be discrete or continuous. A
variable which can theoretically assume any value between two given values is called
continuous variable, e.g. the height of a body at various ages, thus measurements give rise to
continuous data. Whereas, discrete variable is one which is incapable of taking all possible
values, and thus, there are gaps between one value and the next, e.g. number of accidents in a
month, kids in a family.
IV. Classification According to Attributes:
Simple Classification: According to this method, only one attribute is studied which is
further sub-divided into two classes and no more. Thus, all those individuals which possess
particular characteristics are placed in one group and those which do not possess that
particular characteristics are placed in another group, e.g. literacy.
Manifold Classification: Sometimes it is required to study more than one attribute. Then the
whole data are divided into a number of classes, e.g. if we study the problem of literacy sex-
wise, then we will study two attributes i.e. one is literacy and the other is sex. There will be
two possibilities a person can either be literate or illiterate, secondly it can either be male or
female. So it will give birth to 04 classes such as Literate male and illiterate males and
literate female and illiterate females.

TABULATION:
The systematic arrangement of the statistical data in columns and rows is called tabulation or a
systematic presentation of data classified under suitable heads and sub-heads by placing in
columns and rows to make the data easy to understand, facilitates comparison and provides an
effective way to convey information to a reader. It involves the systematic presentation of data to
elucidate the problem under investigation. It is a process between the collection of data on the one
hand, and its final analysis on the other. According to Connor, “Tabulation involves the orderly
and systematic presentation of numerical data in a form designed to elucidate the problem under
consideration.” Thus, tabulation is meant to answer the questions relating to the investigation in
an arranged form which is helpful in drawing the inferences from them. According to Buddington,
“The tabulation of result is the first stage of the work of compilation, and subsequently becomes
the basis of consideration, deduction and application to the present and future problems.”

2
 Role of Statistics in Field Sciences

For tabulation of data, it should be decided beforehand that what type of tabulation should be
done as the type of tabulation depends on the nature, scope and object of the inquiry. The
tabulation of the data should make as such that it suits the nature and object of the investigation, if
is not suitable; its analysis naturally would be defective.
Types of Tabulation According to Purpose: Statistical tables are classified according to purpose
1. General Purpose (Primary): These are large in size, are extensive with vast coverage and are
constructed for reference purposes.
2. Specific Purpose (derived or text): Simple in structure and deal with one or two criteria of
classification only.
Types of Tabulation: Tabulation may be simple or complex depending upon the nature of information.
Simple tabulation gives information regarding one or more independent questions while complex
tabulation gives information regarding two mutually dependent questions.
One-way Table: This table gives the information regarding one characteristic of the data and independent
of each other, e.g. Table-1
One-way Table: 1
Provinces Population (Millions)
Punjab 90
Sindh 65
KPK 50
Balochistan 11
FATA 7

Two-way Tables: These tables give information regarding two mutually dependent questions, e.g. if in the
above table the population of the provinces is divided sex-wise, it would become a two-way table and it
would give information regarding two mutually dependent questions, e.g. Table-2
Two-way Table: 2
Provinces Population (Millions)
Total Male Female
Punjab 90 47 43
Sindh 65 34 31
KPK 50 30 20
Balochistan 11 6 5
FATA 7 4 3

Three-way Tables: Those tables which give information regarding three mutually dependent and
interrelated questions, e.g. the above tables can be converted into three-way table if a new question is
added to it if the literacy is also added, then the table takes the form; Table-3
Three-way Table: 3
Population (Millions)
Provinces Males Females Total
Literate Illiterat Tota Literate Illiterat Total Literate Illiterat Tota
e l e e l
Punjab 25 22 47 30 13 43 40 50 90
Sindh 22 12 34 20 11 31 30 35 65
KPK 20 10 30 12 8 20 30 20 50
Balochista 2.5 3.5 6 1 4 5 4 7 11
n
FATA 1.5 2.5 4 0.5 2.5 3 2 5 7

3
 Role of Statistics in Field Sciences

High Order Tables: High order tables are those which can provide information about a large number of
inter-related questions. High order tables may be of four-way, five-way, etc and such tables are also called
manifold tables.
High Order or Mani-fold Table: 4
Population (Millions)
Provinces Males Females Total
Literate Illiterat Tota Literate Illiterat Total Literate Illiterat Tota
e l e e l
Punjab 25 22 47 30 13 43 40 50 90
Sindh 22 12 34 20 11 31 30 35 65
KPK 20 10 30 12 8 20 30 20 50
Balochista 2.5 3.5 6 1 4 5 4 7 11
n
FATA 1.5 2.5 4 0.5 2.5 3 2 5 7

