Graduation Year

2025

Date of Submission

4-2025

Document Type

Open Access Senior Thesis

Degree Name

Bachelor of Arts

Department

Psychology

Reader 1

Jay Conger

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

Abstract

The United States faces persistent challenges in controlling political fraud. Government agencies exhibit the second-highest fraud rate of any industry, and corruption levels have only worsened over the last decade, contributing to declining public trust in national leadership. Current fraud detection efforts generally catch late-stage, large-scale fraud, highlighting the need for earlier-detection mechanisms. One way this could be accomplished involves identifying rhetorical patterns that are commonly used by political frauds in order to flag high-risk individuals. Leaning on the Theory of Planned Behavior, it was hypothesized that political frauds would exhibit more rhetorical appeals to fear, blame, pathos, and self-promotion than non-frauds. To test this, a matched pairs design was employed, pairing six Congressional frauds with non-fraudulent counterparts. Five media appearances per politician were identified, transcribed, and coded (both deductively and inductively) using computer-aided text-analysis. ChatGPT Plus has been shown to be a fairly reliable coder of qualitative datasets (k = .66-.95) and thus was chosen. 18 rhetorical strategies were coded for raw instances per speaker, transformed into per-1,000-word frequencies, and analyzed with paired-samples t-tests. Results showed that fraudulent Congressmembers exhibited more combative, hyperbolic rhetoric, more national-level discourse, less measured policy discussion, and less local-level discourse than non-fraudulent Congressmembers. This supported our hypothesis and suggests that fraudulent political behavior may be commonly preceded by measurable linguistic patterns. Fraud-fighting agencies may benefit from using these insights to flag potentially fraudulent politicians earlier in their criminality.

Download

Share

COinS