Collaborative Research: Crime and Community in a Changing Society, the National Neighborhood Crime Study 2

Maria Velez
NSF
06/01/2014 – 05/31/2016

Description:

A major part of this project involves the collection of a second panel of neighborhood crime data from police departments for 91 cities throughout the United States; the first wave of data collection was completed approximately ten years ago. Based upon the lead PI's (Krivo at Rutgers University) experience in collecting the first wave of data, we had originally anticipated that the collection, cleaning, and merging of all data for the project would take a full 18 months. However, several differences between the first wave and the current wave of the study lead us to now conclude that we can complete this portion of the work in the first 12-14 months of the project. First, advances in data management and storage practices within police agencies over the last ten years should speed up the process of data acquisition and transfer relative to wave 1.

Second, the division of labor for the project across two sites with a total of three co-PIs will allow data collection to proceed more quickly than in wave 1 where the full effort took place at a single site with one PI managing all police department contacts. Third, Krivo's experience with the first wave of data collection will ensure a smoother and more efficient data collection and management plan. In addition, with NSF support approved, we can leverage these funds to gain some additional resources from the University of New Mexico. One of the Co-PIs, Lyons, is committing research funds awarded to him from the College of Arts and Sciences to cover an additional .25 FTE GRA for Fall 2014 and Spring 2015. This extra assistance will provide more staff during the intense data collection, cleaning, and management period of the project.

The remaining 10-12 months of the project will be devoted to data analysis and writing of papers to evaluate the theoretical model proposed in the research. However, because we will have at least 6 months less time than originally proposed to analyze the data and write papers for publication, we will scale back the breadth of analyses conducted during this phase of the project. Specifically, we will conduct only limited analyses addressing our fifth objective of examining how neighborhood characteristics are both causes consequences of crime. We will limit our modeling of the consequences of crime to a few key neighborhood outcomes of population and economic change. Analyses of how city and metropolitan conditions affect the relationships between neighborhood characteristics and crime (the fourth objective) will also be scaled back to focus on a more limited set of city/metropolitan economic and political conditions.

Research Plan:

We will address five specific objectives.

