Diagrammatic and 6 Graphic Presentation INTRODUCTION SIGNIFICANCE OF DIAGRAMS AND GRAPHS GENERAL RULES FOR CONSTRUCTING DIAGRAMS TYPES OF DIAGRAMS GRAPHS. GRAPHS OF FREQUENCY DISTRIBUTION MISCELLANEOUS ILLUSTRATIONS LIMITATIONS OF DIAGRAMS AND GRAPHS eee eee ee py eeeSTATISNCA, MENON, INTRODUCTION In the previous chapter we have discussed the techniques of Clases and tabulation that help in summarising the collected data and all&tion them in a systematic manner. However, these forms of Presentation ing always prove to be interesting to the common man. Too many tone hot often confusing and may fail to convey the message effectively to tha whom they are meant. » One of the most convincing and appealing ways in which statist, results may be presented is through diagrams and graphs. Evidence gf 41 can be found in newspapers, magazines. journals. advertisements, 41.) There are numerous ways in which statistical data may be displaye, pictorially such as different types of diagrams, graphs and maps. Very Pvc] the problem is that of selecting the best out of several methods that may 2 available. This is a difficult task and requires a great deal of artistic talent and imagination on the part of the individual or ageney engaged in the preparation of diagrams and graphs. It is not practicable to discuss al] the possible forms of charts here. An attempt is made in this chapter to illustrate some of the major types of diagrams, graphs and maps frequently used in presenting statistical data. SIGNIFICANCE OF DIAGRAMS AND GRAPHS Diagrams and graphs are extremely useful because of the following reasons: They give a bird's-eye view of the entire data and, therefore, the information presented is easily understood. It is a fact that as the number and magnitude of figures increases they become more confusing and their analysis tends to be more strenuous. Pictorial presentation helps in proper understanding of the data as it gives an interesting form to it. The old saying ‘A picture is worth 10,000 words’ is very true. The mind through the eye can more readily appreciate the significance of figures in the form of pictures than it can follow the figures themselves. They are attractive to the eye. Figures are dry but diagrams delight the eye. For this reason diagrams create greater interest than cold figures. Thus, while going though journals and newspapers the readers generally skip over the figures but most of them do look at the diagrams and graphs. | Since di is have attraction value, they are very popular in exhibitions, fairs, ces, board meetings and public functions. They have a great memorising effect. The impressions created by diagrams last much longer than those created by the figures presented in a They facilitate comparison of data relating to different periods of time of regions. Diagrams help one in making quick and accurate different een of data. They bring out hidden facts and relationship and can ulate as well as aid analytical thinking and investigation. ‘Comparison of Tabular and Diagrammatic Presentation may be— in the form of tables as well as diagrams and ad nf presenta have their own usefulness for particular the choice of the form of presentation must be madeMATIC AND GRAPHIC PRESENTATION 1 thought and care. The following points may be kept in view in contain precise figures whereas diagrams giv La 2 , ite idea. Exact values can be read from a table. a information can be presented in ph or diagram. one table than cither in one usually require much close readin; — it than diagrams. eee and diagrams have a visual appeal wove (0 more impressive to laymen. ia epee between Diagrams and Graphs is how to distinguish a diagram from a graph. Though there line of demarcation between the two, yet following points of e may be noted: constructing a graph we generally make use of graph paper a diagram is generally constructed on plain paper. In other a graph represents mathematical relationship (though not ly functional) between two variables whereas a diagram S are more attractive to the eye and as such are better " for publicity and propaganda. They do not add anything to ming of the data and, therefore, from the point of view of a or research worker they are not helpful in analysis. ‘on the other hand, are very much used by the statistician he research worker in analysis. In fact, these days it ts to find any research work without the graphic support. presenting frequency distributions and time series, graphs are propriate than diagrams. In fact. for presenting frequency ons diagrams are rarely used. FOR CONSTRUCTING DIAGRAMS general rules should be observed while constructing