Improving Measurement of Community Safety Perceptions with Enhanced Data Inclusivity and Novel Use of Small Area Estimation through Population – Based Hybrid Sampling: A Pilot Study in Detroit, Michigan
Project Summary
The proposed study aims to implement a pilot project in the City of Detroit (n=1,200) with a novel, yet practical sampling approach designed to maximize coverage of microgeographies (Census block groups in our case) as well as inclusivity of groups who may be underrepresented in traditional sample surveys due to, for example, nonresponse and may hold different perceptions of community safety and policing (e.g., younger population) by blending address-based sampling (ABS) and respondent driven sampling (RDS). The sampling approach will rely on the principles of probability sampling, scientifically guaranteeing the representativeness. The various components in the hybrid sample will be blended through our extension of multiple frame estimation methods to fit our unique sample design as well as our major modification to the existing RDS estimators.
Investigators
Sunghee Lee
Project Period
12/31/2027