The focus of this project was to create a publishing quality map
illustrating the relationship between 2014 Composite Mean SAT Scores and the
test participation rates by graduating seniors within the conterminous United
States and the non-contiguous states of Alaska and Hawaii. The map is intended to accompany an article
in the Washington Post discussing high school seniors and college entrance
scores.
In order to best present the bivariate
data, I chose to create a choropleth map with graduated, range-graded symbols. A choropleth map is ideal for displaying
uniformly distributed data within each enumeration unit where change occurs at
the enumeration unit boundaries. The
composite SAT scores are shown by applying a color gradient scheme to the
States. I chose to class the composite
mean test score values using a defined interval of 100 with six classes. While this reasonably represented the data, I
wanted to better convey the high end outliers.
By tweaking the last break point to cover the range between 1,800 and
1,816 I was able to represent the data more accurately. I selected a suitable six class color scheme
using color brewer. (Brewer, 2002-13) Superimposed on each state is a graduated,
range-graded symbol illustrating the graduating senior test participation rate. The 2014 participation rates were classed
using Natural Breaks (Jenks) with seven classes. I chose to show the class ranges using the
feature values to accurately depict the percentages even though this showed
percentage gaps in the legend; the reason for the gaps was explained within the
map subtext. While the other
classification options (Equal Interval, Defined Interval, Quantile, Geometrical
Interval and Standard Deviation) provided reasonably similar results I chose to
use Natural Breaks (Jenks) because this method minimizes the difference between
values in the same class and maximizes the difference between classes. Once I had properly classified the data I
shifted my focus to developing my layout and design. The conterminous states served as the primary
map; I included insets for Alaska and Hawaii since the page size restricted
their visibility within the primary map.
For clarity, I created an inset map of the Northeastern United States so
that the State names and graduated symbols could be easily distinguished. I added all required map elements, but kept
stylization to a minimum within ArcDesktop. Finishing effects were created using Adobe
Illustrator with attention to Gestalt principles. The Gestalt theory, developed in the 1920s,
describes how humans recognize individual components in a graphic image and
organize them into a unified whole. The
Northeastern inset map applies closure which is our ability to complete an
image even with parts missing; in this instance we visually interpret the
dashed line to represent the square in its entirety, without gaps. In order to apply figure-ground, I made the
primary map elements a lighter color than the background. I also applied stylization effects including
feathering and drop shadow so that the objects appeared to be visually closer
to the reader. I feel the feathering
effect helped to emphasize the importance of these items and direct the
reader’s attention to them. By making
the graduated symbols darker than the states, I was able to apply visual
hierarchy so that the symbology was emphasized without detracting or obscuring
the underlying states. While all of the
previously discussed items are important to an accurate map and good
cartographic design they are all for naught if an incorrect projection is
used. Because we were comparing data by
area it was important to choose a projection which preserved area. There are four main equal area
projections: Globe, Sinusoidal Equal
Area, Lambert Azimuthal Equal Area and Albers Equal Area Conic. By looking at the size and directionality of
my subject matter these choices could be further narrowed to those suiting
small regions or countries resulting in either Lambert Azimuthal Equal Area or
Albers Equal Area Conic. Narrowing by
extent, I selected the Albers Equal Area Conic which is suited for large,
mainly east/west areas of mid-latitude.
Additionally, this projection is a good choice for thematic and
presentation maps. ArcDesktop offers
further refinement to this projection by setting the longitudinal center to
-96, the first standard parallel to 20, the second standard parallel to 60 and
the latitudinal center to 40. I chose
this projection, designated as North America Albers Equal Area Conic for my
project.
This project allowed me to
utilize the cartographic methods, techniques and skills I had acquired during
the semester. The importance of
visualizing the data distribution, defining the map purpose and audience became
clear when attempting to design a map with no detailed instructions. While daunting at first, the task quickly
became enjoyable as I was able to recall and apply what I had learned
throughout this course. Because of the
experience gained this semester, I was able to create the ArcDesktop portion of
the map with relative ease. Although I
had not used Adobe Illustrator in the more recent labs, I was pleased to note
that I was able to recall and apply the skills I had learned. As I was less intimidated by the software, I
was able to explore new options and find better ways to manipulate the objects
that I had imported from ArcDesktop.
Comparing the .mxd file to the .ai file leaves no doubt that using Adobe
Illustrator greatly improves the maps presentation value. Not only did this project make me an acolyte
of Adobe Illustrator, it also strengthened my confidence in the skills learned
throughout this semester. I am more
enthusiastic about the remaining coursework and the possibilities that GIS
brings to my future.
