Lab 9 - Bivariate Choropleth & Multivariate Maps
Being productive is always a great feeling, but when you learn to be productive in map-making -- it's even better. In this week's lesson we received a lecture about different types of multivariate maps -- that is, maps that show multiple variables. There are different ways of doing this, such as side-by-side comparison, overlaying or combining multiple map types, and/or multivariate/bivariate choropleth mapping. For example, you might have seen proportional symbols on top of a choropleth map, or perhaps pie-charts on top of a choropleth map. If you haven't, I provided samples below from our lecture slides:
But let me tell you about "bivariate choropleth maps"...
For our lab assignment we learned how to create a "bivariate choropleth" map to examine relationships between two variables: obesity and physical inactivity. Bivariate choropleth maps are awesome and incredibly useful. Why? Because this technique contributes to a better understanding of the relationships between two variables (whatever they may be). These types of maps allow us to visually see strong, weak, positive, and negative correlations in color swatches. Simply put, bivariate choropleth maps allow us to see two variables at once. Instead of depicting two separate maps in one page (each map displaying a univariate choropleth map), we can it get done in one, simultaneously. Specifically, in this week's lab assignment where we investigated the links between obesity and physical inactivity in U.S. counties, the bivariate choropleth map clearly demonstrates where the correlations are high (southeastern states), and where the correlations are low (midwest). The bivariate choropleth map can also be a space saver in an infographic because of its productive quality of showing multiple data at once.
Creating the legend
Though there are different rules depending on what type of data is being displayed, for our lab assignment we created a 3x3 table legend where the relationship between the variables were quantitative and sequential in data ranges (as opposed to divergent). The approach involved using a Quantile Classification, Adding [new] Fields in the attribute table, using the SQL Query Builder to select attributes as well as using the Field Calculator to populate the new fields with "coded" class values. Choosing the colors of the 3x3 table legend involved choosing one single hue with increasing saturation (darkness) for each variable, and then manipulating the HSV percentage values. I used the color schemes provided in Joshua Steven's website, which gave the hexadecimal color codes for the purple/yellow theme I chose (see map at the very top), and then converted these values to HSV using this site: www.color-hex.com.
Furthermore, I decided to use a labeling design similar to the ones from Joshua Steven’s website, where the words “Low” and “High” (with an appropriate directional arrow) indicates the progression of colors with the associated relationship values. Then, underneath the arrows, I used bolded titles for what each side of the 3x3 table represented: % Obese and % Physical Inactivity. I liked this labeling design best because I appreciated it the most when I saw it myself in other map examples. I think it’s one of the easiest ones to read.
For our lab assignment we learned how to create a "bivariate choropleth" map to examine relationships between two variables: obesity and physical inactivity. Bivariate choropleth maps are awesome and incredibly useful. Why? Because this technique contributes to a better understanding of the relationships between two variables (whatever they may be). These types of maps allow us to visually see strong, weak, positive, and negative correlations in color swatches. Simply put, bivariate choropleth maps allow us to see two variables at once. Instead of depicting two separate maps in one page (each map displaying a univariate choropleth map), we can it get done in one, simultaneously. Specifically, in this week's lab assignment where we investigated the links between obesity and physical inactivity in U.S. counties, the bivariate choropleth map clearly demonstrates where the correlations are high (southeastern states), and where the correlations are low (midwest). The bivariate choropleth map can also be a space saver in an infographic because of its productive quality of showing multiple data at once.
Creating the legend
Though there are different rules depending on what type of data is being displayed, for our lab assignment we created a 3x3 table legend where the relationship between the variables were quantitative and sequential in data ranges (as opposed to divergent). The approach involved using a Quantile Classification, Adding [new] Fields in the attribute table, using the SQL Query Builder to select attributes as well as using the Field Calculator to populate the new fields with "coded" class values. Choosing the colors of the 3x3 table legend involved choosing one single hue with increasing saturation (darkness) for each variable, and then manipulating the HSV percentage values. I used the color schemes provided in Joshua Steven's website, which gave the hexadecimal color codes for the purple/yellow theme I chose (see map at the very top), and then converted these values to HSV using this site: www.color-hex.com.Furthermore, I decided to use a labeling design similar to the ones from Joshua Steven’s website, where the words “Low” and “High” (with an appropriate directional arrow) indicates the progression of colors with the associated relationship values. Then, underneath the arrows, I used bolded titles for what each side of the 3x3 table represented: % Obese and % Physical Inactivity. I liked this labeling design best because I appreciated it the most when I saw it myself in other map examples. I think it’s one of the easiest ones to read.
Very creative title! It catches the map viewers’ attention, whilst remaining informative to what the map is about. Overall this blog post is very well done!
ReplyDelete