Rules for Tabulation: Following are the general rules for construction of tables.
a. The tables should be self-explanatory and easy to understand without footnotes.
b. Must have title and should be written above the table in block letters.
c. One table should not be made for all information as this would increase the chances of errors.
d. Thick lines for one class of data from the other, and thin lines for sub-divisions of classes.
e. Main headings should be few in number so that they may not confuse the mind.
f. The columns of comparable data should be kept side by side.
g. Approximation of the data should be made before tabulation.
h. Captions (columns) and Stub (headings of Rows) should be self-explanatory.
i. The unit of measurement under each heading should be clearly mentioned.
j. The total of the rows and column should be total row at right and columns at the foot.
k. For purpose of comparison, items should be arranged either alphabetically, geographically,
chronologically or according to size and importance in order to facilitate comparison.
l. If some portion of the collected data cannot be classified into any class, should be tabulated
separately under the heading miscellaneous.
m. Each table constructed should have separate heading, should clearly tell what this table presents.
n. If certain figures are to be given some importance, they should be placed in a box or circle.
Sketch of a Table:
Title: Table Showing Population of Pakistan (1981).
Prefatory Notes:
Box-head Columns Caption
Unit
STUB Body

Footnotes: Source Notes:

Advantages of Tabulation:
a. The main advantage of tabulation is that a mass of data which is confusing to the mind is
presented in a logical sequence giving the shape of statistical tables which answer all the
questions of the problem under investigation.
b. Tables give a bird’s eye view of the whole mass of data in a concise form which takes less
time in drawing inferences from them.
c. Tables pave a ground for the analysis of the data.
d. The errors and omission if any, are found in tabulation.

4
 Role of Statistics in Field Sciences

3. DIAGRAMATIC/GRAPHIC PRESENTATION OF DATA

The main object of the science of statistics is to simplify the complexity of the quantitative data
and to make it easily intelligible. Statistical methods like classification, tabulation remove the
complexity and simplify the statistical data by putting it in a systematic order so that there is no
difficulty in understanding them. Generally, figures are neither interesting to all nor everyone is
competent to compare them and observe their differences, therefore, it necessitates to have such a
device to presents the quantitative data in such a manner that even a layman may be in a position
to understand and compare the quantitative data.
A. Diagrammatic Representation
Diagrams represent the dry statistical figures by attractive, appealing and beautifully constructed
diagrams that’s why common man always avoid figures and like pictures, graphs and charts for
data representation. As diagrams are being a visual display, so leave long lasting impression on
the mind of a reader. By having just a look on diagrams and graphs, one can draw conclusion at a
glance, because no one neither bother about the figures nor to do some quite mental arithmetic
work. Diagrams are appealing to both eyes and mind and give a bird’s eye view of the collected
data and are considered as visual aids which are helpful for the purpose of comparison. According
to Fred R. Barnard (1927), “A picture is worth ten thousand words” or a Chinese proverb ascertains its
importance as “Picture is worth than 1000 words.”

Besides this, diagrams have some disadvantages such as;

a. Diagrams are less accurate than tables.
b. Diagrams money and time and the amount of information conveyed is limited.
c. Diagrams are excessively used in business and administration.
Some of the important aspects of the diagrams are listed as:
1. Attractiveness: Diagrams are attractive while figures are avoided. One may not like to devote
even a minute to study the numeric, because to him they are dry and boredom. If the same
figures are represented by neat and beautiful diagrams, will attract the readers. The readers
would try to skip over the figures and see the diagrams happily as a consequence of
simplicity and attractiveness.
2. Effective Impression: It is authenticated fact that diagrams leave long lasting effect on the
minds of readers than the figures. The figures studied may be soon out of his mind and may
forget the figures but the diagrams and pictures have strong effects on the mind and may take

