Cartography: Final Project - Thematic Mapping of Two Sets of Data

 Map Content   

     This is the last project for this course. The culmination of this project is a single map displaying two datasets. As a final, it tests our ability to design and implement cartographic analyses independently while integrating skills learned throughout the semester. We were given the choice of two data sources (either SAT or ACT testing data), each with two sets of data to map. I chose to use the testing data published by ACT, Inc. I mapped average test score and percentages of graduates tested, by state.
     I used Excel to create a table that could be subsequently imported into ArcMap and joined to the basemap data. I analyzed the data in Excel to see if it was normally distributed. It turns out that neither dataset is normal. (This can be attributed to the fact that in some states taking this test was a requirement while in other states participation is voluntary. Participation in turn affects the average score since more students [of varying aptitudes] are taking the exams in the mandatory states the average score is representative of a broader student base). This is important when considering how to classify and portray your data. After this preliminary data analysis, I worked on preparing my basemap. The basemap data comes from the US Census Bureau and lack a projection. I used Albers Equal Area Conic as it is appropriate for this landmass and the preservation of area (important for preserving the enumeration unit in choropleth maps which is how I wanted to map part of the data).     

     Perhaps the data could have been better represented in a different way but I intuitively leaned toward a chorpoleth basemap and some sort of proportional or graduation symbology. I decided to map the percentage data on a choropleth map while representing the average score with graduated symbols. Since the data is reported by state that caused a light bulb in my head to go off at the thought of enumeration units (states) which brought me to choropleths and data classification. Some classification methods are better at taking the data distribution into consideration (natural breaks, optimal) while others are not so good (equal interval, quantile). I used classification methods in the latter category because they ultimately provided a better visual representation of the data. For the choropleth data: when I experimented with natural breaks and equal interval (I decided against using standard deviation because it is not as easy for a map user to intuit) with 3-7 classes, most of the country ended up being represented by a single class/color. This is largely due to the fact that a substantial proportion of the country (roughly half) had greater than 70% participation. Since the quantile method places an equal number of observations into each category, a greater distinction between classes emerged and I found that 7 classes nicely represented that data. Additionally, the median of the data will fall in the middle class in a quantile classification using an odd number of classes. For the average score data, I used the equal interval method with 3 classes. The scores are tightly clustered on the number line so I thought the range of data was better served by a few classes that were easily distinguishable and interpreted.  

Map Design     

     At first I used the color ramp and symbolization choices found in ArcMap to asses and plan the overall design of my map. I chose, however, to use CorelDraw x7 to compose the final map. Using CorelDraw allowed me to customize and fine tune my map in greater detail than could be done in ArcMap. I used Color Brewer 2.0 to chose a color ramp. I chose a multi-hued, sequential color ramp that allows the map user to easily distinguish between classes (in this case low to high participation percentages). I went with circular graduated symbols to represent the average score data and gave the symbols a gradient fill to make them look spherical (bringing them off of the page). I tried to design a custom symbol but it was not as easy to decipher, confused the pattern in the data, and looked less cohesive than the circles. (Thus, I placed my custom pictograph as the title border so it felt like my efforts were not in vain). I utilized drop shadows extensively to create figure-ground contrast between the elements of my map and sized the contiguous US as large as possible (while still leaving room for other map elements). You can see my map below and I hope that you find it pleasing to the eye, understandable, and informative.

A map of the percentage participation and average scores achieved
by US high school graduates on the ACT for the year 2013. 

     I learned so much from this course. I look at maps with a much more discerning eye and deeper level of understanding. I tend to see a map almost daily by way of news or social media and I appreciate them much more now that I have begun honing my own cartographic skills. Thank you for visiting my blog and participating in my journey as a cartographer. I have a new way to communicate information and I plan to make thorough use of it throughout my career.