Employee Resource Mobile Application; Evaluating Political Discourse on Twitter Platform

My STS research paper focuses on how and which social groups have asserted pressure such that Twitter’s algorithm amplifies political content moreso than others, while the technical portion explores my experience with developing an employee resource mobile application for IT Concepts, called CECI. The goal of CECI is to help streamline information for employees into one reliable place. Originally, most important data was located on Microsoft Sharepoint, but this application did not scale well as more information was added. Thus, CECI was developed which allows employees to access important announcements, internal tools, and stay in touch with fellow coworkers. Additionally, the motivation behind my STS research paper was to examine how algorithmic bias is present within social media currently. I especially wanted to focus on the political sphere within Twitter since politicians, such as Donald Trump, have had a major influence on people’s actions throughout the United States.
IT Concepts, a software consulting company located in Washington, D.C., determined that access to employee information was heavily decentralized, making it difficult to find information about announcements, pay and other resources. To solve this issue, I worked with a team of software engineers to develop a mobile app which would consolidate necessary information. The main tools used during this process were React Native, JavaScript, and AWS. We also used an Agile framework for the team which aided in adapting to various tasks and unexpected issues. We presented a successful beta version to leading executives, who determined that this application would be extremely useful for everyday employees, by providing users access to a multitude of resources and the ability to interact with coworkers. The app has the ability to centralize data and can be applied to other businesses, as well. Development is ongoing and new features and updates are needed in order to maintain the state of the mobile app.
In regards to my STS paper, I was able to utilize the SCOT framework to determine that the actions of ordinary users, politicians, and influencers along with their motivation to create trending posts have been instrumental in why Twitter’s algorithms amplify political content moreso than others. The literature review covers background information concerning algorithmic bias along with providing evidence regarding how Twitter tends to amplify right-leaning parties. Throughout my analysis, I investigated how various groups of users are interacting with the platform and common patterns in their online behavior. Investigating user activity and the patterns that make certain topics trending on this social media platform have an impact on what is being shown to others. I concluded that users on Twitter are drawn to maximizing elements such as likes, retweets, and comments. Thus, posting content that is shocking or divisive tends to get more attention. Tweets that are political in nature tend to attract opinions from all sides, hence why algorithms tend to favor similar posts.
Overall, I have gained a lot of insight through my experience with both projects. Working on the mobile app especially gave me a glance into the amount of work required to build a functioning product. As software engineers, we continue to test and update our applications, but sometimes there are details and behaviors that we may not realize are harmful until users perform testing. This retrospective insight does influence how I view the issue of Twitter’s algorithmic bias. Obviously the developers did not intend to specifically favor political content, it just so happens that a variety of factors - including how different user groups utilize the social media platform - have affected the resulting behavior of the system. Hence, we see just how important it is to be cognizant of the performance of an algorithm or application. As a future software engineer, I do realize the importance of incorporating diverse viewpoints and being able to adapt to different situations.

School of Engineering and Applied Science
Bachelor of Science in Computer Science
Technical Advisor: Daniel Graham
STS Advisor: MC Forelle
Technical Team Members: Myanh Le