Lab 7: Data Classification and Mapping
3/16/22 and 3/19/22
-
My lowest low percent change is -41.457 in 1900-1910
-
My highest high percent change is 82.0579 in 1970-1980
Is there a significant value in the data set that would make a good initial breaking point?
I chose 0 as my breaking point. One of the purposes of my map is to show population decrease in one color and population increase in another color, so this is the logical breaking point.
​
My classification scheme for my choropleth/graduated color maps:
50 to 83
30 to 50
20 to 30
10 to 20
0 to 10
-5 to 0
-15 to -5
-42 to -15
​
How and why I chose my classification scheme:
I first looked at lots of numbers and tried to find patterns between all the years. I started with the negative numbers first since that would be easiest. There were lots of numbers under -5 and there were a moderate amount of numbers between -5 and -15 but leaning more towards -13 and -14, therefore it would not be worth it to have a class of -5 to -10 only. In addition, there were lots of -20s, but not a lot of numbers between -20 and -15, which is why -15 was chosen as a break point. There were not many numbers below -15 so they all got lumped together in the final class.
​
There were very few numbers between 50 and 82 so they got their own category. Plus, those numbers tended to be in the high 50s and low 60s. My data had a decent number of high 40s compared to a lesser number of 30s and low 40s so the next category extends to 50. Otherwise, there would be no active 30 to 40 category on most maps, so by making it 30 to 50 most maps will have that category.
​
The challenge was figuring out 0 to 30 since that is where the bulk of the data was. So, I made a histogram of my data (note the data for population change from 1900 to 1910 and 1910 to 1920 are not displayed on this histogram because the system had a maximum of 500 numbers).
​
​
​
​
​
​
​
After lots of toggling with the classes on the histogram to see what landed where in different classifications, I decided to make 20 to 30 its own category. The classification 0 to 10, 10 to 25, and 25 to 50 also could work, since there are many high 10s and low 20s, but I decided not to go with that since the low 25s are too similar from the rest of the numbers in the 25 to 50 category. 0 to 10 and 10 to 20 both have lots of data points and the data within each category are spread out fairly evenly.
Finally, a more than +50% change is large no matter what and -31% to -15% shrinking is also large. In general, the big numbers are grouped together. There also tends to be fewer big numbers.
​
​
​
​
​
​
​
​
​
​
​
​
​
My color choices choropleth/graduated color maps and justification:
I used Colorbrewer. I used sequential coloring single hue for red and sequential coloring single hue for green. I didn't use diverging colors because I didn't like how it looked and I couldn't get 3 reds and 5 greens to show. While this is not color-blind friendly, I think the color choices make the most sense.
​
​
​
​
​
​
​
​
​
​
​
​
Problems with this lab and solutions to the problems:
My main problem was not double checking the drop down box to make sure I was using the right data set to be imported upon, but that was very easy to fix. I also had a problem with my 1910 to 1920 exclusion as I accidentally used the exclusion clause for 1900 to 1910 instead and when I tried to fix it I lost my whole map. To fix this, I just had to get rid of the exclusion and reset the data and the map came back.
​
3/28/2022
Classification scale of my graduated symbol maps:
7,700 - 15,000 persons
15,001 - 25,000 persons
25,001 - 35,000 persons
35,001 - 50,000 persons
50,001 - 90,000 persons
90,001 - 150,000 persons
150,001 - 250,000 persons
250,001 - 526,000 persons
​
Lowest low: 1980 - 7,797 persons
Highest high: 2020 - 525,534 persons
​
Justification of my graduated symbol maps' symbolization:​
I chose to use a filled circle as my symbol since it is compact and not too distracting as a symbol. I chose the background of my map to be a terrain-like color: light brown. I chose my circle symbol to be dark blue because I thought it looked nice and the darkness of the circle gives some contrast to the lightness of the background. My dot symbol size goes from 4 to 50. The 4 size is a small, yet noticeable size on my map and the 50 size fills the area of the smaller counties.
​
​
​