5
 Role of Statistics in Field Sciences

some time to forget. Advertisers and propagandists win the public by advertising their
business through pictures as they are considered as mind catching.
3. Intelligible: Attractive and impressive diagrams are readily intelligible, e.g. if a man has
never counted more than a few hundreds of anything, may not imagine how large the city is?
If he is told that the population of the city A is 20 lacs he would not understand but if a circle
is ¼ parts shaded and the rest is left blank, and the reader is told that the shaded portion is the
population of the city, he would understand it at first glance.
4. Comparison: By drawing diagrams and graphs, comparison of the data becomes easier.
5. Time Saving: Diagrams save time because no efforts are required to understand the diagrams
which are otherwise needed in grasping the significance of the numerical figures.
6. Helpful to Economists: Economic theories such as law of substitution / law of diminishing
utility can be properly explained with the help of diagrams and pictures. An economist can
make his idea rich, impressive and attractive if it is displayed with beautiful pictures. A
speaker addressing the audience, if giving numerical figures, would bore the public with dry
figures, while another speaker can make the same topic interesting and impressive by
explaining the facts through diagrams, pictures and graphs.
7. Helpful to Business: Diagrams are useful to businessman because he has hardly time to
devote the study of huge figures and to think and draw inferences from them. If the same
facts and figures are represented to him in the shape of diagrams, it will take few minutes to
grasp the significance of the data.
Requisites of a Good Diagram:
1. Diagram should be attractive, neat, drawn artistically and scientifically according to scale and
must be consistent with the size of paper available and the size of the data to be represented
and the scale must start at zero for the purpose of comparison.
2. Diagram should be precise and complete in all respects to give a bird’s eye view of the data
and for effective impression made comparison easy.
3. Minimum cost over diagram for maximum utility.
4. Proper heading.
5. Diagram should be represented in such a way to avoid necessity of reading foot-note.
6. Important points should be emphasized by the use of different colours.
Limitation of Diagrams:
1. As diagrams give approximate results, so these diagrams should be used in those places
where the purpose is to give a rough idea of statistical data. As compare to layman,
statistician is not of much use for further analysis of the figures.
2. As diagrams need technical hands for construction to present a clear picture, but if
constructed wrongly, it would give fallacious conclusion.
3. When the same data is represented in the same unit by diagrams, then comparison is possible
or other-wise.
Rules for Drawing Diagrams:
There is no such hard and fast rules that can be laid down for drawing diagrams but still some
points are of valuable importance for the construction of a diagram;
1. General: Data, for which the diagrams are represented, should be kept near the diagram or on
a page facing diagram.
2. Heading: A suitable, complete and short heading should be given at the top of the diagram.
3. Size: The size of the diagram should be such to look beautiful i.e. neither be small nor too large.
4. Scale: The selection of the scale should be such that the details of the diagrams should spread
it. The scale should be mentioned either on the right hand top or the left hand bottom.

6
 Role of Statistics in Field Sciences

5. Body: The body of the diagram should be constructed neatly and accurately i.e. colours, cross
hatching, dotting and etc. should be used.
6. Proper Type; A diagram construction should be proper according to the given data. A proper
diagram should be impressive and attractive.
7. Guide: At the bottom or at the top, symbols used should be explained.

Forms of Diagrams: Diagrams are represented by any one of the following forms:
1. Dimensional Diagrams: As the name suggests, these types of diagrams are drawn by taking
into account the dimensions, i.e. length, breadth and thickness. These are;

a. Linear / One Dimensional Diagrams: A diagram in which the size of only one dimension
i.e. length is fixed in proportion to the value of the data is called one dimensional diagram.
These diagrams can be drawn in both vertical and horizontal manner. The related different
bar diagrams differ from each other only in respect of their length dimension, while they
remain the same in respect of their other two dimensions i.e. breadth and thickness. The size
of breadth of each of such diagrams is determined taking into consideration the number of
diagrams to be drawn, and the size of the paper at one’s end. The lengths of the lines / bars
are in proportion to the different figures they represent. Generally, bars are drawn in one
dimensional diagram because they give better visual effect than the line.
Advantages:
i. The chief advantages of a bar diagram can be outlined as under:
ii. It is very simple to draw and read as well.
iii. It is the only form of diagram which can represent a large number of data on a piece of
paper.
iv. It can be drawn both vertically and horizontally.
v. It gives a better look and facilitates comparison.
Disadvantages
i. It cannot exhibit a large number of aspects of the data.
ii. The which of the bars are fixed arbitrarily by a drawer.
i. Simple Bar Diagrams: A simple bar chart is used to represent data of only one variable
classified on a spatial, quantitative or temporal basis. In a simple bar chart, we make bars of
equal width but variable length, i.e. the magnitude of a quantity is represented by the height or
length of the bars. As in bar diagrams, one bar represents only on figure, therefore, the number
of bars /lines are depended upon the number of figures. e.g. Number of students in various
Universities in one year.

7
 Role of Statistics in Field Sciences

Construction: First, the data is arranged in ascending or descending order of magnitude.

Then, the scale is adjusted in such a way that the largest bar can be accommodated easily on
the paper on which the diagram is going to be constructed, leaving sufficient space on all
sides of the diagram for writing scale, heading and key notes. The width of the bar must be
uniform for all the bars with uniform space. Bars may be drawn either horizontally/vertically
but bars must be drawn vertically and be attractive by filling in with special colours.
ii. Multiple Bar Diagrams: A chart depicting two or more characteristics in the form of bars of
length proportional in magnitude of the characteristics. These diagrams are placed side by
side to represent two or more sets of inter-related data in one diagram. This diagram provides
more than single information about the same problem. The bars should be coloured
differently so as to represent good view, e.g. a chart comparing imports and exports of a
country over years.

iii.Composite/sub-divided Diagrams: Bars who’s each bar represents the total of the
components and then it is divided according to the size of each item, e.g. if it is required to
show the total number of teachers in university showing the male and female teachers, then
the bar proportional to the total strength is erected.

