 Tutorial
This Tutorial takes you, step-by-step, through the process
of creating Dynamic Choropleth Maps.
Step 1 - Topics
Click on a Topic to expand it.
The Data Selection Box contains several
data "Topics", subject areas, such as Air, Water, TRI, Human Health,
and Demographics.
Each Topic is a Data Cube which can be expanded by clicking
to access its elements (data sets). Open a topic by clicking it with the mouse.
Step 2 - Data Sets
Select a Data set of interest by clicking it.
In this case "Lung Cancer Mortality" is the selected data set.
The next step is to assign the data set to the map.
Step 3 - Mapping data
Select a map component to receive the data.
Click one of the Map titles (shown boxed in red) to drop the data.
These are the data components of the map. When the mouse moves over a map title
the cursor changes to a "Hand" to indicate that the title can be clicked.
The North data component is the primary
component and determines map coloring. The other two components, the South and East
components, are data filters which are used to investigate interactions with
the primary component. Repeat steps 1 to 3 to load the other components.
You can expand other "Titles" in the data tree to select from other data sets.
Note: If you don't want any filtering,
just uncheck both filters, this will allow you to observe the primary data
component without any interactions.
Step 4 - Data filters
Enable Data filter(s) to investigate interactions with the primary data component.
A data filter is enabled by checking the box next to the slider.
Here we have enabled the South filter and disabled the East Filter to focus
on interaction between "Lung Cancer" the primary data component, and "Percent Black".
Step 5 - Interpreting the Map
Interpret Slider Percentages and Percentiles.
The slider shown above for "Percent Black" shows four percentiles above it, and
three percentages below it. The slider end-points are used to control the
percentages, while the percentiles are automatically calculated.
The percentiles are from left to right:
The minimum county value (in this the minimum is 0 Percent Black);
the second percentile (in this case the 33% percentile is 0.4 Percent Black);
the third percentile (in this case the 67% percentile is 6.2 Percent Black);
and, the maximum county value (in this case 86.7 Percent Black).
This means that county "Percent Black" goes from a low of 0 to a high of 86.7,
and 33% of counties have Percent Black less than 0.4, and
67% of counties have Percent Black less than 6.2.
The middle 34% corresponds to counties with Percent Black between 0.4 and 6.2,
inclusive. All sliders are interpreted in the same way.
Step 6 - Data Interactions
Move sliders to investigate data interactions.
The slider can be moved as a scroll bar by dragging it or it can be resized.
To resize the slider place the mouse over one of the ends
and drag it to the new position. The mouse pointer will change to a splitter
when the mouse is placed over a slider end.
Now move the slider for "Percent Black" to observe the effect on county "Lung Cancer
Rate" as we look at counties with higher and higher "Percent Black."
As the slider is moved only those counties
are shown whose data values are within the range shown above the slider.
For example,
only those counties with "Percent Black" from 0.4% to 6.2%, inclusive, are shown.
The other counties
with values less than 0.4% or greater than 6.4% are filtered out (shown in black).
We also see that 34% of counties have "Percent Black" within this interval.
The map will redraw dynamically as a slider is moved.
Step 7 - Using Sliders
Select Map percentile ranges for the primary data component.
The North slider is used to define the percentile ranges for map coloring.
Move the slider or the slider end-points to alter ranges. The slider
end-points set the percentages of the distribution, and the percentiles
are automatically calculated.
The slider shown below shows four percentiles above it, and
three percentages below it. The percentiles are from left to right:
The minimum county value (in this case 10.7); the second percentile
(in this case the 33% percentile is a rate of 53.5 persons per 100,000),
the third percentile (in this case the 67% percentile corresponds
to a rate of 66); and, the maximum county value (in this case a rate of 171.1).
The extreme values of data can be shown by
moving the lower end-point of the red band to the right, or by moving the
upper end-point of the cyan band to the left. For example, to isolate
the 10% of counties with the highest Lung Cancer Mortality rates
move the slider as shown.
Note: The three color bands can be reduced to two, or one, by merging the
slider end-points.
Step 8 - Mouse over
Mouse over a county to identify it and its data values.
Step 9 - History
Use the history feature to recall previous maps. This is similar to the browser
history feature. Each rendered map is stored in the history and can be recalled
by using the left and right history arrows as shown.
Step 10 - Drill Down
Right click the map to drill down for more information on a county. The first two entries
-- Analyzer and Work Offline -- are not drill downs. TRI is based on EPA TRI Explorer,
facilities reports are based on EPA facility reports, Enviromapper is an EPA facility
mapper, Scorecard is an Environmental Defense Fund site, and Census and Fedstats are
demographic sites containing census information.
Step 11 - Analyzer
Use the Analyzer to transform data using mathematical expressions and Boolean logic.
Right click the map and select "Analyzer" to create data transformations.
Data variables associated with the South, East, and North Sliders, are identified
as X,Y, and Z, respectively. A transformation of the form f(x,y,z) is assigned to
x, y, or z. The function f is formed using the operators +, -, *, /, ^, (, ) and some
intrinsic functions and Boolean operators. For example:
z = 100 * (z-x)/x assigns z as the percent change of z with respect to x. A dialog box
will allow the transform to be input as well as an optional title. Click OK to run the
transform and map the transformed data. Several intrinsic functions are available to
be used in the transform such as abs(), exp(), log(), and sqrt(). The Boolean operators
<, =, >, !, <=, >= are also supported to allow the data to be subsetted using logical
conditions. For example, z=(z ! x) will show only those counties where z is not equal
to x. These Boolean operators can be used to create a general filtering of counties that
sliders cannot. For example, if Z is "death rate all cancers, white females", and X is
"death rate all cancers, white males" then the transformation Z = (Z > X) will show Z
in only those counties for which female death rates for all cancers exceed those of
male, a rather rare occurrence.
|