diagram must be given a suitable title. The title should few words as possible the main idea that the diagrams intend owever, the brevity should not be secured at the cost of ion of essential details. The title may be given either at the or below it. between Width and Height A proper proportion between d width of the diagram should be maintained. If either the is too short or too long in proportion, the diagram would ok. While there are no fixed rules about the dimensions, a u as suggested by Lutz in the book entitled “Graphic be adopted for general use. It is known as “Root-two”re 130 STATISTICAL METHHOn, that is, a ratio of 1 (short side) to 1.414 (long side). Modifications no doubt, be made to accommodate a diagram in the space available. ""™“%: | | | ; ; | 3. Selection of Scale The scale showing the values should be in ever, numbers or in multiples of five or ten, e.g., 25. 50. 75, or 20, 40, 60. values like 1, 3, 5, 7 should be avoided. Again no rigid rules can be taiq down about the number of rulings on the amount scale, but ordinarily should not exceed five. The scale should also specify the size of the unit and what it represents; for example, “million tonnes”. “number of persons in thousands”, “units produced in lakhs”, etc. All lettering should be easily readable without turning the chart up and down. 4. Footnotes in order to clarify certain points about the diagram, footnote may be given at the bottom of the diagram. 5. Index An index illustrating different types of lines or different shades, colours should be given so that the reader can easily make out the meaning of the diagram. 6. Neatness and Cleanliness Diagrams should be absolutely neat and clean. z 7. Simplicity Diagrams should be as simple as possible so that reader can understand their rheaning clearly and easily. For the sake simplicity, it is important that too much material should not be loaded a. e diagram otherwise it may become too confusing and prove ; worthless. Several simple charts are often better and more effective than one or two complex ones which may present the same material in a confusing way. TYPES OF DIAGRAMS In practice, a very large variety of diagrams are in use and new ones are constantly being added. It would be outside the scope of this book to deal exhaustively with the subject and as such only the more frequently used diagrams are disussed. For the sake of convenience and simplicity they may be divided under the following heads: One-dimensional diagrams, e.g., bar diagrams. _ Two-dimensional diagrams, e.g., rectangles. squares and circles. _ Three-dimensional diagrams, e.g., cubes, cylinders and spheres. One-dimensional or Bar Diagrams E most common type of ms used i tice. A width is shown merel ea tienuicnuithertear e it is only the length of the bar that matters number of items is large, lines may be~ AND GRAPHIC PRESENTATION m of bars to economise space. The special merits of bar 95M ne following: are readily understood even by those unaccustomed to a charts or those who are not chart-minded. - the outstanding advant that th th ee e : tage ey are the simplest large number of items are to be compared thi are the wher fm that can be used : . bar diagrams. the following points should be kept in mind: width of the bars should be uniform throughout the diagram. between one bar and another should be uniform Year : 2004-05 2005-06 2006-07 Fixed Assets : 415 481 607 bar (b) Sub-divided bar diagrams vp ed iltiple Sa coors (a) Percentage bar diagrams (6) Deviation bars. age ee wr. ————STATISTICAL METHODS 12 (a) Simple Bar Diagrams A simple bar diagram {s used to represent only one variable. For example, the figures of sales. production. population, etc., for various years may be shown by means of a simple bar diagram, Since these are of the same width and only the length varies. it becomes very easy for the reader to study the relationship. Simple bar diagrams are very popular in practice. This can be either vertical or horizontal. In practice, vertical bars are more popular. However, an important limitation of such diagrams is that they can present only one classification or one category of data. For example, while presenting the population for the last five decades, one can only depict the total population in the simple bar diagrams, and not its sex-wise distribution. Illustration 1. Following table gives the birth rate per thousand of different countries over a certain period : GCouney Both Rate Country Birth Rate ee 33 China 40 Germany 16 New Zealand 30 UK. 20 ‘Sweden 15 Represent the above data by a suitable diagram. Solution. The above data can suitably is represented by simple