History of Statistics One view is statistics is originated from the Latin word ‘status’. Another view is that it originated from the Italian word ‘statista’. German word ‘statistik’ or French word ‘statistique’ Reference of births and deaths in ‘Arthashastra’ by Kautilya’s; Records of agriculture in ‘Ain-i-Akbari by Abu Fazl. Meaning The word statistics is understood in two ways ~ plural and singular. In plural sense it may be defined as data qualitative as well as quantitative, that are collected, usually with a view of having statistical analysis. In singular sense it may be defined as, the science of collection, presentation, analysis and interpretation of numerical (quantitative) data. Statistics is both art and science. Variables The characteristics of population are called variables and they may vary for each item. Variables can be divided into two types: a) Quantitative variables. b) Qualitative variables. Quantitative variables represents numerical facts. Qualitative variables are such which represent qualitative phenomena like fairness of skin, etc. Variables can be classified in another manner: i) Discrete variables or Discontinuous variables and ii) Continuous variables. Data Collection Data are quantitative and qualitative. Data are secondary and primary. Secondary Data Data, which were collected earlier for purposes other than the present investigation, are known as secondary data Secondary data are collected from internal and external sources and such sources are referred to as secondary sources. They may be available in the form of printed papers, books, articles, newspapers, CDs or on Internet.presentation of Data in the process of decision ~ making, vast variety of data has to be gathered from numerous sources. To understand them clearly they need to be simplified and classified. Classification of data refers to the division of data into some classes. Similar types of data are grouped in one class. The purpose of classification is to ensure that data be grouped and presented in a manner so that their analysis can be conducted properly. It can be summarized as follows: i) It condenses the data and ignores unnecessary details; ii) It facilitates comparison between data; iii) It help in studying the relationships between several characteristics; iv) It facilitates further statistical treatment. Data may be classified into four categories 1. Quantitative or Ordinal Classification: When the data are classified in respect of a variable, say height, weight, profits, salaries etc, they are known as quantitative classification 2. Qualitative or Cardinal Classification: These are such facts that cannot be measured on natural numerical scale but are grouped on the basis of some distinctive characteristics or attribute. For example: Fairness of skin of Indian girls, nationality, gender, smoking habit. 3. Chronological or Temporal or Time Series Classification: Sometimes data are classified according to time. For example: exports of a firm in different years; agricultural production in India in various seasons. 4, Geographical or Spatial Classification: Data are also arranged according to the geographical boundaries. Number of cars in India, Nepal, Pakistan, etc, is an example of geographical classification. Frequency Distribution Frequency distribution refers to the division of quantitative data either individually or in to different groups on the basis of their magnitude. In other words, it is can arrangement of ordered array ina manner so that data are written along with their frequency / frequencies. Formation of frequency distribution is essential for the purpose of proper analysis of data This arrangement in done in the form of three series: Y Individual series or raw data Y Discrete series or ungrouped frequency distribution Y Continuous series or grouped frequency distribution Term associated with frequency distribution 1. Class Limit: The two ends of a class. Each class has two limits, lower limits and upper limits. 2. Class Boundries: The two ends of a class after making it continuous type. 3. Class width: Class width refers to the difference between upper and lower class boundries.Class Mark: Mid point of each class is class mark. Lower limit + Upper Limit Class Mark = _ Or Lower Boundry + Upper Boundry 2 5. Relative Frequency equency of a class Relative Frequency = Fal eee 6. Percentage Frequency Percentage Frequency = Relative Frequency x 100 7. Frequency Density Fre class Frequency Density = reauency of « class Width of the class Open Ended Classes: Ifthe first class is less than 10, subsequent classes are 10-20, 20-30 and last class is 30 & above, such an arrangement is referred to as open ended classes. Mode of presentation of Data 1) Textual presentation 2) Tabular presentation 3) Diagrammatic presentation 4) Graphic presentation Textual Presentation Data presented in paragraph / paragraphs. It does not require any special knowledge of statistics. Conversation with exact magnitude can be presented in this way. This is the first method or beginning for other methods. However, it is not preferred by statistician because it is dull, monotonous and cannot be used for comparison. Tabular Presentation Presentation of data in the form of table is very useful. The following points should be kept into consideration while constructing a statistical table: 1) Table Number: Every table is allotted a number. 