Theses and Dissertations

Date of Award

2026

Document Type

Dissertation - NSU Access Only

Degree Name

Doctor of Philosophy (PhD)

Department

Abraham S. Fischler College of Education and School of Criminal Justice

Advisor

Steven Hecht

Committee Member

Grace Telesco

Committee Member

James Pann

Keywords

accountability mechanisms, age, binomial generalized linear models, Black/African American respondents, demographic disparities, demographic-disaggregated reporting, force outcomes, gender, generalized linear models, household income, intersectionality, likelihood-ratio chi-square tests, national data source, police-initiated encounters, police-public contact survey, police use of force, PPCS, predictive modeling, race, social categories, SPSS, structural accountability, threat of force, traffic stops, use-of-force reporting

Abstract

This study examined how the intersection of demographic characteristics predicts differences in police-initiated encounters resulting in force used, threat of force, or neither. Using data from the 2022 Police-Public Contact Survey (PPCS), the study employed an intersectional framework to analyze how age, race, and gender independently and collectively influence police use of force outcomes during policeinitiated contacts, specifically traffic stops, pedestrian stops, passenger stops, and trafficaccident responses. While existing research has examined these factors separately, limited scholarship has investigated how they intersect to shape unique experiences during police-initiated encounters using the PPCS data. This study addressed that gap through eight research questions building from single-factor analyses to three-way intersectional effects.

The analytic sample (n = 6,894) was drawn from PPCS respondents aged 18 or older whose most recent contact was police-initiated. Binomial generalized linear models (GLMs) were conducted in SPSS to predict three mutually exclusive, dichotomous dependent variables: force used, threat of force, and neither. Independent variables were gender (male, female), age group (18–24, 25–44, 45–64, and 65 or older), and race (Black/African American vs. Other), with household income analyzed as a covariate. Significance was evaluated using likelihood-ratio chi-square tests, with pairwise contrasts conducted to compare levels of the categorical predictors and line plots used to interpret significant effects. Physical force was a rare event in the sample (n = 91; 1.3%), and threat of force was rarer still (n = 7; 0.1%).

Results indicated that gender, age, and race each independently and significantly predicted force outcomes. Male respondents and Black or African American respondents were significantly more likely to experience physical force, and respondents aged 65 and older were significantly less likely to experience physical force than all three younger age groups. These main effects persisted after controlling for household income, which was itself a significant predictor. None of the two-way interaction terms (gender × age, gender × race, age × race) reached statistical significance. The three-way interaction models failed to converge due to complete separation caused by insufficient force events within demographic subgroups, leaving the three-way hypothesis indeterminate rather than rejected. Findings indicate that demographic disparities in police use of force are distributed across multiple independent social categories and support policy measures including demographic-disaggregated use-of-force reporting, structural accountability mechanisms, and continued investment in the PPCS as a national data source.

Download
To access this thesis/dissertation you must have a valid nova.edu OR mynsu.nova.edu email address and create an account for NSUWorks.

Free My Thesis

If you are the author of this work and would like to grant permission to make it openly accessible to all, please click the Free My Thesis button.

  Contact Author

Share

COinS