bar diagram. Birth rate per thousand of Different CountriesATIC AND GRAPHIC PRESENTATION is The above dala can be represented by a simple bar diagrarn DIVIDEND (%) (Rs. Cr.) OF OSWAL MILLS LTD. | | In a sub-divided bar diagram each bar is further sub-divided in134 vertical base. CO iniana Marine PRODUCTION (Lakh tonnes) 8 target for 2006-07: Textiles Gems and jewellery Engineering goods Agricultural products Chemical products Leather products Ores and minerals Marine products Plantations Electronics Petroleum products Raw cotton GROWTH OF PRODUCTION OF FISH 1998-99 to 2004-05 (c) Multiple Bars In a mult 2005-06 15185 8839 6505 5514 5419 3692 2146 1743 1339 611 1379 196 STATISTICAL METHOD, Solution. The above data can be represented by a sub-divided bar diagram (drawn on ‘ Mlustration 4. The following data relate to exports in 2005-06 (provisional) and exports Figure (Rs. crore) 2781 2080 1600 1200 861 544 Represent the data by- horizontal bars. (Please see page 135 for diagram) tiple bar diagram two or more sets of—— |AGRAMMATIC AND GRAPHIC PRESENTATION Exports Target for 2006-07 Figures in Rs. crore Teste oe ee gS SSL GL IE 2989] ST Jewnllery KT 17056 — Goods QT 8000 “pris LSTA Products KS Qa 7216 Prac Ses Leather Products Sire0 Ores & 2146 Minerals 2781 Marine_[—]1743 Products Kw 2080 a 1339 1600 611 [] Exports in 2005-06 (Provisional) eiectronies 18 500 Exports Target for 2006-07 Petroleum. 1379 Products” [Ja61 Raw _[]196 Cotton” F544 Sree on ee bem Zone Year West North East South 2005 78.4 88.9 83.7 89.9 ie 75. 62.5 103.6 75.5 6 121.2 116.5 107.6 123.9 135 Milustration 5. The regional rainfall indices during the year 2005 to 2007 are given below: aus? “Represent the data by a multiple bar diagram. (Please see page 136 for diagram) Wustration 6. Draw a multiple bar diagram from the following data: Sales Gross Profit Net Profit a (000 Fs.) (000 Fs.) (000 Fs.) 2004 120 40 20 2005 135, 45 30 2006 140 55 35 2007 150 60 40 (B.Com, S.V. Univ., B.Com., Gulbarga Univ.) (Please see page 136 for diagram)eea ==F43 5 : Ff ge Go | £8 POS Ee | us SSS) gs / =v — > ae a INS 5 gy g ——, gfATIC AND GRAPHIC PRESENTATION Particulars 2005 2006 wages 45.0 50.0 Other costs 30.0 333 Polishing 15.0 16.7 Total costs 90.0 100.0 Sale price 100.0 100.0 Profit or loss +10.0 SALES (In Ris.) . Take the sale price per chair as As. 100 and ges. The percentages $0 obtained are given below: express the other figures in 2007 525 35.0 175 105.0 100.0 -5.0138 STATISTICAL METHODS SALES & NET PROFITS OF PVT. INDUSTRIAL COMPANIES (% CHANGE OVER PREVIOUS YEAR) (0) sates NET PROFITS Broken Bars In certain series there may be wide variations in values—some values may be very small and others very large. In order to gain space for the smaller bars of the series, the largest bar may be broken. Mustration 9. The following data relate to the Additional Taxation during the year 1997-98 to 2006-07: Additional Taxation (Rs. in Crores) ADDITIONAL TAXATION (Rs. in crores) UMEacramMartc AND GRAPHIC PRESENTATION i jonal Diagrams shed from one-dimensional diagrams in which only the Jength ts taken into account, In two-dimensional diagrams the length ‘as the width of the bars is considered, Thus the area of the bars the given data. Two-dimensional diagrams are also known as diagrams or area diagrams. The important types of such diagrams ingles, (b) Squares, and (c) Circles. This form is quite popular. Since the area of a rectangle Is to the product of its length and width, while constructing such a oth length and width are considered. When two sets of figures be represented by rectangles, either of the two methods may be We may represent the figures as they are given or may convert to percentage and then sub-divide the length into various onents. The latter method is more popular than the former as tt comparison to be made on a percentage basis. The following would illustrate both these methods of constructing rectangular 40. Represent the following daia on details of cost of two commodities by the’ — 7 Commodity A Commodity 8 per unit a ee sold 40 units 30 units of raw materials Rs. 52 Rs, 50 expenses of production Rs. 64 Rs, 60 Rs. 44 Rs. 40 : (B.Com., Karnataka Univ.) Let us calculate the cost of raw materials, other expenses ana protit per unit. Y Costs and Profits per unit ‘of Commodities A & B | RAW MATERIALS | OTHER EXPENSES le ¢ | PROFITS. noe ewConmrantey A Total (Ra) Par Una a) esDIAGRAMMATIC AND GRAPHIC PRESENTATION iI area diagram is more difficult to read than to construct because of qe problem of judging areas. The rectangular method of diagrammatic presentation is , to use where the values of items vary widely. For example, if in ‘{ilustration given above the number of units sold of commodity A and 20 and 240 respectively, the width of the rectangles would be in the £5; 60 or 1: 12. If this ratio is taken, the diagram would look very . It is in order to overcome this difficulty that squares are used. - method of drawing a square diagram is very simple. One has to take ., “square-root of the values of various {tems that are to be shown in the eae and then select a suitable scale to draw the squares. 