Access to Prenatal Care

Grant Dixon

For this project I decided to use a data set from the World Development Indicators from the World Bank. The data set chosen indicates the percentage of women with access to pre-natal care. I thought this data would be an interesting case study for information visualization. It is able to display both the advantages and disadvantages of the visualization of information. I created four visualizations of this dataset. One is a bubble chart, one is a bar graph, one is a world map, and one is a matrix chart. These four visualizations display the information in the data set in different ways and at the same time reveal information that was previously unattainable.

The first visualization I created was the bubble chart (Figure 1). This chart excels at visualizing the change in number of counties polled over time. As you adjust the “Bubble Size” parameter the amount of bubbles, which signify, countries shifts. This graph demonstrates how over time more countries were being polled about their women having access to pre-natal care. The chart not only shows which countries have been polled, but it also displays the percentage of women in said country with access to pre-natal care. There is a drawback to this chart though. As more countries are added, the maximum size of the bubbles decreases. This creates the impression that the counties polled in 1990 have a higher percentage than those collected in 2000, when in fact more countries had larger percentages in 2000. In the end this visualization provides quite a bit of information. It allows us to see the change in amount of countries polled and see the results of those polls compared against other polls of the same year, but it suffers from the lack of uniformity in its scales.

The second visualization is a bar graph (Figure 2). While this type of visualization can be considered “simple” it is really quite the opposite. This visualization in particular reveals that a majority of counties polled over the years have had relatively high percentages of women with access to pre-natal care. It also reveals the differences between the countries polled. It is more striking to see a tall bar next to a short one than it is to see a high number next to a smaller one. The concept of difference between the two are heightened. This chart also presents a flaw in the dataset. When you switch from year to year in the legend, you begin to see an inconsistency in the data collection. A country with collected data one year does not have any the next. It makes it difficult to map out the change in access to pre-natal care over time and frustrates the viewer.

The third visualization is a world map (Figure 3). The data in the dataset was transferred into map form. The country’s percentage is displayed as a specific color. This map is useful because you can pinpoint where advances in access to pre-natal care is changing. It shows areas such as in Asia where access to pre-natal care expands into new countries. This visualization could be a powerful tool, but it comes up the shortest of all the visualizations. This visualization depicts a profound flaw in the data collection process. You are able to see that some areas of the world have been completely ignored or have only been polled once or twice, while others are polled every year or so. It shows that the data collected is not completely reliable. It is impossible to describe the state of access of pre-natal care for women around the world if there is not data from every country. The visualization depicts an incomplete and flawed data set.

The fourth visualization is a matrix chart (Figure 4). This visualization is the most complete and decipherable. It presents the information in a clear and concise manner. By having the countries represented as different colors and placed at a certain percentage, one can immediately know what percentage a specific country had in a certain year. This chart also depicts the countries not polled that year. This is helpful because none of the other visualizations adequately provided this information. This chart allows for the comparison between different countries and for the critique of the dataset as a whole. It shows all of the data in the dataset in one place and also displays the deficiencies of the dataset in a cumulative manner.

I find these visualizations interesting because as you look at the visualizations next to each other, the data being visualized projects different meanings. Figure 1 displays how many countries are being polled every year in a simple manner while Figure 3 displays how many countries are not being polled on a regular basis. Basically, Figure 1 hides the fact that countries are not being counted. It only focuses on the data collected. On the other hand, Figure 3 focuses on all the countries and vividly depicts areas where there is a lack of data. If you analyze Figure 2 for example, you are slightly overwhelmed. There are many bars in the graph and it is hard to read. You easily see the difference between countries with higher percentages compared to those with smaller ones. On the other hand if you place that same data in Figure 4 you are able to visually see where every country is at percentage wise. Instead of being categorized by country, the figure shows the percentages as the constants and makes the countries the variables. As you change the graph the countries move instead of the percentages like in the other examples.

Overall, information visualization is an important tool. It allows for a person to examine data in unconventional ways and it allows a person to create new ideas or new data from existing datasets. It also enables a person to be able to see how complete their data collection is and lets them address areas that need improvement with relative ease. In the end, information visualization should be used in any field which uses data to make arguments because it is a conduit to simply communicate information to others thru visual means.