10. Working with Charts
Chart Element Properties Dialog box 
10. Working with Charts
Charts are another data
visualization tools. We will learn how to display the data or the relationship
between sets of data in charts in ArcView chart document.
Suppose you are
interested in the demographic characteristics for each county in New York
State. With ArcView, we can visualize different attributes in charts as well as
in maps. You can download the demographic data from U.S. Census Bureau Census
2000 gateway as well as spatial boundary from U.S. National Atlas. County boundary
(shapfiles) is added in a view, and Census data (dbase file) is added in a
table. Since we are only interested in NY state rather than the entire country,
we can select features using Query Builder, and then we can convert the
selected features into a separate theme. (The Convert to Shapefiles… from Theme
menu converts the selected features when some features are selected, and
converts all features when nothing is selected.)
Add converted shapefiles
as a theme. You might want to delete one feature that does not have an
attribute. You can do so by choosing the Delete Records from the Edit menu in a
table. (You can do the same thing in a view as well.)
Summarize a theme table
on the field [Fips]. As for county, FIPS is a 5 digit code that is used to
identify administrative boundary in the U.S.. For example, 36029 is the FIPS
code for Erie County, where the first two digit refers to New York State. In
other word, FIPS is the unique identifier of each county feature. A summary
table shows the number of records for each FIPS. Unfortunately, one county may
not necessarily have one feature like shown below. Actually we can merge them
based on FIPS.
The Summarize has an
option for merging shape if the table is a theme table. In Summary Table definition
dialog box, choose Shape for Field, choose Merge for Summarize by, and press
Add button. In the similar way, you can add more field such as County name to
the summary table. When you choose Merge_Shape, it will become a shapefile
rather than just a table. Now 7 features consisting of Jefferson County are
merged to one feature. The census data will be joined to this merged county
data, where the common field is FIPS code, the census data is a source table,
and the county data is a destination table.
Based on the attributes
joined from census, you can make a couple of thematic maps such that they can
help you explore easily how different age groups are distributed in each
county.
It may be useful to take
a closer look at some features in a chart. Suppose you are in charge of
reporting the health care status in each county. You may want to compare the
proportion of elderly persons in some counties to investigate the possible
correlation between health care spending and the proportion of a certain age
group. Just select the feature using the Select Feature 
tool, or whatever the way you prefer.
Then open the theme table 
.
Now Table is an active
document. Click the Create Chart button 
. Chart Properties dialog box will pop up.
In the Chart Properties
dialog box, you can add data groups or series by choosing any numeric fields
available in Fields list and adding them into Groups: list. A chart data group is a set of related values that describe
the same variable. By default, ArcView considers the data from a field in a table as a data group. A chart data series is a set of related values used for
comparison in a chart. By default, ArcView considers the data from a record in a table as a data series. The Label series
using: list allows you to specify a field used for labeling each item in data
groups or series on the chart.
|
FIPS |
COUNTY |
PERCENT_OVER_65 |
|
36025 |
Delaware County |
19 |
|
36029 |
Erie County |
16 |
|
36041 |
Hamilton County |
20 |
|
36057 |
Montgomery County |
19 |
In this case, we use the
field [Percent_over_65] as a data group. Each record (county) can be seen as a
data series, and the field [County] is used to label each item. The field names
become the group labels as shown on the x-axis. In a chart, different colors
represent a data series (in this case, records – which county).
Now we are going to
modify the look of the chart using Chart Element
Properties . The
y-axis is scaled between 16 and 20. To change the minimum and maximum scale of
y-axis value, click the Chart Element Properties tool 
, then
click the mouse on the y-axis area. Chart Axis Properties dialog box will pop
up. Type 5 for Scale min text box, and type 20 for Scale max text box. You can
also add grid to make it easy to read the value, or axis label. Press OK button.
To change the title,
click Title on the top.
Type the new title and
press OK button.
You can change the
color. Click the Chart Color tool 
. Choose
any color, the click your mouse on the bar you wish to work on.
The chart data series are
associated with the records in a theme table as well as map features. If you
want to add more counties in the chart, you can just select features in a view
or records in a table.
Conversely, you can
deselect in chart as well. Click the Erase tool 
, and
then click the bar you want to deselect. For example, deselect Hamilton County.
The deselected one
(Hamilton county) is deselected in a view also. You can select and deselect in
a view, table, and chart interactively.
This time, suppose you
want to draw the bar chart for two fields such as percentage age group under 5
and over 85. Chart Properties dialog box allows you to specify the fields used
for data groups. You can choose more than one field depending on what types of
chart you have in mind.
The field name is
labeled on an x-axis by default. If you use an alias for the field name in a
Table Properties dialog box, the alias will be used for labeling the x-axis.
ArcView groups the data
by the fields by default. What if you want to group them by the record, in this
case County? You can switch data groups and series by clicking the Series from Records/ Fields button 
.
Can we make other types
of chart like Pie chart? Yes. For example, you are interested in the race
composition in each county. Unlike Bar chart, pie chart is usually good for
displaying multiple related fields. We add one field Click the Chart Properties
button 
. Delete
fields in Groups list. Select race percentage fields in the Fields list, and
add them to the Groups list by clicking Add button. You can change the order of
fields as well by clicking and dragging each field in the Groups list.
Click the Pie Chart
Gallery button 
.
Remember how to select
fields (how many fields, in which measurement scale such as normalized (e.g.,
percentage) or not (e.g., number)) depends on the chart type. For example, if
you want to draw x, y scatter plot, you should choose two fields. As for pie
chart, choose more than one like shown above. The following chart shows the
example of making population pyramid using bar chart.
The followings are the
example of making a chart on the summary table. Suppose you are interested in
how fatal accident occurrence has changed over time, in Erie County. Use Query
Builder to select records that meet your criteria.
Summarize the selected
records on [Year] field.
The summary table shows how many
accidents occurred and how many people died from the accident by year. Now you
can create a chart from this summary table. Select Create Charts button 
.
Select Sum_Fatals for
Groups list, and Year for Label series using list.