42. Represent the following data by @ suitable diagram : Total Public Sector Outlays (Rs. ‘000 crores) 1 " um Vv v vl vil vii 4.672 8.577 16,566 35,595 69,380 1.80,000 3.25,000 ce there is a very big gap between the First Plan and Eighth Plan oullay, a may be quite suitable here. The size of one side of a square will be the square root of the value to be represented and the size of the sides of shall be proportional to the square roots of the various quantities to be CALCULATIONS FOR DRAWING SQUARE DIAGRAM ‘Outlay ‘Square-root ‘Side of the square in cms 1,960 44.27 0.22 4,672 68.35 0.34 8,577 92.61 0.64 16,566 128.71 0.64 35,595 188.67 0.94 69,380 263.40 1.31 1,80,000 424.26 212 3,25,000 570.08 2.85 Tigure of the square-root has been divided by 200 and the side of the square < PLANS >ur STATISTICAL, METHOD, (c) Circles Another way of preparing a two-dimensional diagram ts in an of circles. In auch depen both the total and the component man or sector can be shown, The area of a circle is proportional to the square of its radius, As in the construction of squares, the square-roots of various figures are worked out while constructing the circle. However, in the latter case the radius of the circles can be obtained by dividing the value of pic and taking square-root. Circles can be used in all those cases in which squares are Used, However, in both these types of diagram it is difficult to Judge the relative magnitudes with precision. Mlustration 13, Represent the data of Illustration 12 with the help of circles, lutlay i ve" 22 bynes (7) ' 1960.00 624.20 24.98 " 4672.00 1487.90 38.57 i 8577.00 2731.53 52.26 Vv 16566.00 5275.80 72.63 v 35595.00 11335.99 106.47 vi 69380.00 22095.54 148.65 vil 1,80,000.00 57324.84 239.43 vill 3,25,000.00 103503.18 321.72 Note: Area of a circle = x 7? = AfeaATIC AND GRAPHIC PRESENTATION Pie diagrams are very 7 ‘ populat used in ractice to percentage breakdowns. For seca teams the help of a ple ones Uitives : a the expenditure of the Government is oot ren eads like Agriculture. Irrigation. Industry : Aan ete. Similarly through a pie diagram we can show ae Itures incurred by an industry are divided under eads Ike raw materials, wages and salaries. selling and expenses, etc. The pie chart is so called because the looks like a pie, and the components resemble slices cut making comparisons, pie diagrams should be used on @ fe basis and not on an absolute basis. since a series of pic showing absolute figures would require that larger totals be ed by larger circles. Such presentation involves difficulties of sional comparisons. However, when pie diagrams are on a percentage basis, percentages can be presented by al In size. It may be noted that this problem does not arise in a single pie diagram. out the sectors for pie chart, it is desirable to follow some J arrangement or sequence. It is common practice to begin the { Component sector of pie diagram at 12 O'clock position on the ually the other component sectors are placed in clockwise in descending order of magnitude. except for catch-all like “miscellaneous” and “all others” which are shown last. ralues : () 6O per cent, (i) 25 per cent, (ii (360/100 = 3.6). the ig values of the four components in the illustration are (60) < (25) x (3.6) = 90°; (10) x (3.6) = 36%; (5) x (3.6) = 18. ond step is to draw a circle of appropriate size with a compass. of the radius depends upon the available space and other factors ation. d step is to measure points on the circle representing the size of or with the help of protractor. The ordinary protractor is based ‘ale in which the total circle is 360 degrees, but it is possible to a protractor in which the entire circle is divided not into 360 but parts so that the angle representing any desired percentage can ectly. g out the sectors for a pie chart it is desirable to follow some gement, pattern or sequence, For example, it is a common to arrange the sectors according to size, with the largest at the others in sequence running clockwise. An essential feature of pie the careful identification of each sector with some kind of ory or descriptive label. If there is sufficient room, the labels can bd inside the sectors; otherwise the labels should be placed in us positions outside the circle, usually with an arrow pointing the appropriate sector. We have divided the outlays by h to get the square of radius.