2. Areal / Two Dimensional Diagrams: The two dimensional diagrams are constructed when
the items of the data to be compared becomes great. In case of proportional diagrams, we

8
 Role of Statistics in Field Sciences

used to take the area into account, i.e. their length and width both, e.g. rectangles, squares
and circles. Area Diagrams are of two types:
a. Simple Area Diagrams:
i. Rectangles: In drawing these diagrams, we take into account the lengths and a breadth, as
the area of a rectangle is equal to the product of its length and breadth. The lengths of the
rectangles are kept equal and their breadths are proportional to the size of the total figures.

ii. Squares: When it is required to compare two quantities which are large in magnitude and
bear a ratio of 100:10,000. This large proportion in the data size would not favour
rectangles being not a good representative of such type of data. If we construct square
diagrams whereas the sides of a square are determined by taking the square root of the
given figures, e.g. to compare100 to 10,000, the sides of which are in the ratio of √100
√10000 or 1:10. Now with these square roots, squares are constructed with a convenient
scale and with a uniform space in between them.

iii.Circles: When the magnitude of two quantities to be compared is very great, then the construction
of square diagrams becomes not only difficult but laborious too. Then we use to make circular
diagrams. The area of a circle is directly proportional to the square of its radius, e.g. if the radius of
a circle is twice to that of another, then its area will be four times the area of the later. The method
of construction of a circular diagram is the same as that of the square diagram. The square roots of
the various items are calculated, divided by some convenient figures so as to obtain the radii of the
various circles.

Advantages:
i. Circles are easy to draw.
ii. Circles give clear concept than squares.
iii. Circles are attractive to look.
Disadvantages:
i. Difficult to find the area of the various circles from the diagrams at a glance.
ii. Very labourious to find the cubic roots of the data.
b. Sub-divided Area Diagrams:
i. Sub-divided rectangle diagrams: In sub-divided rectangles, first length of the rectangle
is fixed. Then the total units by the units of length to find out the units of the breadth.
Then the lengths of the various components are find out in the same manner and the
sub-divided rectangles are constructed, e.g. when the percentage of the various
components to the total are given, then the sub-divided rectangles can be constructed.
The length is taken as 100 and then adjusting the breadth to make up the given areas.
Advantages: This method is of great use when it is required to compare the expenditure of
the same items in two different families’ budgets.

9
 Role of Statistics in Field Sciences

Details Family A Income (Rs.500$) Family B Income (Rs.400$)

Expenses %age Cumulative Expenses %age Cumulative
Food 140 28 28 120 30 30
Clothing 80 16 44 80 20 50
House rent 100 20 64 60 15 65
Education 30 6 70 40 10 75
Fuel 40 8 78 20 5 80
Misc. 40 8 86 20 5 85
Saving 70 14 100 60 15 100
Total 500 400
ii. Sub-divided square diagrams: when comparison is made, square diagrams do not give
the exact idea about the type of data. In order to grasp the full significance of the data
the total is represented by a square which is sub-divided into rectangles which are
proportional to the size of the various components.

Advantages:
a. It is good for the purpose of comparison
b. Takes lesser space than square diagrams
Disadvantages:
a. It takes a lot of time
b. It is laborious to construct.
iii.Sub-divided circle diagrams: When the circle is divided into various parts it is known
as sub-divided circle. In order to construct a sub-divided circle, first find out the various
angles which represent the various sectors by the formula:
%age Share = Individual x 100 Angle = Component x 360
Total Total

Advantages: This type of diagram presents the data as world population, expenditure
Disadvantages:
i. Construction of this type diagram is difficult.
ii. If the number of items is large, circular diagrams become complicated.
iii. Comparison of the various sectors is difficult.
3. Cubic / Three Dimensional Diagrams: These diagrams are based on three dimensions, i.e.
length, width and thickness. As the volume is taken into account, so also called volume diagrams. The
volumes of these diagrams are in proportion to the given figures, e.g. Cubes, Blocks and Spheres. In
order to construct a three dimensional diagram, we take the cube roots of the data. The cube root of

10
 Role of Statistics in Field Sciences

the figures will give the length of the sides of the various cubes. At first cube roots of the various
figures are taken, and then these cube roots are divided by 50 to obtain the sides of the cubes.

Advantages:
i. Gives a clear view of the data.
ii. Beautiful and attractive.
Disadvantages:
i. To draw these types diagrams, a skill in drawing is necessary.
ii. It is laborious to find the cube-roots of the figures.
4. Pictograms: When the relative values of items are represented by pictures, then termed as
pictograms The size of the different pictures should be in proportion to the values of the
various items, which are represented in the pictograms, e.g. beds in hospitals in the ratio of
Years 1981 1991 2001 2011
No. of Beds 100 140 180 200
Ratio scale (1bed for 20) 05 7 9 10

5. Cartograms: Numerical data are generally shown in a beautiful and attractive manner by
means of maps or cartograms. The distribution of rainfall in various regions of Pakistan and
the production of wheat in the various districts of Pakistan can be shown with the help of
maps, density of population. Seerist made the following suggestions for making diagrams.
1. Choose illustrations which are least liable to be misunderstood, and to faithfully and
correctly interpret the facts.
2. The facts and representation agree that all diagrams are provided with concise, clearly
stated and appropriate titles.
3. Avoid figures which must be read in more than one dimension.
4. Indicate on diagrams the scales of values used to avoid confusion.
5. Include as a component part of the diagrams the data which they illustrate.
6. In expressing the different parts of a total, use lines or bars rather than sectors of circles.
7. In map representation, divide the frequencies and not the number of districts or divisions.
8. In statistical maps representing a series, incorporate as a part of the legend the frequency
with which the units of measurement occur.

