Exploring the spatiotemporal effects of domestic violence in Puerto Rico during the COVID-19 period: a geographically and temporally weighted regression approach

Domestic violence is an epidemic with profound impacts on both the mental and physical well-being of victims. In Puerto Rico, reported cases have risen rapidly in recent years, particularly after the 2020 COVID-19 pandemic. Despite its urgency, domestic violence remains understudied in the context of spatial information science, with limited research exploring how its relationship with sociodemographic factors varies across space and time. This study addresses this gap by investigating the spatiotemporal effects of domestic violence in Puerto Rico from 2020 to 2022, the COVID-19 pandemic period using geographically and temporally weighted regression (GTWR). Police-reported domestic violence data from the thirteen police regions of Puerto Rico showed a decline in 2020. Following the COVID-19 lockdown measures and mobility restrictions, it remains unknown whether this reduction is due to disruptions in reporting access and service availability or a decrease in underlying domestic violence. Results revealed statistically significant spatial clustering of police-reported domestic violence at the police-region level, with San Juan and Bayamón emerging as persistent hotspots. The female population showed strong positive associations (GTWR coefficients ranging from 0.5 to >1.0 across Puerto Rico), while separated populations exhibited moderate to strong positive relationships (0.25 to >1.0), with localized negative associations observed in the regions Aguadilla and Mayaguez. However, the population with moderate income stress demonstrated strong negative associations (coefficients < -1.0). These spatially varying relationships were not captured by the global ordinary least squares model(OLS). Overall, GTWR improved model fit (Δ Adj. R 2 = 0.0907). These results show that socioeconomic associations with domestic violence are spatially and temporally non-stationary, and can help public health officials prioritize region-specific prevention and intervention strategies to guide the effective allocation of public health resources in Puerto Rico.

Authors

Institutions

Publication Details

Journal
Frontiers in Public Health
Published
2026-04-28
DOI
https://doi.org/10.3389/fpubh.2026.1776190
Primary Topic
Intimate Partner and Family Violence
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Exploring the spatiotemporal effects of domestic violence in Puerto Rico during the COVID-19 period: a geographically and temporally weighted regression approach

Jarron M. Saint Onge, Carmen Caiseda, Joshua Atsu, Folashade B. Agusto et al.
Frontiers in Public Health
Intimate Partner and Family Violence
article

Exploring the spatiotemporal effects of domestic violence in Puerto Rico during the COVID-19 period: a geographically and temporally weighted regression approach

Jarron M. Saint Onge, Carmen Caiseda, Joshua Atsu, Folashade B. Agusto, Jeremis N. Morales Morales, Phyllis N. Muniu, Atinuke Adebanji
article en

Abstract

Domestic violence is an epidemic with profound impacts on both the mental and physical well-being of victims. In Puerto Rico, reported cases have risen rapidly in recent years, particularly after the 2020 COVID-19 pandemic. Despite its urgency, domestic violence remains understudied in the context of spatial information science, with limited research exploring how its relationship with sociodemographic factors varies across space and time. This study addresses this gap by investigating the spatiotemporal effects of domestic violence in Puerto Rico from 2020 to 2022, the COVID-19 pandemic period using geographically and temporally weighted regression (GTWR). Police-reported domestic violence data from the thirteen police regions of Puerto Rico showed a decline in 2020. Following the COVID-19 lockdown measures and mobility restrictions, it remains unknown whether this reduction is due to disruptions in reporting access and service availability or a decrease in underlying domestic violence. Results revealed statistically significant spatial clustering of police-reported domestic violence at the police-region level, with San Juan and Bayamón emerging as persistent hotspots. The female population showed strong positive associations (GTWR coefficients ranging from 0.5 to >1.0 across Puerto Rico), while separated populations exhibited moderate to strong positive relationships (0.25 to >1.0), with localized negative associations observed in the regions Aguadilla and Mayaguez. However, the population with moderate income stress demonstrated strong negative associations (coefficients < -1.0). These spatially varying relationships were not captured by the global ordinary least squares model(OLS). Overall, GTWR improved model fit (Δ Adj. R 2 = 0.0907). These results show that socioeconomic associations with domestic violence are spatially and temporally non-stationary, and can help public health officials prioritize region-specific prevention and intervention strategies to guide the effective allocation of public health resources in Puerto Rico.

Frontiers in Public HealthVol. 14
University of Kansas (US), InterAmerican University of Puerto Rico (PR), Purdue University West Lafayette (US), North-West University (ZA), Kwame Nkrumah University of Science and Technology (GH), University of Kansas Medical Center (US), Kwame Nkrumah University (ZM)
National Science Foundation
Gender equality
Openalex Percentile: Top 9%
Intimate Partner and Family Violence
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.