Purpose – To create a Bivariate Choropleth map that simultaneously represents two variables that are commonly used to gauge socio economic development, by viewing each map separately then compare both maps together, to identify relationships that may exist.

 

Method – The data was downloaded from The City of Calgary Open Data Portal and re projected to 3TM WGS 1984 W114. Using the Children Under 5 and Seniors Over 65 fields within the attribute table.

 

Three classes were chosen for each variable for the Bivariate map, which created nine classes, represented as a three-by-three matrix. Showing a combination of low, medium and high values. The Quantile Classification method was chosen as there were no empty classes and the data values were treated appropriately.

 

A new instance of the demography layer was added to the data frame with two new fields added to the attribute table. The newly created field for Children under 5 was re classed into 3 classes of A=low, B=medium and C=high values. Next the same was done for Seniors Over 65 field, naming the re classed data 1=low, 2=medium and 3=high values.

 

Next, a new field was added to the attribute table and named CH_CODE (Children under 5). An attribute query was used to select the percentage of children that were equal or smaller than seven percent. While the attribute query was open, a Field Calculator was used on the CH_CODE to update the value to “A”. The same was done for the next two classes of using the Select by Attributes tool for the percentage of children equal or smaller than 13% and 48% and reclassified as “B” and “C” in the Field Calculator.

 

The process above was used for the Seniors Over 65 variable, using the following values:  <= 7% and re classed a “1”, <= 13% and re classed a “2” and <= 48% and re classed a “3” into a new field called SE_CODE.

 

A new field was added to the attribute table and named BIVAR_CODE and the following field calculation was used: “[CH_CODE] & [SE_CODE]” to concatenate the values.

 

Under the symbology tab for the demographic layer, the Unique Values option was selected and the BIVAR_CODE field was selected.

 

A generic legend was then created. All elements were ungrouped, and all labels deleted. Under the legend item properties tab, the override default patch size check mark was removed. The number of columns were set to three, 30 points

square with zero gaps between the patches.

 

In the legend properties, under the symbology tab the classes were then re-ordered from top to bottom: A3, A2, A1, B3, B2, B1, C3, C2, C1. The primary colours of CMYK red (horizontal colour ramp) and CMYK blue (vertical colour ramp) were chosen and when combined, created CMYK purple as the secondary colour. Colour increment occurs along each vertical, horizontal and diagonal axis. A frequency box was added above the matrix to show the frequency of each class.

 

Lastly, in the layout view, the data frame was split into 4 separate frames. Each individual map was scaled to accommodate all the required layout elements of, legend, titles, sub titles, scale bar, north arrow and shadows.