11
 Role of Statistics in Field Sciences

Graphic Representation:
Statistical Series:
An arrangement of statistical data in a systematic manner is known as series or when a number of
statistical facts are arranged in accordance with some law, this would form a statistical series or
statistical distribution. According to Connor, “if two variables quantities can be arranged side by
side so that the measurable differences in the one correspond to the measurable differences in the
other, the result is said to form a statistical series”, e.g. production of wheat in the country for the
last 8 years are arrange systematically, this would constitute a statistical series
Types of Statistical Series:
Spatial Series: when the collected data do not change in relation to time but in relation to place or
geographic location, the series is termed as spatial series, e.g.
Places Population (in Million)
Quetta 21
Pishin 15
with respect to time. The variable in this case is years and also termed as historical series because
the data arranged in it relate to either present or past. The figures of population in Balochistan
during the last seven censuses would form a time series. The changes in the level of phenomena
measured are related to the changes in time.
1901 21
1911 30
Condition series: It is an arrangement of statistical data on the basis of changes in some condition.
If the data relating to heights of 30 students, they would form a condition series, as the figures are
neither on the basis of time nor place, but a particular condition namely height, like-wise data
relating to income, expenditure, marks, weight.

B. GRAPHIC REPRESENTATION OF DATA

Generally, diagrams are used for the purpose of propaganda and publicity and give a rough idea
about the data, but from statistical point of view, its value is not more than providing of fairly
good devices to compare two or more quantities except their relationship. To avoid this
complexity or pitfall of the data representation, graph is considered as a better tool for data
representation. Graph is a pictorial presentation of the relationship between the variables. Graphs
have some advantages over diagrams, e.g.
1. Graphs give more accurate and precise results than diagrams.
2. Graphs are of immense use for studying time series and frequency distribution.

12
 Role of Statistics in Field Sciences

Graphics construction:
1. Rectangular Co-ordinates:
a. Draw two straight lines intersecting each other at right angle in a plane of paper.
b. The point of intersection (O) is called origin.
c. If we take any point in the plane, its location can be taken with reference to the point of
intersection point (O) and the lines XOX´ and YOY´ line is termed as coordinates. The line
XOX´ is called X-axis and YOY´ as Y-axis.
d. If any point assumes P is taken in the plane, then we can find its location. Draw PM
perpendicular PM on X-axis and PN on Y-axis, if the distance OM= t and ON = c , then t,c
will be the co-ordinates of the point P. For the sake of simplicity these lines are denoted by
(x, y) where x=OM and y=ON. The distance c along X-axis is called Abscissa of the point
P and distance c along Y–axis is called Ordinate of the point P.
e. The axes divide the plane of paper into four parts each one is termed as Quadrant, e.g.
YOX = First Quadrant (the coordinates of X & Y are Positive, e.g. 2, 3)
YOX´ = Second Quadrant (x co-ordinate is negative and y is positive e.g. -2. 3)
X´OY´ = Third Quadrant (x and y are co-ordinates are negative e.g. -2, -3)
Y´OX = Fourth Quadrant (x coordinate is positive and y coordinate is negative, e.g. 2, -3)

2. Functional Relationship: Whenever a point is located by means of co-ordinates, it means

that two values are involved, and every point ties together and expresses a relation between
two factors termed as variables. These two variables are;
i. Independent Variable: If a variable either increases or decreases by increment
arbitrarily determined is called independent variable and is plotted along X-axis.
ii. Dependent Variable: The value against the time is called dependent variable.
In short, the dependence between the two variables may be real in the sense that the values
of the second variable are definitely determined by the value of the independent variable.
When the two variables x and y are related, then the value of y can be calculated by a given
value of x, then y is said to be function of x expressed as the general function for a
relationship as y=f(x).
If the values of the independent variable are taken along X-axis and the corresponding
values of the dependent variable along Y-axis, a graph representation of the function will
be obtained in the form of a curve /parabola or straight line.
3. Choice and Adjustment of Scales: While drawing a graph, the choice and adjustment of
the scale is an important aspect of graph. The scale should be chosen in the light of the data

