Você está na página 1de 6

www.statistics.me.

uk

How to Present Information


In many areas in life including business and management, government, health and social services, and science and engineering masses of data is being generated. This could be primary data such as a set of readings made by a measuring instrument or a pile of completed questionnaires or secondary data such as a set of company accounts or government statistics. Masses of raw data cannot normally be directly interpreted by the human brain and hence we often need to find and communicate the meaning of a given set of data. The following factors should be considered in the presentation of data: (i) (ii) (iii) (iv) (v) The subject matter should be clearly presented, The purpose of the presentation, The amount of detail that is required, The method or presentation, Articulation of the most important messages.

Typical methods of presenting information include tables, graphs or charts. For example consider the following data presented using these methods. An estate agency sells and lets property. The income from its sales business over the four quarters of the year for two years are 2010: Q1: 15,000, Q2: 18,000, Q3: 14,000, Q4 9000 and in 2011 Q1: 14,000 , Q2: 15,000, Q3: 11,000 and Q4: 7,000 whereas its income from its lettings business in 2010 was Q1: 5000, Q2: 8,000, Q3: 10,000, Q4: 12,000 and in 2011 was Q1: 15000, Q2: 20,000, Q3: 22,000, Q4: 23,000. The data may be tabulated as follows. Year Quarter Sales Lets Q1 15,000 5,000 2010 Q2 Q3 18,000 14,000 8,000 10,000 Q4 9,000 12,000 Q1 14,000 15,000 2011 Q2 Q3 15,000 11,000 20,000 22,000 Q4 7,000 23,000

The same data can be represented in a graph1, for example as follows

Income from Sales and Lets 20102011


25 Income 1000s 20 15 10 5 0 2010 Q1 2010 Q2 2011 Q3 2011 Q4 2012 Q1 2012 Q2 2012 Q3 2012 Q4
1

Lets Sales

Plotting a Graph

www.statistics.me.uk

Whether the data is preferred as a table or graph, a number of conclusions can be drawn from the representations of information. The income from lets is steadily rising, whereas the income from sales appears to be on a downward trend. There seems to be some seasonality in the data with higher income in the summer months. There are a number of methods for illuminating the point that the income from lets is increasing over the income from sales. One method is by using pie charts. The following pie charts shows the changes in the proportions of income for the two years. Since sales were greater in 2011 then the pi chart for 2011 is shown to be larger.

2010 TOTAL

2011 TOTAL

Sales Lets

Sales Lets

Classification of Data Primary data may be obtained from a number of sources for example from surveys, from measurement or they may be computer generated. Before we can represent data we often have to process it and classify it. Firstly there are two basic types of data that we have to handle; discrete and continuous variables. The term variable is used in the sense that numbers are being handled but the numbers can take any one of a set of values and hence the term variable. Discrete Variables Discrete means separate. If any possible value of the variable to be presented takes one of a set of fixed values then the variable is said to be discrete. Most often, discrete variables are whole numbers. For example if we surveyed the number of occupants in cars passing a point in the road then all the values would be whole numbers. However, discrete variables do not have to be whole numbers. Continuous variables If we measured and recorded the heights of a set of people then the height may take any value within a range, although the data may still be recorded to a set accuracy, for example to the nearest cm (in which case they are technically discrete variables).

www.statistics.me.uk

Frequency We are often interested in how often something happens or its frequency. As an example let us assume that 20 students in a class sit a test and their marks are as follows. Student Mark Student Mark 1 5 11 9 2 6 12 6 3 3 13 8 4 8 14 4 5 4 15 7 6 5 16 5 7 7 17 6 8 7 18 2 9 3 19 8 10 7 20 7

A more compact way of representing the data is to count the number of students that achieved each mark, as follows. Mark Frequency 0 0 1 0 2 1 3 2 4 2 5 3 6 3 7 5 8 3 9 1 10 0

Table 1. Frequency of test marks. Classification or grouping of Data With large amounts of data it is useful to group data into intervals. The example above shows an example of grouping data. Let us look at another example. In an exam 120 students received a percentage mark. In order to condense the quantity of data, the student marks are tallied into intervals of 10%. Mark range 1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 Frequency 1 4 6 12 20 35 26 11 3 2

Table 2. Frequency of exam marks in 10% ranges. The advantage with a table like Table 2 is that is can reduce a large quantity of data to a small table. However, although the essence of the information is maintained, there is some reduction of information; Table 2 does not tell us what the actual marks are.

www.statistics.me.uk

Histograms and Frequency Polygons A histogram is a graphical representation of information, in which frequencies are represented by bars, the width of each bar is equivalent to the width of each interval, the area of each bar is equivalent to the frequency of the data within that interval. A frequency polygon is obtained by linking up the tops of the histogram blocks. The following diagrams show a histogram and a frequency polygon representing the data in Table 1.

Histogram
6 5 Frequencye 4

3
2 1 0 0 1 2 3 4 5 Mark 6 7 8 9 10

Histogram based on the data from Table 1.

Frequency Polygon
6 5 Frequency

4
3 2 1 0 0 1 2 3 4 5 Mark 6 7 8 9 10

Frequency polygon based on the data from Table 1.

www.statistics.me.uk

The following diagrams show a histogram and a frequency polygon representing the data in Table 2.

Histogram
40 35 30 Frequencye 25 20 15 10 5

0
1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 Mark

Histogram based on the data from Table 2.

Frequency Polygon
40

35
30 Frequency 25 20 15 10 5 0 1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 Mark

Frequency polygon based on the data from Table 2.

Cumulative Frequency Curve Another useful chart that can be created from data like that in Table 1 or table 2 is the cumulative frequency curve. In order to develop a cumulative frequency curve, the first step is to form a table based on the frequency table in which the frequency accumulates. For example for the data in Table 1 the

www.statistics.me.uk

cumulative frequency for mark 2 is 0+0+1=1, the cumulative frequency for 3 is 0+0+1+2=3. The following table is the cumulative frequency table based on Table 1. Mark Frequency 0 0 1 0 2 1 3 3 4 5 5 8 6 11 7 16 8 19 9 20 10 0

Table 3. The cumulative frequency data based on Table 1.

Cumulative Frequency
25 Cu mulative Frequency 20 15 10 5 0 0 1 2 3 4 5 Mark 6 7 8 9 10

Cumulative frequency curve for Table 1 Similarly the following graph is the cumulative frequency curve for Table 2. The graphs have a characteristic squashed-S shape.

Cumulative Frequency
140 Cu mulative Frequency 120

100
80 60 40 20 0 1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100 Mark

Cumulative frequency curve for Table 2

Você também pode gostar