School Work


of 19
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
   Survey Methods for Transport Planning Richardson, Ampt & Meyburg 395 11. Finalising the Survey   Once the data has been collected and the analysis has been completed, there is a natural tendency for you to think that the project is nearly finished, and that it is time to move on to something new. However, there are two major tasks still to be completed. First, the results of the survey must be communicated effectively to the sponsors and other interested parties. Obviously there is little point to conducting surveys if this task is not handled well. Second, there is a need for you to make sure that the data are stored in a manner which ensures easy retrieval by others. While you may not want to make further use of the data right now, it is most likely that either you, or more likely someone else, will want to use the data in the future; therefore it has to be easily accessible. This ease of accessibility covers both the physical media on which the data is stored (tapes, disks) plus documentation which will allow someone else to make sense of the contents of the data set. This chapter will address the issues and methods involved in the presentation of the results of the survey, in the storage of the data, and in the documentation of the survey method.  Chapter 11 396 11.1 PRESENTATION OF RESULTS It is a well-worn phrase that data is not the same as information. Data must be analysed, interpreted and presented properly before it can be classified as information. The analysis of data and the interpretation of results have been covered in the previous chapter; this chapter describes some of the techniques by which the information content of data may be effectively presented. In doing so, it draws upon two of the most readable and interesting books on the topic of graphical presentation of quantitative data that it has been my pleasure to read (Tufte, 1983, 1990), and which should be on the bookshelf of anyone involved in data collection and analysis. Tufte (1983) describes data graphics as devices which visually display measured quantities by means of the combined use of points, lines, a coordinate system, numbers, symbols, words, shading, and color . Although data graphics are a relatively recent invention (about 200 years old), the recent explosion in multi-media computer graphics capabilities has given many people the illusion that they can create attractive and meaningful graphics. If you are one of these people, then you should buy and read Tufte's books - quickly. As an underpinning to data graphics, Tufte (1983) lists several principles of graphical excellence which help to communicate complex ideas with clarity, precision and efficiency. These principles are: ã show the data ã induce the viewer to think about the substance rather than about methodology, graphic design, the technology of graphic production, or something else ã avoid distorting what the data have to say ã present many numbers in a small space ã make large data sets coherent ã encourage the eye to compare different pieces of data ã reveal the data at several levels of detail, from a broad overview to the fine structure ã serve a reasonably clear purpose: description, exploration, tabulation, or decoration ã be closely integrated with the statistical and verbal descriptions of a data set.   Finalising the Survey 397 Tufte states that graphics reveal data , and in this he is closely aligned with the ideas of Tukey (1977) and the role of graphics in exploratory data analysis; the first rule of any data analysis is plot the data . Thus, good graphics can be useful both in analysis and presentation; indeed, good graphical presentations encourage the reader (viewer) to conduct exploratory analysis beyond what is reported by the srcinal analyst. 11.1.1 Avoid Distorted Graphics One of Tufte's primary rules is to avoid the use of graphics which distort the data they are trying to represent. There are three classic distortions which unfortunately appear far too often in practice; the three-dimensional (3-D) pie-chart, the use of non-zero srcins, and the use of multi-dimensional objects to represent uni-dimensional data values. The 3-D pie-chart is often used to add glamour to a presentation which would normally only use a 2-D pie-chart. A comparison of 3-D and 2-D pie-charts which use the same data is shown in Figure 11.1. It is clear from the 2-D pie-chart that the values at the top, the right and the bottom are all the same size (they are each 8% of the total). However, in the 3-D pie-chart, it looks as though the 8% segment at the front (bottom) is bigger than the 8% segment at the back (top) which is, in turn, much bigger than the 8% segment on the right. The perspective view used by the 3-D pie-chart has distorted the data in the pie-chart (which is exactly what perspective views are supposed to do!). Unfortunately, many analysts use 3-D pie-charts without realising the distortion involved, because they know the values that they put into the chart and they don't see the distortion with an unbiased set of eyes. The distortion is particularly severe when value labels are not printed next to the pie segments, as in Figure 11.1. Figure 11.1 Comparison of 3-D and 2-D Pie-Charts  Chapter 11 398 The second major type of distortion is the use of a non-zero value as the srcin of the vertical axis of charts. The non-zero srcin distortion is used so frequently that one would think that it is an accepted part of graphical presentations. An example is shown in Figure 11.2. By having the vertical axis start at a non-zero value (in this case, 396), the trend in the data is much more pronounced in the chart on the left than when it is plotted at natural scale on the right. While close inspection will show that the data in the two charts are the same, that is not the impression given at first glance. When such charts are used in audio-visual presentations, where the viewer does not have much time to read the axis labels in detail, a very false impression can be conveyed. Unfortunately, the chart on the left is the one that was produced automatically by a popular spreadsheet package; the chart on the right was only obtained by manual intervention. 3963984004024044064084100100200300400500   Figure 11.2 Comparison of Non-Zero and Zero Origin Charts The third major type of distortion is the use of multi-dimensional objects, such as circles, cylinders and spheres to represent a one-dimensional value. Tufte (1983, pp 69-73) gives some lovely examples, but a simple example will suffice. In transport, one often tries to represent populations or trip destinations in a graphical format. A method that is often used is to use circles centred on the location, the size of which are proportional to the number of trips (or the population of an area). An example is shown in Figure 11.3. In the diagram on the left, the diameter of the circles is proportional to the number of trips, whereas in the diagram on the right the area of the circles is proportional to the number of trips. In both cases, the biggest circle is five times the smallest circle (either in diameter or area). In neither case, however, does the  biggest circle look five times bigger. When diameters are used, it looks much more than five times bigger, and when areas are used it does not look as though it is five times bigger. Neither representation is correct, however, because we are trying to use a two-dimensional object (a circle) to represent a one-dimensional quantity (the number of trips). This problem is becoming more widespread as
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks