- Overall reported forcible rape rates from 1991–2023 exhibit high persistence, suggesting shocks are long lasting when the full sample is considered.
- A single structural break in 2010 divides the series: pre-2010 shows positive trend and near unit root behaviour; post-2010 shows negative trend.
- The post-2010 reduction in persistence likely reflects methodological changes and a revised forcible rape definition affecting counting practices.
Violence Vict. 2026 Jul 15:VV-2025-0122.R1. doi: 10.1891/VV-2025-0122. Online ahead of print.
ABSTRACT
In this article, we examine the historical evolution of the reported forcible rape rate in the United States using annual data from 1991 to 2023. Using fractional integration methods, we determine if shocks are expected to be transitory or permanent. The results indicate high levels of persistence when using the entire sample period; however, allowing for breaks, a single break is detected in 2010, with a significantly different behavior before and after the break. Thus, before that year, the trend is significantly positive, and the order of integration is very close to 1; after 2010, however, the trend becomes negative, and the order of integration is much lower than 1, showing a certain degree of reversion to the mean. This may be explained by methodological changes in the way sexual assaults are counted and by the revised definition of forcible rape that was introduced.
PMID:42457383 | DOI:10.1891/VV-2025-0122
Share Evidence Blueprint
Save to Google Notes
Search Google Scholar
Save as PDF
⭐ My Revision List
