Data visualization

DATA VISUALIZATION 6

Data visualization

Introduction

Data visualization is the presentation of data in a graphical or pictorial format (Cleveland, 1993). With data visualization, decision makers are able to identify new patterns, grasp difficult concepts and present the analytics visually. Visualization’s earliest seeds began with geometric diagrams represented in tables to show the positions of celestial bodies and stars of which the maps assisted in exploration and navigation.

History of data visualization

Ancient Egyptian surveyors used the idea of coordinates in laying out towns. Moreover, by 200 BC heavenly and earthly positions were located by something related to longitude and latitude (Maas & Morgan, 2005). For example, Claudius Ptolemy in 85BC-165BC presented in Alexandria a map projection of a spherical earth into longitude and latitude that would, until the 14th century, serve as reference standards. In the 17th century, some of the significant challenges were related to physical measurement such as space, distance and time that were used in territorial expansion, navigation, map making, surveying and astronomy (Ziemkiewicz, & Kosara, 2008). This period saw the dawn of practical application and a new growth in theory which engendered the rise of coordinate systems and analytic geometry.

Data visualization

(Source: Friendly, 2006)

Figure 1: History of data visualization

In the 18th century as shown in figure 1 above, the idea of graphic representation, data of interest and importance and some rudiments of statistical theory were established. In this period, these aspects were further expanded to new graphic forms and new domains. Map-makers in cartography showed more features in map and not just geographical position. New data representations, as a result, a number of inventions such as thematic mapping of physical quantities, contours and isolines took place (Friendly, 2005). The first half of the 19th century was evidenced by innovations of technique and design, and witnessed explosive growth in thematic mapping and statistical graphics. Nothing could equal this rate until the current times. By the mid-1930s, Wainer and Velleman (2001) observe that there were enthusiasm for visualization and few graphical innovations characteristic of the late 1800s. This was supplanted by the rise of formal statistical and quantification of models in the social sciences. Standard error became precise with parameter estimates and numbers (Friendly, 2006). In the 21st and the last quarter of the 20th century, the concept of data visualization grew into a multi-disciplinary research area that is mature and vibrant with the use of software tools for a wide range of data types and visualization methods using iPad, laptops or desktop computers.

Theory of visualization

Visualization in various empirical studies has been limited in use because there is infinity of different designs and that it applies to a specific design. According to Cleveland (1993), experimental studies use theory to generalize results, and in data visualization, the theory of perception is largely applied. Transforming data into visual patterns makes visualizations realistic and enable people to solve problems using pattern perception skills. Mapping data into visualizations is provided under theory-based guidelines based on scientific understanding of human pattern perception (Young, et al. 2006). A theory of visual thinking algorithms in addition to the theory of pattern perception are needed to describe processes that involve both computation activities and cognitive or perceptual activities. Perceptual activities are incorporated in visual thinking algorithm and involve epistemic actions, pattern perception and visual working memory. These include aspects like generalized fisheye views, brushing dynamic queries using sliders, computation activities and mouse or eye movements (Ziemkiewicz & Kosara, 2008). Making a decision to use a specific interactive process arises from the analysis of the efficiency of these computer and human algorithms. The theory needed for a discipline of visualization design is provided by applied perception and distributed cognitive algorithms.

Data Visualization designs

. For example, data of international visitors touring Australia can communicate their intention and expenditure. In the figure 2 below, readers can immediately know that most visitors to Australia come for holiday followed by those who come to visit their friends and relatives. Maas & Morgan, 2005) century, has reached the masses through commercial software products after being popularized in tragically ineffective ways. The products that promote data visualization are superficially appealing, sense-making, effective data exploration and aesthetics (stData visualization since the turn of the 21

Data visualization 1

Figure 2: Visitors by main purpose of visit

. In the pie chart, it does not show any values of expenses by state although it can easily be learned that New South Wales had the highest expenditure. Shneiderman, 1996)Pie chart is one of the traditional modes to display information graphically. As shown in figure 3 below, the pie chart indicates the nature of relationships clearly and makes obvious how people should utilize information. However, it does not compare quantities, or rank order of values, and does not compare quantities easily (

Data visualization 2

Figure 3: Expenditure by state

On bar charts, as shown in figure 4 below, it displays the same information which can be perceived readily. It presents the data accurately, indicates the nature of relationships, compares quantities easily, ranks order of values and allows people to utilize information such as the case of visitor numbers to Australia.

Data visualization 3

Figure 4: Bar chart of top markets by visitor numbers

With regard to the line graph, it is more powerful because apart from presenting the data accurately it allows for trends and patterns to be observed. For example, the researcher can be able to follow the total expenditure on education for tourist over a span of 10 years as shown in the figure 5 below.

Data visualization 4

education’s share of total trip spendFigure 5: line graph of

Conclusion

Data visualization, since the medieval times to the modern times has facilitated encoding of information, and is only successful in a way the human brains can understand and eyes can discern. As seen in the examples of tourist data in Australia, getting this right is not art but science. This can only be achieved by studying human perception and visual representations that are meaningful, accurate, efficient and easy. By doing so, it makes it easy to translate and decode abstract information.

References

Cleveland, W. S. (1993). Visualizing Data. Hobart Press.

Friendly, M. (2005). Milestones in the history of data visualization: A case study in statistical historiography. In C. Weihs and W. Gaul, eds., Classification: The Ubiquitous Challenge, (pp. 34–52). New York: Springer.

Friendly, M. (2006). A Brief History of Data Visualization. York University. Springer-Verlag.

Maas, H. & Morgan, M. S. (2005). Timing history: The introduction of graphical analysis in 19th century british economics. Revue d’Histoire des Sciences Humaines, 7(2): 97–127.

Shneiderman, B. (1996). The Eyes Have It: A Task by Data Type Taxonomy for Information Visualizations. John Wiley and Sons.

Ziemkiewicz, C. & Kosara, R. (2008). The Shaping of Information by Visual Metaphors, Transactions on Visualization and Computer Graphics. Proceedings InfoVis, 14(6): 1269- 1276.

Wainer, H. & Velleman, P. F. (2001). Statistical graphics: Mapping the pathways of science. Annual Review of Psychology, 52(2): 305–335.

Young, F. W., Valero-Mora, P., & Friendly, M. (2006). Visual Statistics: Seeing Data with Dynamic Interactive Graphics. New York: Wiley.