2) Heading: Every table should be given a heading. It should not be too lengthy. 3) Head-note: At the right hand side of the top of the table a head note may be given to specify the units of measurement. 4) Captions and stubs: Data in statistical tables are given in columns and rows. Column headings are known as captions and row headings are known as stubs, 5) _ Body of the table: This is the main part of the table and contains all data. The numbers of the columns and rows of the table should be such that all information may be easily fit in Footnote: if some explanation is required to be given about a particular figure, in the footnotes. 6) it may be given7) Sources: If data are collected from some secondary source, it is better to mention the details of such a source. Diagrammatic Presentation One of the most popular manners of presenting data is through graphs. They are quite eye-catching and well understood by common people who do not have sound statistical background. Any hidden trend present in the data can be noticed only through diagram but compared to tabulation this is less accurate. Data can be presented through a large variety of graphs. While drawing a diagram appropriate horizontal and vertical scales should be chosen. The drawing should be neat and simple. 1. Line diagram / Line Chart / Historiagram Values are represented by either vertical or horizontal lines, The length of a line varies as per the value. Distance among various lines is arbitrarily fixed and the same distance is maintained among all lines. Example: Represent the export figures for a period of six years for ABC Ltd. TET * 1% Solution: Export 2001 2002 2003 2004 2005 2006 Year 2. Bar Diagrams / Bar chart Diagrammatic presentation of data is very popular in business Presentations and has become a ‘avorite technique In case of computer presentations. There are a number of bar diagrams some of them, which are more in use, are explained here.Single Bar Diagrams To show individual values of a single type we draw single bar diagrams. The height of ie bar depends upon the value. Width of all the bars is taken to be equal and the gap between different bars is the same. Single bar diagrams can be drawn vertically or horizontally. Through single bar diagrams one can show quantitative data as well as qualitative attributes. : Example: The following figures represent the number of employees working in difference departments in a manufacturing company in a particular year. Present these figures through single bar diagram (horizontally and vertically). Category of employees No. of employees Electrical department 8 Mechanical department 12 Heating department 5 Packing department 7 Transport department {oD | 14 No, of employees working in different department No. of Students Electrical Mechanic Healing Packing | 8 12 5 7 9 | = Series 1 Transport Category of employees Multiple Bar Diagrams / Compound Bar Diagrams When more than one feature of an observation or time are to be shown simultaneously, such a presentation is done through multiple bar diagrams. For example, if we want to show the number of presentation is done students who appeared for different categories of ICAI examinations in May, 2004, multiple bar diagrams are a suitable technique. Like single bar diagram, they can be shownhorizontally as well as vertically. Height of each bar depends upon the number of items in each observation. Example: Draw a multiple bar diagram (horizontal and vertical both) from the following information Number of male and female students appeared in May, 2004 examinations of ICAI, Delhi: : P.E-2 : El FINAL _ Male | 32,000 24,000 Female | 8,000 16,000 | 35,000 "Vale Female 30,000 25,000 20,000 No. of Students. 15,000 10,000 5,000 PEI PEI Final Exam 3. Pie Charts In pie chart a circle (pie) is drawn to represent data. Sum of all data is equated to 360 degrees and each data is converted into degrees. Then data are plotted in the circle in proportion to their degrees. Pie charts are often used to represent the budgetary allocations to various departments of the government of India, expenditure on different groups in a family, etc. Example: The following table gives the distribution of outlay in a Five Year Plan of India under the major heads of development expenditure: Head of expenditure Expenditure (in crores of 2) Agriculture and community development 8,000 Irrigation and power 4,000 Industry and mining 7,000 Transport and communication 5,500 Miscellaneous 2,500 Total 27,000 Represent the above information by a pie chart.Solution: For drawing a pie chart the total expenditure is equate expenditure are converted into corresponding degrees which are calculated 2s: Expenditure (in crores of ) Head of expenditure Agriculture and community development Irrigation and power Industry and mining Transport and communication Miscellaneous Total Pie chart is drawn according to the degrees. 8,000 4,000 7,000 5,500 2,500 27,000 d to 360 and then !Neividyal Degrees 107 53 93 B D Agri. & Comm. Dev., [ Irri. & Power; Ind. & Mining; Trans. &Comn.Q Misc. Graphic Presentation of Frequency Distributions Frequency distribution can also be presented graphically. Histogram, frequency polygon, and ogive curves are used graphically show frequency distributions. 1. Histogram Histogram is a graphic representation of a frequency distribution — continuous series. Each class is represented by a rectangle. The height of the rectangle on Y - axis (vertical — axis) shows the frequency of the class and while the class intervals are shown on X ~ axis (horizontal ~ axis). The number of rectangles in a histogram is equal to the number of classes of the frequency distribution. It is important to remember that histogram is not suitable for individual series. Example: Draw a histogram from the following data Profits (in & Millions) 10-20 20-30 30-40 40-50 | 50-60 | No. of firms 36 45 68 19 1280 70 60 505 40 No. of Firms 30>} 20>] 10 10 20 30 40 50 60 Profits (% Million) ( 2. Frequency Polygon Frequency polygon is a curve drawn from a histogram. The midpoints of all the classes are plotted on the X-axis and above these midpoints dots for frequencies are marked on the Y axis. Then all the frequency points are joined like a curve. The ends of the curve are extended in both sides. 50 | 45 40 35 30 No. of shirts sold 36 38 40 42 44 46 Size3. Ogive : Ogive is a graphic presentation of cumulative frequency distribution. For drawing 2n OBIve, class intervals are shown on X ~ axis and cumulative frequency is plotted on Y~ axis. Ogive are employed to show positional values such as median, quartiles, etc. Ogive can be drawn in two ways Less than ogive: To draw this type of ogive, we choose the upper limits of the clases; against the first upper limit we put a dot on Y — axis to show its frequency. For the second upper limit obtain the cumulative frequency by adding the first and the second - class frequencies and mark the cumulative frequency on the graph and repeat this step for all subsequent upper limits. All the points on the graph are jointed. This gives a smooth rising curve. Example: From the following data of monthly house rent paid by certain number of families draw a less than ogive. Monthly house Rent No. of Families ‘Monthly house Rent | No. of Families | paid paid (in) _ 7 100-149 So mnnt20 450-499 40 150-199 27 500-549 35 200-249 36 550-599 25 250-299 47 600-649 7 300-349 49 650-699 14 350-399 44 700-749 3 400-449 43 Solution: First of all this inclusive series is converted into exclusive series then cumulative frequencies are obtained. 350 + Cumulative no. of families y S Ss PPPOPPEOPPEOH Rent paidMore than ogive: For a more than ogive chose lower limit of all class; for the first lower limit select the total frequency, for the next lower limit subtract the first frequency from the total frequency, repeat this procedure for all subsequent lower limits. Plot the frequencies on the graph; joint all the points and that gives a smooth falling curve. Example: Draw a ‘more than’ ogive curve for the following data _No. of children No. of families More than 0 - 150 | More than 1 72 More than 2 50 More than 3 28 | More than 4 12 | More than 5 8 | More than 6 5 Solution: 160 More than 140 » 120 3 = 100 = 80 6 g 6 8 2 40 8 2 2 5 co No. of children 4. Line Graph , In line graph, first of all scales are fixed for X and Y axis. X axis represents time or independent variables and Y axis shows dependent variables. Dependent variables are plotted through points on the graph paper and then these points are joined. The line / shape emerging from the joined points is known as line graph. ; Since the line graph begins from zero value on Y axis, therefore, in case of very high values the line graphs looks very shabby and disproportionate. In such cases we draw a false base line. A false base line is a set of two zigzag lines parallel to X axis beginning from a convenient value on Y axis leaving aBap for some values. A break in Y axis is shown to represent values not mentioned IN fs axis. in the following example, the values begin from 130, therefore, on Y axis we begin our scale from zero and then break it, create a false base. This false base may begin either from 100 or 130. Example: Closing balance of advances to big industries by a nationalized bank in different years in shown in the following table. Draw a line graph from these figures. ] ] [Years 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | 2003 | 2004 2005 | Advances (in&] 130 | 160 | 190 | 240 | 240 | 290 | 300 | 310 | 350 | 370 | crores) 2 Solution: 400 350 7 300 + 250 -P 100 Advances (% Cr.) 8 \ t 50 7 T T T T T T T T T 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Year Two - lines graph or Balance graph For one independent variable, sometimes there may be two independent variables. For these two independent variable, points are plotted on the graph and one line is drawn for one variable and the other line is drawn for the other variable. The area between these two lines represents the balance of the two variables. This type of graph is suitable to show profit (area), revenues and cost; balance of payment (area), exports and imports, etc. Example: Export and import figures in trillions of rupees are mentioned below. Represent these figures through a two — lines graph.2003 | 2004 | 2005 7a aon! 2000 | 2001 | 2002 | | | |__*e2rs_| 1996 | 1997 | 1998 | 1999 | Exports | 35 | 45 | 51 ) ag 62 58 64 | 67 | | Imports 20 34 41 45 | 56 52 | 60 | 70 | 62 | 60 | Solution: 80 + 70 + 60 +> c S = ~~ : rT TT a 2003 2004 2005 1996 1997 1998 1999 2000 2001 2002 Year - Export Import