Access Type

Open Access Dissertation

Date of Award

January 2019

Degree Type

Dissertation

Degree Name

Ph.D.

Department

Economics

First Advisor

Stephen . Spurr

Abstract

Since the infamous riot of 1967, high crime rates and negative media reports have labeled the city of Detroit as one of the most dangerous cities in the U.S. For a better understanding of crime situation in Detroit, we discuss the spatial interactions of crime rates among block groups as well as the impact of socioeconomic variables on crime by using spatial autoregressive models, crime data, and socioeconomic data. In particular, we introduced a geographical methodology: inverse distance weighting (IDW), which is mainly used to estimate distance based weighting from a scattered set of points. This paper is an essential and policy-relevant topic that has seldom been studied in the past. The study not only enhances our understanding of spatial interactions in regional criminal activities but also provides implications for policymaking, especially for re-allocation of police resource in a certain patrol area based on the fact of spatial interactions of different types of crime. Strong spatial interactions are found in this paper except for weapons offenses. Considerable significant parameters of the socioeconomic variables on crime are discussed as well.

The evidence of the existence of peer effects is scarce for endogeneity issues like selection bias. This study analyzes peer effects in student academic performance of a junior high school in Xinjiang Province, China where students were randomly selected to each classroom and dormitory. Based on unique data from a remote area of China, we find that both positive and negative peer effects exist and are also significant in both classroom and dormitory analyses, which provide new insight and implications for educational policy making. According to two comparable SAR models, this paper provides relatively rational and considerable empirical results and shows more accurate interdependence results. In particular, consistent and unique dormitory peer effect coefficients are presented in our model. It also empirically confirms and supports the theoretical work of previous studies in SAR modeling. Limited but considerable exogenous variables are discussed in this paper as well.

Download

Share

COinS