13
 Role of Statistics in Field Sciences

size that could be easily accommodated. On graph, independent variable is taken along X-
axis and dependent variable along Y-axis. To avoid distortion and misleading comparison,
no need for horizontal scale to be started with zero but the vertical scale must start from
zero. The X-axis is divided into equal parts, and each part should represent periods of equal
length, while equal space on Y-axis represents equal amounts if ratio scale is used. There
are no hard and fast rules regarding the ratio between the abscissa and the ordinate scale.
Bowley states, “It is difficult to lay down rules for the proper choice of scales by which the
figures should be plotted out. It is only the ratio between the horizontal and vertical scales
that need to be considered. The figures must be sufficiently small for the whole of it to be
visible at once, if the figure is complicated, related to long series of years and varying
numbers, minute accuracy must be sacrificed to this consideration, e.g. horizontal scale is
decided, the vertical must be chosen so that the part of the line which show the greatest rate
of increase is well inclined to the vertical which can be managed by making scale
sufficiently small, however, the fluctuation must be clearly visible for which the scale may
need to be increased.”
4. Plotting of Data: At this stage the values of x and y co-ordinates of the various points are
plotted on the graph paper for the whole period. Obviously there will be as many numbers
of the points as the number of years. After plotting, all the points are joined. If the figures
relate to continuous variable then the plotted points should be connected by smooth curve
but if the variables are discrete, then the plotted points should be joined by straight lines.
Types of Time Series Graphs:
In time series graphs, values of a variable are given at a different periods of time. When a graph
of such series is drawn, it would give changes in the value of a variable with the passing of time
and the graph is termed as Historigram.

Types of Historigram: There are two types of Historigram.

1. Absolute Historigram: If the absolute values of the time series are plotted, then the graphic
representation obtained is called absolute Historigram. Absolute Historigram are of one or
two or more variables.
i. One Variable Absolute Historigram: Draw X and Y axis on plain paper. Take time along X-
axis and figures along Y-axis. Plot the various points and join it either by straight or curve lines.
ii. Two Variables Absolute Homogenous Historigram: If there are two or more variables of
homogenous units (same units), all of them can be shown on the graph paper by taking
horizontal and vertical scale as common. The method of plotting the curve is exactly the
same as that of Historigram of one variable.
iii. Two Variables Absolute Homogenous Historigram: If there are two or more variables in
different units, even then can be shown on the graph paper. In this case only horizontal scale

14
 Role of Statistics in Field Sciences

is taken as common but as the values of the dependent variables are measured in different
units, so vertical scale may be in different units, separately on both sides of the graph.
a. The scale should be so adjusted to bring the two variables nearer to each other for the
sake of comparison.
b. The scale should be so adjusted that the average values of both the scales should come in
the centre of the scale.
2. Index Historigram: When instead of plotting the absolute values of the time series, their
indices are plotted on the graph paper termed as Index Historigram. This type of graph is
drawn for the purpose of comparison in order to study;
i. Changes in one variable over a period of time
ii. Changes of two or more variables over a period of time
3. Mixed Graphs: When two or more variables are expressed in different units on the same
graph paper by two different graphs, then called as mixed graph. According to this method,
one variable is presented in the form of bar diagram and the second variable is plotted in the
form of a curve making the graph very attractive.
4. Graphs on Ratio Scale: Graphs drawn on a natural scale along with values of y are scaled
in proportion to their actual values which show absolute movements in the values of a
variables and failed to study the relative movement in values. However, in economic
investigations, the study of the relative movements is becoming more important which gives
birth to ratio scale or logarithmic scale instead of natural scale. “Ratio scale is that scale
which studies the relative movements in the values of a variable. The difference between the
ratio scale and the natural scale can be noticed from the opposite figure. In the natural scale
equal differences are measured by equal distances on this scale, whereas in ratio scale
proportionate movements are measured. Natural scale is based on arithmetic progression,
whereas the ratio scale is based on geometric progression.
5. Logarithmic Curves: The relative movements may be presented graphically by the
following two ways:
a. By plotting the logarithm of the given values on a natural scale. In this method, the logarithm
of the variable y is taken and is plotted along y-axis on the natural scale. The plotted points
are joined by straight lines and termed as logarithmic curve.
b. By plotting the given values on logarithmic scale. In this method, instead of taking the
logarithms of the variable y, actual values of the variable are plotted on logarithmic paper
along y-axis which is designed on the ratio scale principle while the horizontal scale of this
logarithmic paper is in the arithmetic ratio, and the graph is called semi-logarithmic graph.
Instruction for Reading Graphs on Ratio Scale:
a. If a logarithmic curve is increasing, it indicates that the rate of growth is increasing.
b. If a logarithmic curve is falling downward, it means that the rate of growth is decreasing.
c. If the logarithmic curve is a straight line, reflects that the rate of growth is uniform.
d. If the curves rise more steeply in one portion than in another portion, then it means that the
rate of increase in the former case is more than in the later.
e. If the curve falls more steeply in one portion than in another portion, means that the rate of
decrease in the former is more than that in the later.
f. If two curves are parallel to each other, then they represent equal percentage rates of change.
g. If one curve is steeper than another, its rate of change is higher than that of the other curve.
Advantages & Disadvantages of Ratio Scale:

