Undergraduate Writing

All Semesters

literature review


The Model Minority, Online: Examining AAPI Identities on Reddit [Research Proposal] (Nicolas Lab, Spring 2022)

In spite of America’s growing diversity, there is a lack of research considering relations between non-White minority groups. Notably, there are comparatively fewer works exploring Asian American people, who, in particular, hold a distinctive place in the social hierarchy in which they can realize their own identity in a way that either embraces or rejects Whiteness and consequently impacts other racial minorities with less privilege. Meanwhile, in digital environments, racialized communities thrive. Asian American people have taken to Reddit to explore their identities, and in the midst of social movements like Black Lives Matter, the question arises: how do AAPI people constructing their identities perceive people of other races? Racial identity theory informs how people perceive themselves and others through race, as well as how they behave in their ingroups. Intergroup contact lies adjacent to it, providing a framework to analyze interactions between groups. Through these lenses, one can perform a thematic content analysis on moderated discourse in notable AAPI subreddits on Reddit to explore the perceptions of other races from an AAPI perspective and how they may have changed over time. Completion of this study can result in enormous benefits to the digital activist space by informing Reddit moderators of strengths and weaknesses in their practices, creating a more structural understanding of online communities, identifying which aspects of discrimination are most salient, and more.

Social Media Bias: A Review of Psycho-Digital Feedback Loops (Nicolas Lab, Fall 2021)

People have always consumed media in selective and partisan ways. As social media become dominant forms of news and entertainment, users introduce personal biases to the platforms they use, and platforms respond by amplifying those biases. Recommendation systems established for commercial gain aim to increase engagement and consequently create feedback loops of bias through individual and collaborative personalization that affects digital communities on both local and global scales. Literature reveals that social media algorithms create filter bubbles, homogenize groups, and prioritize popular opinions over minority ones, thereby increasing the strengths of pre-existing beliefs, polarization, ingroup loyalty, and misrepresentation. A related, albeit weaker, consequence of social media influence manifests in the backlash effect, in which fact-checkers have counterproductive consequences and have further impacts on civic engagement. Despite dire implications, researchers demonstrate that various factors can mitigate the effect of bias and, and they work toward developing systems geared toward fairness and diversity to alleviate sociopolitical strife.

The Quality of Q&A Communities [Literature Review] (InfoSeeking Lab, Spring 2019)

In recent years, online Q&A services have grown in popularity. By posing and answering questions, users have created whole communities based around information exchange. Researchers seek to understand these communities and must consider the problems users face, as well as the factors involved in a successful search for information. A literature search was conducted to assess the different approaches to improve community Q&A platforms. Three publications were selected for the review. They examine datasets from popular sites to study various factors. The methodologies implemented in the studies primarily feature predictive frameworks and mixed-method analysis. All three studies achieved conclusive results and thus greater understandings of community Q&A services and how to enhance them.

Term Papers

The Digital Nudge [Research Proposal] (Cognition and Decision-Making, Spring 2022)

A branch of Richard Thaler and Cass Sunstein’s “nudge theory” considers dark patterns, which are digital features “crafted to trick its users into doing things that are not in their best interests” (Özdemir, 2020). While different in intention and ethics, traditional nudges and dark patterns are similar in the cognitive functions they activate. This research proposal will consider the question: does awareness of a dark pattern lessen its effect? The study will consider a range of dark patterns (bait-and-switch, misdirection, trick questions, etc.) and modalities (mobile and desktop devices) to investigate the ways in which awareness of a pattern differs depending on the pattern, as well as how this form of digital nudging compares to traditional nudging.

Digital forms of nudging are not novel, and in the twenty-first century, choice architectures that exist within technology are increasingly relevant. While more benevolent forms of nudging may involve “e-mail, SMS, push notifications, mobile apps, social media, gamification, e-commerce, e-government, location services” (Özdemir, 2020), social media platforms, with business models that rely on user engagement, implement mechanisms to increase activity, user interaction, and overall consumption, even at the expense of user welfare (Chaney et. al, 2018). Thus, it is important to further research dark patterns to better understand their manipulative strategies, detriments to human choice, and ethical implications. In doing so, we can begin to consider alternative platform models that preserve user autonomy, encourage transparency, as well as protect larger establishments of justice and democracy. The hope is that this study could provide insights into the similarities and differences between dark patterns and traditional nudges and determine whether disclosure of a dark pattern is an effective mitigation strategy.

News Consumption and Voter Turnout: A Statistical Analysis in Stata (Political Science Methods, Fall 2021)

Few things are as elusive as a person’s decision to vote. Civic engagement is arguably one of the most important aspects of American democracy, as those in office only derive power from the people they represent. In modern times, the polarized state of the country has driven experts to toil over voter turnout, its causes, and its complexities. Within this discipline, several obstacles lie in the way: dishonest people being counted in polls, voter suppression, accessibility, general human fickleness. Regardless, by studying this topic, researchers can gauge what compels a person to vote and subsequently encourage political participation.