\ ea “ ont, ny, Muereien 14. Kran 4 ye Gagan bo 6 hong Gate A orth Pred atid Boag Ot Peat Han, PPO Mast at Ranh (nevengiant 12% Wenagetein we, 2% Crergy AP Wem y 008 Mona is 164% Trenegon, Commanncaon, 169% ood Gerices and Others 161% 0 Com, Carigilety imp, Sotution. Tha arpa mt corte © Gren by q Peroage OM , 146) « Porcortage otiay «36 COAAPUT ATION FOR Pie inGtihen a ————_—— ——$—_$—___. Seriar Per eniaye ‘Angle cxalary Agpamsne 9d Furn evelopment 129 129446 =sc AND GRAPHIC PRESENTATION Me 15. The following data relate to how the Rupee comes and goes: ee Rupee goes ae 22 Central Plan 25 18 Interest 15 18 Defence 13 14 Share of taxes 4 7 Other non-plan expenditure 12 State & UT Plan assistance 12 Subsidies 5 Non-plan assistance Fupee comes tom Corporation Tax Detence 19p. States’ share of Other non-Plan Taxes & duliles Expenditure 12p. portray a more than five or six categories by means of a pie chart. If, for ‘there are eight, ten or more categories it may be very confusing to the relative values portrayed specially when the several small of approximately the same size. This type of diagram, although used, appears upon comparison inferior to simple bar diagram, the diagram or a group of curves. ional diagrams mensional diagrams, also known as volume diagrams, consist of ders, spheres, etc. In such diagrams three things, 5 and height, have to be taken into account. Such range of difference between the smallest and the ‘or example, if two values are in the ratio of bars are used to represent them, the shortest bar would enerees Mo STATISTICAL METHOD, be of one thousandth part of the largest bar. If squares or ci then the side of net square or the radius of on proportionately too large or too small than the other. He used then their sides would be in the ratio of 1:10. This example makes it clear that three-dimensional diagrams have an important role to Play when the gap between the smallest and the largest value is very large. rcles are Used le circle would be pwever, if cubes are Limitations of Three-dimensional Diagrams ‘The three-dimensional dia grams have the same limiting features as the two-dimensional diagrams and to an even greater degree, Thus the side of a cube must be propor. Uonate to the cube-root of the magnitude to be represented. It 1s very diffi cult for the eye to read precisely such diagrams and hence they are not recommended for statistical presentation, Pictographs and Cartograms Pictographs are very popularly used in not abstract presentations such as lines or bars but really depict the kind of data we are dealing with. Pictures are attractive and easy to comprehend and as such this method is pafticularly useful in presenting statistics to the layman. When Pictographs are used data are represented through a pictorial symbol that is. carefully selected, ae constructing a pictograph the following points should be kept in Presenting statistical data. They are mi * The pictorial symbol should be self-explanatory. If we are telling a Story about ‘aeroplane, the symbol should clearly indicate” an aeroplane. The following points should be kept in mind while selecting a pictorial symbol: (a) A symbol must represent a general concept (like man, woman, child, bus) not an individual of the Species (not Hitler, Akbar or Dr. Sharma's car). (b) A symbol should be clear, concise and interesting. (c) A symbol must be clearly distinguishable from every other symbol. {d) A symbol should suit the size of Paper, te., it should be neither too small nor too large. (e) Last, but not the least, an artist should use the principles of a good design established by the fine and applied arts when drawing a pictorial symbol. + Changes in ni Ss are shown by more or fewer symbols, not by saith tres + should be simple to understand and convey essential ae ets w _Mustration 16. The following table gives the military balance between India and Pakistan:(MATIC AND GRAPHIC PRESENTATION Net inward Countries Net outward migration mipmalion 1,160 Mexico 400 istan 428 China 390 405 Pakistan a 362 ny 220 India 280 210 Iran 276 192 Indonesia 200 137 Philippines 180 120 Ukraine 140 alia 100 Kazhakhstan 120 80 ‘Sudan IN-BOUND VS. OUT-BOUND NET INWARD MIGRATION NET OUTWARD MIGRATION MEXICO USA AFGHANISTAN CHINA span [i eos PAKISTAN GERMANY @ 220 INDIA canaoa fet iran RR 27 uae fj 92 inoonesia | 200 UK i 137 PHILIPPINES UKRAINE AUSTRALIA [fj 100 KAZHAKHSTAN fj 12° RussiA J 90 