15
 Role of Statistics in Field Sciences

a. Ratio scale does not begin with zero line as relative rates of change are compared on it,
while natural scale begins with zero as absolute values are compared on it.
b. In ratio scale it is not necessary to have a base line, so any curve can be shifted upward or
downward without affecting the proportionate changes in its values. To compare two or
more curves on the ratio scale, they can be brought close to each other to facilitate comparison.
c. When the range of the data is very high, it can only be represented on ratio scale. No doubt
natural scale can be used even if false baseline is used but this would give fallacious conclusions.
d. Two or more series measured in different units can be represented on ratio scale.
e. On ratio scale only aggregate and its component parts can be studied.
f. Ratio scale cannot study absolute changes.
g. Natural scale cannot study relative changes.
h. Logarithmic scale is of great importance in the case of index Historigram as index
Historigram studies relative changes and can be drawn better on ratio scale. On the other
hand, if drawn on natural scale, they would give fallacious impression.

2. FREQUENCY DISTRIBUTION: Frequency distribution can be defined as “the

organization of a set of data in a table showing the distribution of the data into classes together
with the number of observations in each class or group is called a frequency distribution. When
the statistical data are grouped according to size or magnitude, the series so formed is known as
frequency distribution which generally consist of class intervals and their corresponding
frequencies falling in each class known as frequency denoted by f. Data presented in the form of a
frequency distribution are called grouped data while the data in original form is called ungrouped
data. When the data is arranged in ascending or descending order of magnitude is said to be an
array. The purpose of a frequency distribution is to produce a meaningful pattern for the overall
distribution of the data from which conclusion can be drawn. Following requisites are needed for
the construction of a frequency distribution table.
Types of Frequency Distribution: Frequency distribution is either symmetrical or asymmetrical.
Forms of Frequency Distribution:
Series of Individual Observation: If the items are listed singly after observations, then it will
constitute a series of individual observations, e.g. Marks are obtained by 8 students are given
individually, then the frequency table would be:
Roll Nos. 1 2 3 4 5 6 7 8
Marks: 10 15 25 30 20 40 50 60
Discrete Series: In this series, the various units are capable of exact measurement and each unit of
the data is separate and complete, and definite breaks are visible between different units, e.g.
counting of persons of salary of 500. The data of this type would give rise to discrete series, e.g.
Marks: 11 12 13 14 15 16 17 18
Students: 5 10 8 6 11 15 12 4
Continuous Series: In this series the statistical units are arranged in groups because they are not
exactly measurable and are only shown by approximations, e.g.
Marks obtained No of students
0-10 14
11-20 10
Classification According to Class-intervals: This type classification is based on the data where
direct quantitative measurement of the data is possible, e.g. income, height, weight height of a

16
 Role of Statistics in Field Sciences

persons in categories tall and short are placed in a range of heights called class interval
5.5 ___7.5 feet so proceeding in this way the data are divided into a number of classes each of
which is called class interval. The starting side is called lower limit of a class and the ending side
is called the upper limit of a class. The difference between the two class limits is called the
magnitude and the number of items which comes under any class interval are called Class
frequency. Suppose if the heights of 80 students lie in the range of 5.5-7.5 feet, then 80 is the
frequency of the class 5.5 – 7.5 feet.
Magnitude of the Interval: For finding magnitude of the class, the best way is to find the range of
the data (difference between maximum and minimum values) and divide this range by the number
of classes desired. If we have a range of the data of 50 and we desire to have 10 groups, then the
magnitude of each class interval would be 5. So, magnitude should be such selected not to distort
the important characteristics of the data. The human minds always prefer easy figure like
2,5,10,15,20 as magnitude, so we must take equal interval of the magnitude because class interval
with unequal magnitude are unfit for the purpose of comparison and always give misleading results.
Class Limits: The class limits should be so chosen that the mid-point of a class interval and the
actual average of items in that interval should be as close to each other as possible, other-wise the
class limits would distort the main characteristics of the data.
Framing of Class Intervals: class intervals can be framed by the following methods:
I II III IV
0-10 1-10 0-9.9 under 10
10-20 11-20 10-19.9 under 20
If the values are identical with the upper limit of a class interval are excluded and included in the
next class and is considered as not a good method because there is a danger of overlapping.
Counting the Number of Items: When classes, magnitudes and classes limits are decided, then
the next thing is to count the number of items in each class.
By Tally Sheets: According to this method the class intervals are written on a sheet of paper
called Tally Sheet. Then a stroke is marked for each time against the class interval in which it lies,
on doing like this but mark a horizontal line after every 04 strokes in a class which indicates the
fifth items. This would form a group of five which are easy to count.
By Mechanical Tabulation: many mechanics are available in hands both hand and electrically
operated. By hand operated one is called needle sorting by which a large number of items can be
sorted under a number of headings by means of cards of convenient size or shape with a number
of holes in it. Each hole in the card represents particular values. The cards are stacked and the
needle passes through some hole giving some value which represents the value of that hole. These
cards are later on taken away and counted. In this way the frequencies of the various classes are
found out by the repetition of this method.
Construction of a Frequency Distribution Table: There are no hard and soft rules for deciding
on the number of classes which actually depends on the size of data. Statistical experience reflects
that there is no less than five and not more than 20 classes are generally used. H.A. Sturges has
proposed an empirical rule for determining the number of classes into which a set of observations
should be grouped as K= 1 + 3.3 log N
Types of Frequency Distribution: These types of graphs are drawn to represent the data in a
compressed and tabular form. Readers, stakeholders and decision makers appreciate the relative
sizes of a number of items more easily when the data are represented graphically rather than

