UROP Project

A Taste of Power: An Experimental Investigation

Power; Competitive Environment; Preference Change; Research Design; Experiments

Research Mentor: Dr. Ou Kai Ou,
Department, College, Affiliation: Department of Political Science, Social Sciences and Public Policy
Contact Email: kou@fsu.edu
Research Assistant Supervisor (if different from mentor):
Research Assistant Supervisor Email:
Faculty Collaborators:
Faculty Collaborators Email:

Looking for Research Assistants: Yes
Number of Research Assistants: 2
Relevant Majors: Open to all majors; political science and economics students have priority.
Project Location: On FSU Main Campus
Research Assistant Transportation Required:
Remote or In-person: In-person
Approximate Weekly Hours: 5-10 hours, During business hours
Roundtable Times and Zoom Link:
Not participating in the roundtable

Project Description

A striking regularity in daily life is that people seem to change once they gain power. For example, the humble colleague becomes imperious after a promotion, the fair-minded official grows self-serving after election. This observation underlies a common narrative: power corrupts. Yet this narrative, however intuitive, may be masking a more complicated and more important truth. When we see behavior shift alongside power, we are observing a correlation, not yet a mechanism. Several very different mechanisms could produce the same surface pattern. Power may genuinely change what people want, reshaping their goals, values, or risk tolerance. Alternatively, power may not create new preferences at all, but simply remove the social and institutional constraints that had been suppressing preexisting ones, in which case power reveals character rather than corrupting it. It is also possible that the people who seek and attain power differ systematically from those who do not, so what looks like personal transformation is really a selection effect. Yet another possibility is that behavior shifts because accountability and oversight change with position, not because anything internal to the person does. Finally, apparent corruption may reflect nothing about individual psychology at all, but rather a failure of institutional design, such as weak checks and balances that would produce similar outcomes regardless of who holds power. Adjudicating between these explanations is not merely an academic exercise: each implies a different remedy. If power itself corrupts preferences, the solution lies in selecting and developing better leaders; if power merely disinhibits or escapes oversight, the solution lies in strengthening accountability structures; and if the true culprit is institutional design, then focusing on individuals may be a costly distraction from fixing the systems that surround them. This project aims to disentangle these possibilities through experimental investigations.

Research Tasks: The UROP research assistants will engage in various research-related tasks, including literature reviews, data collection, data entry, conducting laboratory experiments, reaching out to individuals for interviews, and other pertinent assignments. These tasks will be structured as weekly assignments. Mentees will receive email notifications detailing their assignments at the start of each week, typically on Sunday or Monday, with the expectation that the work will be completed by Friday afternoon. Some tasks can be executed asynchronously and remotely, and some tasks should be performed in person on campus, depending on the assistant's role in the team. Regular research meetings between the mentor and mentee will be held every week. Mentees are anticipated to possess the necessary skills for their assigned tasks, although they may be encouraged to acquire new skills as needed to effectively complete their assignments. Most work will be conducted and submitted electronically. Therefore, it is expected that mentees have proficiency in using software such as MS Office, Adobe, and any other relevant applications.

Skills that research assistant(s) may need: Required skills vary depending on each mentee’s assigned tasks. The only universal requirement is the ability to conduct literature reviews without using large language model tools (e.g., ChatGPT), which, while useful, can produce false information or cite non-existent sources. Priority will be given to individuals with experience in conducting interviews, running lab experiments with human subjects, or developing websites.