suoan [os 16. The following table gives the net inward migration and net outwardee ee eee vee ee en r 2 ee ne as STATISTICAL Site Merit. Compared with other types of diagrams, pictographs have a greater attraction value and, therefore, where the attention of masses Is to be drawn such as in exhibitions, fairs, they are very popularly used. They stimulate interest in the information being represented. + Facts portrayed in pictorial form are generally remembered longer than facts presented in tables or in non-pictorial charts. . Limitation They are difficult to construct. Besides, it is necessary to use one symbol to represent a fixed number of units which may create difficulties. Thus if one symbol is representing 10 crore persons, the question is how to represent a population of 31.85 crores. In such a case either the symbol should be proportionately smaller or the figure approximated to 30 crores. In either case, error is introduced. + Pictographs give only an overall picture, they do not give minute details. For greater accuracy we should write actual figure under above or on one side of the symbol. Cartograms Cartograms or statistical maps are used to give quantitative information on a geographical basis. They are thus used to represent spatial distributions. The quantities on the map can be shown in many ways, such as through shades or colour, by dots, by placing pictograms in each geographical unit and by placing the appropriate numerical figure in each geographical unit. The cartogram given below shows Delhi's growing gender gap :ATIC AND GRAPHIC PRESENTATION = maps should be used only where ; Geographic comparisons are of 1S Salen approximate ole will ined For more ; poate . bar charts are preferable. To be sure, maps umes combined, which are drawn in the appropriate areas. of a Suitable Diagram ac. eae ones to select in a given situation is a a a * loice would primarily depend upon two factors. ie ao, ae data; and (i) the type of people for whom the n a g nature of the data would depend whether to use limen 1s o-dimensional or three-dimensional diagram, and if it pee . whether to adopt the simple bar or sub-divided bar, le bar or some other type. As already stated, a cubic diagram would preferred to a bar if the magnitudes of the figures are very wide apart. of people for whom the diagram is intended must also be . For example, for drawing attention of an uneducated mass, and cartograms are more effective than cubes, circles, etc. it types of diagrams such as bars, rectangles, cubes, pictographs charts have specialised uses, However, bar diagrams are most in practice. There are different types of bars and the appropriate of bar chart can be divided on the following basis: 4 bar charts should be used where changes in totals are __ required to be conveyed. "« Component bar charts are better suited where changes in totals as I] as in the size of component figures {absolute ones) are required ‘be displayed. ‘ c composition bar charts are better suited where changes tive size of component figures are to be exhibited. / bar charts should be used where changes in the absolute ‘of the component figures are to be emphasised and the over is of no importance. “multiple and component bar chart should be used only when more than three or four components, as a large number of e the bar charts too complex to enable worthwhile visual gained. When a large number of components have to be art is more suitable. particularly useful where it is desired to show the relative e figures that go to make up a single overall total. Unlike not restricted to three or four component figures although tends to dwindle with more than seven or eight charts cannot be used effectively where a series of figures is number of different pie charts are not easy to compare. Nor in the overall total be shown by changing the size of the circles are used to represent size. But it is difficult to and they should not be used when it ts possible to use ause it is easier to compare the lengths of lines or bars areas or volumes. peer aoa = STATISTICAL METHOD, Cubes should be used in those cases where the difference betwe smallest and largest values to be represented is very large. In other cases, wet should not be used because comparison is too difficult with the help of cubes. Pictographs and cartograms are very elementary form of visual presentation, However, they are more informative and more effective than other forms for presenting data to the general public who, by and large, neither possess much ability to understand nor take interest in the less attractive forms of presentation. The pictograph is admirably