17
 Role of Statistics in Field Sciences

looking at a table. The method of drawing the graph of frequency distributions is exactly the same
as is used in the time series. Frequency distribution is represented in two forms;
1. Frequency Graph for Discrete Series: The graph drawn for showing discrete series carries
the size of items is measured along X-axis and the corresponding frequencies along Y-axis.
The whole data are plotted and thin lines /bars of uniform thickness are erected above each
value on the x-axis. Some blank space is left in between the thin lines/bars for the sake of
distinguishing from each other.
2. Frequency Graph for Continuous Series: This type of data is mostly shown by histogram.
Histogram is a bar graph with no space in between the bars. This is drawn on a pair of co-
ordinate axes with the unit of measurement for the observations being measured along the
horizontal axis and the number of observations being measured along the vertical axis. The
vertical scale or y-axis is usually taken to begin at zero. The horizontal scale –x-axis can
begin at any convenient number at which to begin the classes. For continuous series, the
class intervals are plotted along the horizontal scale and their corresponding frequencies
along the vertical scale. Various bars are erected above the lower and upper limits of the
various class intervals measured on the horizontal scale. The various bars are joined together
and no blank space is left in between them giving the impression of continuity of the
distribution. The class intervals should have uniform size otherwise it would give
misleading result. No class interval should be left without plotting even if it has no
frequency against it.
3. Histogram of Unequal Class Intervals: If the frequency distribution consists of unequal
class intervals, the areas of the rectangles in the histogram are taken directly proportional to
their frequencies. It means that there will be rectangles with unequal width. If the magnitude
of an unequal interval is m times the magnitude of one of the equal intervals, then the
ordinates representing its frequency will be proportional to 1/m times its frequencies.
4. Frequency Polygon: This graph is formed by placing a dot at the mid-point i.e., the class
mark to of each rectangle of the histogram and connecting the dots. There are two methods
of drawing frequency polygon. The joining of the mid-points by straight lines gives a
frequency polygon which does not give a continuous curve, with the result that it cannot
show the regularity and uniformity of variation in the magnitude
i. In the first method, bars are erected for the given data in the same way as in the case of
frequency graphs. The mid-points at the top of each bar are joined together by straight lines
and the resulting figure is the required frequency polygon.
ii. In the second method, instead of erecting bars, only points are marked against mid-values of
the various class intervals. The points obtained are joined together by straight lines which
give the required frequency polygon.

18
 Role of Statistics in Field Sciences

5. Frequency Curve: Frequency curve is drawn by two methods.

i. First the polygon is drawn and later it is smoothened out. While smoothing the curve, the top of the
smoothened curve should pass above the highest point of the frequency polygon without distorting
the original shape of the curve.
ii. In second method, the frequency of each class interval is plotted against the mid-point of that class
interval. The plotted points are then joined by a smooth curve which gives the required frequency
curve.

6. Cumulative Frequency Curve/Ogive: As this curve is based on cumulative frequencies that is why
it is known as cumulative frequency curve. The graph of cumulative frequency distribution is called an
Ogive. For drawing this graph, frequencies are calculated by adding the frequencies of all the succeeding
classes to the frequency of a class which results into cumulative frequency series. This graph is same in
drawing to frequency curve with a difference that cumulative frequency of each class interval is plotted
against the upper limit of the class interval. The plotted points are then joined by straight lines and
cumulative frequency curve is obtained.

19
 Role of Statistics in Field Sciences

Graphs
Graphs are charts consisting of points, lines and curves. Charts are drawn on
graph sheets. Suitable scales are to be chosen for both x and y axes, so that the entire data
can be presented in the graph sheet. Graphical representations are used for grouped
quantitative data.
Histogram
When the data are classified based on the class intervals it can be represented by a
histogram. Histogram is just like a simple bar diagram with minor differences. There is
no gap between the bars, since the classes are continuous. The bars are drawn only in
outline without colouring or marking as in the case of simple bar diagrams. It is the
suitable form to represent a frequency distribution.
Class intervals are to be presented in x axis and the bases of the bars are the
respective class intervals. Frequencies are to be represented in y axis. The heights of the
bars are equal to the corresponding frequencies.
Example
Draw a histogram for the following data

20
 Role of Statistics in Field Sciences

Seed Yield (gms) No. of Plants

21