Introduction
Thus far I have inputted full data for 38 of the claims, from the files that I photographed at the National Archives during the summer before class. I have partial data for 26 more--the numbers that Dr. Peter Jonas listed in his dissertation and that I have entered into my database (Jonas lists many more). The visualizations on this page are based on the 38-claim dataset.
Map
Initially I was planning to use TimeMap, following the lovely tutorial by Erin Bush. My classmates rightfully warned me that creating a map where you could use a timeline might be a bit much. So instead, I followed Laura O'Hara's lovely tutorial for making a Google Map with my data points and a historical map overlaid.
I could have stopped with the BatchGeo portion, but I thought I wanted to include a full historical overlay because of the change in borders; some of the incidents took place in what was then Mexican Texas. I chose an 1826 map of North America, as it shows an independent Mexico, but before Texas split away in 1836 and the United States seized the northern portion of that country in 1848.
But alas, although MapWarper worked beautifully (as it should when one plots 112 points for alignment), the map came out badly (blurry and pixelated) in Google Maps. So in the end, I chose not to use it.
This map shows the home port of the ships involved in the claims (only if, of course, a ship was involved), and the locations of the incidents. You can click on the pin for the place and see a case (I haven't had luck getting it to show multiple cases) related to that place. Click on the case to find out more details.
As this map shows with the clustering of incidents in port cities, many of the incidents involved trade. Indeed, quite a few of them involved disputed customs duties.
Note: the dots in the pins indicate the locations.
Graphs
I tried to use this tutorial from StackOverflow so that the charts would automatically query the database, and thus automatically update as I input more data. Alas, I didn't have luck getting it to work, although I may try in the future so that each chart is automatically updated as the database is.
So instead, I needed to get the data and code it for the chart. I have two types of charts: the number of incidents per year, and the total amounts claimed and awarded by year. I exported the data from my queries as CSVs, formatted into the arrays for which Google Charts calls. I did a bit of manual manipulation (since there were only 38 claims, and many fewer years, it was quicker than writing a Python script). I then put those into the JavaScript for each chart.
Number of claims by year
This chart shows the number of claims filed for incidents that took place in each year. Because the dataset is so small at the moment, not as many conclusions can be drawn. However, certain peaks and troughs are visible: claims are particularly high between 1815 and 1821, when U.S. citizens supplied arms to rebels and sometimes participated directly in Mexico's War of Independence. 1824 has more claims because several merchants protested an increase in tariff, which they were forced to pay in spite of not being informed when their ships left port. Meanwhile, in 1835 and 1836 Mexico was engulfed in rebellions (most notably, in Texas) and U.S. merchants found their ships confiscated and had other issues.
Total claim amounts by year
I originally ran a chart showing the amounts of each of the 38 claims for which I have incident dates. However, that chart proved too large and unwieldy. So instead, I ran a query for the sum of the total claim amounts, and sum of the amounts actually awarded, for each year in which I have an incident date.
This chart shows much larger peaks and troughs, mainly due to certain outliers that sought extremely large claims.
For that reason, I chose to use one of Google Charts' features: the logarithmic scale. This scaled the chart in a way like the Richter scale: by multiples of ten. It smooths the lines, but of course can be a deceiving visualization if you don't look at the scale!
