The values of accuracy of data used and coverage of full data is very important for graphical presentation of the data. The graph is presented to the top management for decision making. If it is prepared with wrong data or incomplete data then it will lead to wrong decision making which will be detrimental to the interest of the organization.

Following steps are taken.
(a) Add mid values of two successive classes.
(b) Divide it by 2.
(c) It is upper limit of preceding class and lower limit of successive class.
(d) Find the difference between mid values of two successive classes.
(e) Deduct it from the upper limit obtained to get lower limits of the preceding class.
(f) Repeat it for all classes if classes are unequal (if difference between different mid values is different) and if classes are equal (if difference between different mid values is equal) simply keep on adding the difference between two successive mid values to find the upper limits of different classes.

Usually, when we draw any graph, the scale on which the graph is measured starts from zero on the y-axis. However, under the situations when the data to be plotted on graph starts from a value which is far above zero, results in the problem of shortage of space on graph. To overcome this problem of shortage of space, a false base line is plotted. False base line is a line which is drawn to grasp the attention of the reader on the fluctuations which usually remains unnoticed.

General rules to construct a Time Series graph
(a) Since time can never be in negative values, II and III quadrant are not used in making of time series graphs.
(b) Time period (week, Month, Year) is taken on X- Axis. And Variable under study is taken on Y-Axis.
(c) We start Yaxis with Zero and decide the scales for both the axis.
(d) Different values are plotted on the graph.
(e) By joining these points, we get a time series graph.
Time series graph can be of three types:
(a) One Variable Graph
(b) Two or more than two variable Graph
(c) When False Base Line is used
(d) Graphs of Different Units

When class intervals are unequal, we follow the steps given below:
(a) Take the class which has the lowest class interval.
(b) Do not adjust the frequencies of the lowest class.
(c) Frequencies of other classes are adjusted according to the lowest class interval.
(d) Formula for adjusting frequency is:
\(\frac { Frequencyofthegivenclass\times Lowestclassinterval }{ ClassIntervalofgivenclass } \)
(e) Keep the width of the rectangle according to class interval of the concerned class and length of the rectangle according to the adjusted frequency.

The following steps are to be followed to construct a histogram.
Step 1: Mark class intervals on X-axis and frequencies on Y-axis.
Step 2: The scales for both the axes need not be the same.
Step 3: Class intervals must be exclusive. If the intervals are in inclusive form, convert them to the exclusive form.
Step 4: Draw rectangles with class intervals as bases and the corresponding frequencies as heights.
Step 5: The class limits are marked on the horizontal axis and the frequency is marked on the vertical axis. Thus, a rectangle is constructed on each class interval.
Step 6: If the intervals are equal, then the height of each rectangle is proportional to the corresponding class frequency.
Step 7: If the intervals are unequal, then the area of each rectangle is proportional to the corresponding class frequency.

A frequency polygon can also be converted back into a histogram by drawing vertical lines from the bounds of the classes shown on the horizontal axis, and then connecting them with horizontal lines at the hieghts of the polygon at each mid-point. Frequency Polygon can be drawn in two ways:
Method:
(i) Draw a suitable Histogram as explain in the previous section following all the basic principles.
(ii) Get the mid points of the upper horizontal side of each rectangle.
(iii) Join these mid points of the adjacent rectangles of the histogram by straight lines.
(iv) Ends of the frequency polygon are to be extended to mid point of classes at both sides.
(v) Labelling should be properly done and scale of measurement should be clearly shown.
An example is shown below

This is a modification of the Ogive when the variables and the cumulative frequencies are expressed as percentages. It serves to measure the evenness of the distribution and is useful in picturing the distribution and dispersion of wealth, sales and profits etc. If we get a line which is a diagonal of the square, it implies that variable is evenly distributed amongst population. Farther is the line from this diagonal, higher is the inequality and vice versa.

The difference between Histogram and Bar Diagram is given below:

Usually, when we draw any graph, the scale on which the graph is measured starts from zero on the y-axis. However, under the situations when the data to be plotted on graph starts from a value which is far above zero, results in the problem of shortage of space on graph. To overcome this problem of shortage of space, a false base line is plotted. False base line is a line which is drawn to grasp the attention of the reader on the fluctuations which usually remains unnoticed.
A kinked line is used on x axis for the same purpose for which false base line is used for y axis. It means when variable starts from a higher value, we use kinked line and when frequency starts with a first higher number followed by smaller gaps, we use false base line.

Table showing distribution of production in two fortnights amongst internal consumption, exports and stock. (in tones)

In case of equal class interval simply we draw the frequencies of equal width but when class intervals are unequal these frequencies need to be adjusted. It will be clear from the example given below:

Depending on the number of variables involved, we determine whether we should use one way, two way, three way or multifold table. To present non workers in urban areas, we will use two way table.

(a) Simple bar Diagram
(b) Pie Chart
(c) Pie Chart

The x co-ordinate of the point of intersection of less than and greater than cumulative frequency curve is the median.
Therefore, first we need to draw ogives.

(a) Frequency Distribution

Marks	Tally Bars	Frequency (f)
0-10	I I I	3
10-20	1	6
20-30	I I I	8
30-40	I	6
40-50	I	6
50-60	I I I	3
60-70	I	6
70-80	I I	2

(b)
 
(c) Out of 40 students 23.5% (90ut of 40) get pocket money less than 20 and 20% (8 out of 40) get pocket money more than 60.
Frequency Curve: It is described as a smooth frequency polygon. A frequency curve is described in terms of its (i) symmetry (skewness) and (ii) degree of peakedness (kurtosis). Two frequency distributions can also be compared by superimposing two or more frequency curves provided the width of their class intervals and the total number of frequencies are equal for the given distributions. Even if the distributions to be compared differ in terms of total frequencies, they still can be compared by drawing per cent frequency curves where the vertical axis measures the per cent class frequencies and not the absolute frequencies.

It is shown below:



Histograms: These diagrams are used to graph grouped data. It is one of the most popular and commonly used devices for charting continuous frequency distribution. It consists of erecting a series of adjacent vertical rectangles on the section of X-axis with bases of equal width of the corresponding class intervals and the heights are so taken that the area of the rectangle are equal to the frequency of the corresponding classes.