One topic of concern is that of the voter profile — which individuals are more likely to vote, and why? There are a variety of factors that people recognize as crucial factors, such as age, gender, and education levels. Nevertheless, behavior also plays a significant role. For example, people all over America consume various amounts of news, and through social media, television, newspaper, and more, people learn about the political, social, and economic states of the country. The accessibility, framing, and agenda-setting aspects of the news can raise awareness and evoke emotion, and can have a potential impact on a person’s decision to vote. This background begs the question: to what extent does news consumption affect voter turnout?

YouTube, PragerU, and You: A Three-Way Street (Psychology and Politics, Spring 2021)

The Internet landscape is vast, diverse, and bold – a breeding ground of community as much as one of controversy. People share beliefs freely, and from them, build whole populations centered around particular causes. Prager University (colloquially known as PragerU) is a non-profit that prides itself in its creative analysis and promotion of American values. Infamous for its work, the organization looks through the lenses of conservatism and offers judgements that many regard as offensive, manipulative, or outright incorrect. Although one may respect and admire its forthright contributions to the marketplace of ideas, the effects of PragerU’s content are worrisome. When one considers how a volatile audience may react to biased representations of politics, philosophy, and economics, it is reasonable to believe that even simple ideas have claws and can use them to oppress. Through its use of the YouTube medium, its choice of source, and its storytelling methods, PragerU inspires its viewers to adopt ideologies with dangerous implications that pave the way to bigotry and ignorance. Nevertheless, despite the online power the organization holds, one can learn to become less susceptible to misinformation and come away wiser.

Dynamic Teachers, Dynamic Students [Teaching Statement] (Pedagogy of Peer Instruction, Spring 2021)

You can place a student in a classroom, you can ask them to learn. Although an instructor may have the best intentions while teaching, who is to guarantee that information will settle in the student’s memory in meaningful, long-term ways? After all, countless factors impede one’s ability to learn, such as pre-existing learning habits, a lack of adequate support, and an inability to motivate oneself. With so many forces potentially working against the student, it is not enough for an instructor to want to convey ideas effectively. Instead, they must take actionable measures to mitigate the effects of these forces and promote productive forms of learning. Specifically, when instructors establish comfortable learning environments, they foster successful classrooms that demand dynamism and flexibility from themselves and inspire it in students. Through this technique diversification, students can take control of their futures as stronger thinkers and learners.

Women in Technological Careers (Issues in Women’s Leadership, Fall 2019)

In this day and age, people are becoming more and more reliant on technology. In all parts of society, we see developments in artificial development, big data, personal devices, and more, and naturally, technical fields of study have seen dramatic increases in popularity. Millions of people wish to pursue careers in computer science, information technology, and related disciplines. Often times, they will pursue one of two post-graduation tracks: graduate school, in which they seek higher education, or industry, in which they seek to become an employee of a company. In this paper, I will dissect and assess the two tracks through women-oriented lenses, given that this minority group in particular has unique experiences navigating the culture that saturates the technology field.

Policy Memorandums

Antitrust Reform Memo (NYU Policy Case Competition, Spring 2022)

Currently, US antitrust regulation favors affordability and consumer accessibility above all else, thereby creating unfavorable conditions for Big Tech to exploit. Their true product is user data, which they collect for free and sell to other firms for profit. According to Credit Suisse, the number of companies listed on U.S stock exchanges has decreased from 7,322 to 3,671 in the last 20 years. Despite the massive increase in consolidations, the Department of Justice (DOJ) and the Federal Trade Commission (FTC), responded to the OECD’s concerns by stating that they find the claims of increased concentration unsupported by the data for meaningful markets. To reinvigorate the impact of antitrust laws and address the manipulation of user data, we propose the following three-step policy approach:

  1. The FTC should deprioritize the use of the consumer welfare standard in the identification of monopolistic power.
  2. Congress should give users control over the collection of their data and levy a tax on businesses that save and sell it.
  3. To lower barriers to entry, the FTC should provide public resources to start-up firms.
Unemployment Insurance Memo (Public Policy Formation, Fall 2021)

At the cusp of the Great Depression, President Franklin D. Roosevelt signed the Social Security Bill and established some of modern America’s most important social programs. Of these initiatives, unemployment insurance aimed to safeguard against economic decline and assist individuals suffering from the uncertainty and restrictions of unemployment. Through its federal-state system, the U.S. Department of Labor left the states to fund the program, define their own eligibility requirements and benefits, and administer compensation.

After decades of maintaining a largely unchanged system, the New Jersey Department of Labor was ill-prepared for the waves of unemployment that accompanied the COVID-19 pandemic, and long-standing flaws in the program’s design presented themselves. The outdated unemployment insurance system fails to adequately support the true population of unemployed individuals and incentivize employers to pay into it. The state should implement an expanded, partially privatized system with a more robust online infrastructure.