AI-assisted grading and personalized feedback in large political science classes: Results from randomized controlled trials
Tobias Heinrich, Sanghoon Park, Navida Wang, Spencer Baily, Jack DeOliveira
Available here: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0328041
Grading and providing personalized feedback on short-answer questions is time consuming. Professional incentives often push instructors to rely on multiple-choice assessments instead, reducing opportunities for students to develop critical thinking skills. Using large-language-model (LLM) assistance, we augment the productivity of instructors grading short-answer questions in large classes. Through a randomized controlled trial across four undergraduate courses and almost 300 students in 2023/2024, we assess the effectiveness of AI-assisted grading and feedback in comparison to human grading. Our results demonstrate that AI-assisted grading can mimic what an instructor would do in a small class.
Ticket to Bribe: Municipal Targeting by Cartel and Corruption as Facilitation (in progress)
Spencer Baily
Organized criminal groups in Mexico strategically target municipal officials when their revenues depend on local, place-bound markets like fuel theft. My study shows that municipalities exposed to oil pipeline taps report higher levels of corruption, as local officials become collusive partners who tolerate theft and protect illicit operations. This corruption reduces extortion by substituting bribes for coercion but does not curb lethal violence driven by cartel rivalries. Overall, the findings demonstrate that corruption embeds criminal rents into local governance, stabilizing economic rackets while failing to contain broader violence.
Here Comes the Funds: Rebel Financial Ecosystems and Diversification (in progress)
Spencer Baily
Some rebel group maintain diverse funding portfolios while others are quite limited. What explains this variation? I argue that it is a partly determined by two interacting factors, their relationships with civilians and variability of existing funding. Groups with strong ties to locals are better able to diversify, still it is when current funding becomes threatened that they are most likely take advantage of these close ties.
Sustaining Rebellion: Resources and Rebel Conservation (in progress)
Kenneth Aba, Spencer Baily, Chelsea Estancona
This research project examines the conditions under which rebel groups attempt to sustain consistent levels of funding from limited, yet renewable resources. We investigate when and why rebels sustainably extract these important assets. In our paper, we argue that groups that exhibit state-like behavior will be more inclined to also sustainably extract resources due to their interest in gaining support from civilians and in maintaining other operations. It is also important for groups that rely on the local population for labor to restrict their practices.
Research Assistant
1. Summer 2023: I reviewed and coded data for Lindsay Reid and Chelsea Estancona for a paper on pro-government militias and peace deals.
2. Summer 2022: I conducted a literature review for Matthew Wilson and Amanda Edgell on election management bodies (EMBs) in all forms of government.
Skills
– R
– Latex
– Python
– Microsoft Office Suite
– Econometrics
Spencer Baily 