I Know Numbers, I Have the Best Numbers
Description
Presidential press conferences are a central site of political communication, forcing leaders to explain policy, performance, and crises in real time. This paper examines how often U.S. presidents use numbers in these settings, analyzing 2,452 press conference transcripts from 1923–2025 (8.35 million words). I introduce Numerical Claim Density (NCD), defined as the proportion of substantive numerical references in a transcript. Contrary to expectations of steady modernization, numerical rhetoric does not increase monotonically over time. Instead, it peaks during the early 1930s under Herbert Hoover, declines through much of the late twentieth century, and rises again in the 2020s under Donald Trump. Across the full sample, presidential identity explains roughly 25% of the variance in numerical density, while economic recessions contribute little once composition effects are accounted for. Overall, the findings suggest that numerical rhetoric in presidential communication is episodic and personality-driven rather than a simple function of historical period or crisis conditions.
Replication materials (processed data, code, and outputs) are available on OSF: https://osf.io/wr7dz. Source transcripts are from the American Presidency Project: https://www.presidency.ucsb.edu/.
Files
NCD_Presidential_News_Conferences_Final.pdf
Files
(172.1 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:5af353caf26fb0e2f6c4946d6d6eb37a
|
172.1 kB | Preview Download |
Additional details
Additional titles
- Alternative title (English)
- Numerical Rhetoric in US Presidential Press Conferences: A Computational Analysis, 1923–2025
Identifiers
Related works
- Is derived from
- Dataset: https://www.presidency.ucsb.edu/ (URL)
- Is supplement to
- Dataset: https://osf.io/wr7dz (URL)
Dates
- Created
-
2026-01-26
References
- American Presidency Project. Presidential Press Conferences. University of California, Santa Barbara. https://www.presidency.ucsb.edu/
- National Bureau of Economic Research. Business Cycle Dating. https://www.nber.org/research/business-cycle-dating