Plant-Capture Methods for Estimating Homeless Population Size From Uncertain Plant Captures.

Area of Science:

• Epidemiology
• Statistical Modeling
• Population Dynamics

Background:

• Plant-capture methods are used to estimate population sizes, notably for homeless populations via street surveys.
• Existing methods often overlook uncertainty in plant capture status and site heterogeneity.
• Accurate population estimation is crucial for resource allocation and policy development.

Purpose of the Study:

• To develop and evaluate novel hierarchical modeling approaches for plant-capture studies.
• To formally incorporate uncertainty in plant capture status and survey site heterogeneity.
• To improve the accuracy of population size estimates, particularly for vulnerable populations.

Main Methods:

• Proposed three hierarchical models of increasing complexity to address uncertainty.
• Incorporated plant capture status uncertainty and survey site heterogeneity.
• Applied methods to US Census Bureau's "S-Night" study data for homeless population estimation.

Main Results:

• Developed robust frequentist and Bayesian implementations of the proposed models.
• Demonstrated improved accuracy in population size estimation by accounting for uncertainty.
• Empirical evaluations confirmed the statistical performance of the new modeling approaches.

Conclusions:

• The proposed hierarchical models offer a more accurate and reliable approach to plant-capture studies.
• Accounting for plant status uncertainty is critical for precise population size estimation.
• These methods have significant implications for epidemiological studies and social service planning.