suited to the illustrations of exhibits or articles in newspapers and Magazines or for dressing up annual reports, Cartograms or statistical maps are particularly effective in bringing out the geographical pattern that may lie concealed in the data. GRAPHS A large variety of graphs are used in practice. However, here we shall discuss only some important types of graphs which are more popular. Broadly, the various graphs can be divided under the following two heads: + Graphs of time series, and * Graphs of frequency distributions. Constructing charts and graphs is an art which can be acquired through practice. There are a number of simple rules, adoption of which leads to the effectiveness of the graphs. However, before discussing these rules the elementary procedure of constructing a graph is considered. Technique of Constructing Graphs For constructing graphs, we make use of graph paper. Two simple lines are first drawn which intersect each other at right angles. The lines are known as coordinate axes. The point of intersection is known as the point of origin or the ‘zero’ point. The horizontal line is called the axis of X or ‘abscissa’ and the vertical line the axis of Y or ‘ordinate’. The alternative appellations are Y-axis and X-axis respectively. The following are the two lines: x scale on this kind of graph is in arithmetic progression, hence the term arithmetic line graph or arithmetic grid.GRAPHS OF FREQUENCY DISTRIBUTIONS A frequency distribution can be presented graphically in any of the following ways: « Histogram, + Frequency polygon. « Smoothed frequency curve, and * ‘Ogives’ or cumulative frequency curves.TIC AND GRAPHIC PRESENTATION 165 several methods of presenting a frequency distribution graphically, m is the most popular and widely used in practice. A histogram 1s yertical bars whose areas are proportional to the frequencies constructing histogram the variable is always taken on the the frequencies depending on it on the Y-axis. Each class 1s nted by a distance on the scale that is proportional to its al. The distance for each rectangle on the X-axis shall the same in case the class-intervals are uniform throughout. If mare different they vary. The Y-axis represents the frequencies of ‘lass which constitute the height of its rectangle. In this manner et a series of rectangles each having a class-interval distance as its and the frequency distance as its height. The area of the n represents the total frequency as distributed throughout the histogram should be clearly distinguished from a bar diagram. The lies in the fact that where a bar diagram is one dimensional, ‘the length of the bar is material and not the width, a histogram ensional, that is, in a histogram both the length as well as the portant. gram is most widely used for graphical presentation of a distribution. However, we cannot construct a histogram for with open-end classes. Moreover, a histogram can be quite if the distribution has unequal class-intervals and suitable its in frequencies are not made. que of contracting histogram is given below (i) for distributions class-intervals, and (i) for distributions having unequal als. ass-intervals are equal, take frequency on the Y-axis, the the X-axis and construct adjacent rectangles. In such a case of the rectangles will be proportional to the frequencies. s-intervals are unequal, a correction for unequal class intervals made. The correction consists of finding for each class the density or the relative frequency density. The frequency density juency for that class divided by the width of that class. A ‘or frequency density polygon constructed from these density d have the same general appearance as the corresponding splay developed from equal class intervals. the adjustment we take that class which has lowest and adjust the frequencies of other classes in the following one class-interval is twice as wide as the one having lowest we divide the height of its rectangle by two, if it is three we divide the it of its rectangle by three, etc., ie., the ill be proportional to the ratio of the frequencies of the width of illustration would clearly indicate the manner in which the has to be made: » o ram {s a bar chart or graph showing the frequency of occurrence of each M being analysed. —OpermannMustration 25. Marks 0-10 10-20 20-30 30-40 40-50 STATISTICAL METHop, Represent the following data by a histogram: No. of Students Marks No. of Students 8 50-60 60 12 60-70 52 22 70-80 40 35 80-90 30 40 90-100 5 Solution. Since class-intervals are equal throughout no adjustment in frequencies jg required. HISTOGRAM 60 2 8 . g 50 i 45 40 ‘4 & 35 30 8 2 3h bo 