A New View on Immigration
How visualization can help people to better understand immigration issues.
Based on my personal experience as an immigrant in the United States, I would like to understand some of the official information available. I believe that an effective data visualization can help us all to better understand and analyze data that otherwise can only be interpreted by professionals and scholars. My goal is to create a better visualization of current immigration data that is publicly available and distributed by official channels and a report system maintained by Syracuse University.
Aside from focusing on strictly immigration-related data, I would like to cross reference with election results by state. This will paint a better picture of the relationship between deportations/removals and political affiliation of the states where the deportation/removal orders are issued.
This project was divided into three stages where the main goal was to come up with a familiar yet effective data visualization that was easy to understand by anybody, including people who are not fluent in English.
The first stage is to create color-coded data tables. This makes it easier to sort the data and brings to light some patterns that will generate new questions. Many of the questions in this stage were related to the connection between the amount of deportations by state immigration court. There are multiple variables that play a role in this issue, such as the political affiliation of that state in the last presidential election as well as the physical location of the state with relation to the points of entry and the bordering countries.
The second stage included generating views in draft form, first coming up with design ideas of multiple visualization techniques that can encompass the requirements of the projects, while addressing some of the questions from stage 1. The most natural way to represent this turned out to be a map of the United States. Since just a map seemed to not be enough, the design phase resulted in a dashboard with three elements. This dashboard displays information for election years from 2008 to 2017 (2018 and 2019 statistics have not yet been released by the US Department of Homeland Security). The colors that dominate the dashboard are red and blue to characterize the two political parties. These simple elements are intuitive for any audience at any education level. Therefore, making the map the most prominent element in the display will allow the user to analyze the amount of removal orders issued by state and the political affiliation of the state at first glance. The other two elements in the dashboard follow the design lines from the map, and are a bar chart to represent the margin by which party won that state in the last election, and a table that shows the statistic on immigration decisions by each immigration court in every state, the number of pending cases, and the number of granted decisions by state.
The final step on this stage was to generate the views on draft form using Tableau. Thanks to its flexibility and immediate response in visualizing different datasets, it was very easy to obtain a working prototype of the final product.
The third stage of the project generated the views in a website using the JavaScript library D3.js and basic HTML/CSS. The data was contained in a CSV file for this project, but the script is flexible enough to handle a different formats such as a JSON file for future implementations. This could enable using a web service to deliver up-to-date data on immigration and future elections.
Among all the resources of immigration data publicly available, the closest related to my project are projects that have been released by nonprofit organizations such as TRAC and the Pew Research center. They mainly utilize information from the U.S. Department of Justice and U.S. Department of Homeland Security to generate reports. While the charts and tables presented in TRAC are great— and partly the motivation for this project — they are intended for people who know the field and understand the data. They may be too complex for users who are not professionals.
By far the most elaborate and interesting sector to represent data on immigration and politics is the media, utilizing multiple techniques to achieve content delivery. Big news corporation like the New York Times have utilized elegant data visualizations to bring to light information retrieved mostly from official sources. But none of the ones reviewed in my research create a link between the political leanings of a state and the immigration data reported by that state immigration courts in a determined period of time.
This project resulted in a simple-to-use tool that empowers the viewer to easily question and find patterns in the data provided by official channels, moving away from understanding technical language and how the immigration system works in the United States. Using just a map, a bar chart and a table, this data visualization allows the user to intuitively find information about immigration and elections results without the need to visit multiple websites or scrolling through official data charts. By simply hovering the mouse pointer over a state, the bar chart and table are dynamically populated with the immigration and election data corresponding to that estate. The bar chart colors represent republican and democratic results in the state, and the table in the lower right shows immigration rates for that particular year and estate.
The final product of this project can be found here.