15 n 10 5 oO of © 10 20 30 40 60 60 70 80 90 100 F MARKS ‘26. Represent the following data by means of a histogram: , No, of Weekly Wages No. of a. Workers (in Rs.) Workers ante: be ol 7 30-40 12 40-60 12 60-80 8 (8. Com, (H)., Delhi Univ., B. Com., A.P. Univ.) rvals are unequal, frequencies must be adjusted otherwise ve a misleading picture. The adjustment is done as follows. The doi, hae 4000 bya nd YSAND GRAPHIC PRESENTATION 167 HISTOGRAM ° 10 15 2025 30 40 60 WEEKLY WAGES (IN Rs.) . Draw the histogram for the following data: Frequency Variable Frequency 1 140-150 S| 28 150-160 20 36 160-170 8 49 (B.A., Coimbatore Univ.) HISTOGRAM \ FREQUENCY of frequency distribution. It has more than Pee eating two or more frequency* We may draw a histogram of the given data and en. the mid-points of the upper horizontal side of each” PY -straighe the adjacent ones. The figure so formed is called juency an accepted practice to close the Polygon at both ends of ch with a PrPotetcy zero. This extension is made with the object of making the area polygon equal to the area under the Corresponding histogram "a readers are advised to follow this practice, » The + Another method of constructing frequency pol jon is mid-points of the various class intervals acd roe lot the (the corresponding to each Point and to join all these points by lines. The figures obtained would éxactly be the pont as obtat by method No. 1. The only difference is that here we have not ¢ construct a histogram. :. + By constructing a frequency polygon the value of mode can be ascertained. If from the apex of the polygon a Perpendicular js drawn on the X-axis, we get the value of mode. Moreover, frequency Polygons facilitate comparison of two or more freque distributions on the same graph. as Frequency polygon has certain advantages over the histogram: + The frequen: ons of several distributions may be plotted on the same axis, ee certain comparisons possible, whereas cannot be usually employed in the same way. To compare we must have a separate graph for each distribution. Because of this limitation for purposes of making a graphic comparison of frequency distributions, frequency polygons are preferred . + The frequency polygon is simpler than its histogram counterpart. + It sketches an outline of the data pattern more clearly. * lygon becomes increasingly smooth and curve-like as we eee the number of classes and the number of observations. juency faced construction polygon the same difficulties are hay histograms, fe cannot be used for distributions BE a classes and suitable adjustment, as in case of histogram, necessary when there are unequal class-intervals. i | from the following Mlustration 28. Draw pheooe and frequency polygon no 4ere ypaatic AND GRAPHIC PRESENTATION w pA ¢ class intervals are unequal, we shall have to adjust the frequencies. For Oo lass 20-40 would be divided into two parts 20-30 and 30-40 with frequency , xh class: Histogram & Frequency Polygon NO. OF STUDENTS ‘0°10 20 30 40 50 60 70 80 90 100 MARKS tthed Frequency Curve smoothed frequency curve can be drawn through the various points of the gon, The curve is drawn freehand in such a manner that the area included the curve is approximately the same as that of the polygon. The object drawing a smoothed frequency curve is to eliminate as far as possible ys be kept in mind. This would imply that the top of the curve would the highest point of the polygon particularly when the magnitude of dass-interval is large. The curve should look as regular as possible and sudden turns should be avoided. The extent of smoothing would, however, upon the nature of the data. If it is a natural phenomenon like tossing coin, smoothing may be freely resorted to as such phenomenon normally symmetrical curves, but if the phenomenon is social or economic the curve is generally skewed and as such smoothing cannot be carried too far. For drawing a smoothed frequency curve it 1s necessary to first draw the polygon and then smooth it out. As discussed earlier, the polygon can be constructed even without first constructing histogram by plotting the frequencies at the mid-points of class-intervals. This may save some time but the smoothing of the polygon cannot be done properly without a . Hence, it is desirable to proceed in a sequence, Le., first to draw 2 then a polygon and lastly to smooth it to obtain the smoothed frequency curve, This curve should begin and end at the base line and as a Tule it may be extended to the mid-points of the class-intervals just the histogram. The area under the curve should represent the total ‘umber of frequencies in the entire distribution.