  1. The first objective is to examine how changes in neighborhood socioeconomic and housing conditions relate to changes in neighborhood crime (increases and decreases). Socioeconomic conditions in neighborhoods, such as poverty, joblessness, and low status work, are central sources of crime levels within communities (Peterson and Krivo 2010b; Sampson 2012). However, the Great Recession increased financial and housing precariousness for many and heightened public awareness of extreme inequalities between affluent and middle class or poor individuals (Holzer and Hlavac 2012; Rosenbaum 2012). Widespread economic strain, rising and falling housing prices, increases in foreclosures, and declines in homeownership that can destabilize the forces that keep crime at bay are experienced unevenly across cities and local neighborhoods within urban areas. That neighborhoods experience unequal circumstances is in part a product of elite decision making guided by interests in capital accumulation and growth such as through the promotion of high cost loans (e.g., Logan and Molotch 1987; Squires and Kubrin 2006). African American and Latino communities have been particularly hard hit by predatory lending, subsequent forecloses, and decreases in homeownership and prime (non-risky) external investments (Woodstock Institute 2008, 2009). Yet in the absence of over time neighborhood crime data, we do not know how differential changes in socioeconomic, housing, and political economic conditions influence changes in serious criminal activity.
  2. Our second objective is to test whether changes in the demography of neighborhoods relate to changes in neighborhood crime. Immigration to the United States continues to increase ethnoracial diversity almost everywhere but patterns of change are highly differentiated across neighborhoods (Farrell and Lee 2011; Friedman 2008; Logan and Zhang 2010). In some places, concentrations of immigrants remain dominant while in others immigrants are increasing the mix of racial and ethnic groups present. At the same time, the relative size of the African American population is relatively stable and segregation of this group from others is still high in many places (albeit steadily declining) (Parisi, Lichter, and Taquino 2011; Logan and Stults 2011; Wright et al. 2013). These demographic changes have yieldedmany more racially and ethnically diverse neighborhoods alongside homogeneous African American, White, and Latino areas (Wright et al. 2013). Some communities have also suffered population losses while others have grown substantially. The consequences of demographic shifts may range from competition and conflict among groups with distinct interests to neighborhood decline or revitalization (Bean and Stevens 2003; Waldinger 1997), all of which could alter crime differentially. Political changes and elite decisions about where to place urban development projects or zoning, for instance, also shape the spatial location of groups and encourage highly divergent investments across areas (Harvey 2007; Squires and Kubrin 2006). Population dynamics may thus reshape or reinforce the divergent structural fates of segregated ethnoracial groups. As the United States moves toward a nation that will be majority non-White (Passel and Cohn 2008), it is imperative that we garner more comprehensive knowledge about the consequences of such changes for the safety and criminal risks experienced in local areas.
  3. The third objective is to test whether changes in nearby neighborhood socioeconomic, housing, and demographic conditions correspond to changes in neighborhood crime and ethnoracial disparities in crime. Neighborhoods are not islands that are divorced from the areas and conditions that surround them (Mears and Bhati 2006; Morenoff, Sampson, and Raudenbush 2001; Vélez and Richardson 2012). Demographic, economic, and political circumstances of neighborhoods influence crime in nearby areas because crime can spill over fluid neighborhood boundaries and the circumstances of areas in close proximity to one another confer resources and/or disadvantages that spread beyond their borders (e.g., Peterson and Krivo 2009a, 2010b). Changing spatial arrangements regarding ethnoracial and economic segregation, rising gentrification, and the demolition of many public housing projects are altering who and what local conditions are close to one another (Fagan 2008). Yet, we do not know how these evolving spatial patterns are altering crime overall and across racially and ethnically distinct areas.
  4. The fourth objective is twofold: to explore how changing city and metropolitan conditions influence changes in neighborhood crime rates; and to examine how changing city and metropolitan conditions affect the relationships between neighborhood characteristics and crime. Cities and metropolitan areas differ significantly in the types of social, economic, and political changes that they experience, including changes in policies relevant to the support of local populations and neighborhoods (Bartels 2010; Rich 1980). Some cities continue to experience the effects of deindustrialization while others have expanded or altered their industrial base. Moreover, receptive political structures at the city level such as the presence of sympathetic elites, minority representation, and inclusive policies can significantly influence the socioeconomic realities of minority and poor neighborhoods. For example, minority representation in city government may lead to policies and resources that advance the social, economic, and political standing of minority communities. Analyses of the NNCS1 have shown that some of these macro- conditions independently affect crime in neighborhoods (Crutchfield, Krivo, and Peterson 2012; Krivo, Peterson, and Kuhl 2009; Peterson and Krivo 2010b; Ramey 2012; Vélez and Lyons 2012). For example, neighborhoods in more segregated cities have more violent and property crime and those in places with a larger manufacturing base have less violent crime. Yet, the city and metropolitan context may also enhance or diminish the connections of changing neighborhood conditions with crime. Indeed, a thriving economy and supportive policy environment could diminish the harmful effects of crime-inducing neighborhood characteristics while weak economic and policy environments may enhance such relationships (Crutchfield et al. 2012; Vélez, Lyons, and Santoro 2012). While theoretically important, these conditional effects have rarely been explored and only in static models.
  5. The fifth objective is to examine the extent to which changing neighborhood characteristics are both causes and consequences of changes in local crime. Neighborhood crime research has mainly examined theoretical relationships whereby social structural conditions are considered as sources of crime. Yet the cross-sectional design of most studies cannot support causal claims about the effects of social structure on crime as opposed to the effects of crime on social structure. Indeed, the small literature on dynamic crime patterns for neighborhoods and cities suggests that crime does have consequences for demographic and social change within communities (see Hipp 2010; Kirk and Laub 2010). Higher crime in cities leads to more outmigration, particularly among Whites (Cullen and Levitt 1999; Frey 1979; Liska, Logan, and Bellair 1998). Higher neighborhood violent crime also decreases population and housing values (Hipp, Tita, and Greenbaum 2009; Morenoff and Sampson 1997; Tita, Petras, and Greenbaum 2006). While these studies suggest that relationships between changes in neighborhood conditions and crime may be reciprocal, they have largely analyzed cities as units or studied areas within a single city. The exception is Hipp (2010) who explored a convenience sample of neighborhoods in thirteen cities. Compiling and analyzing the NNCS2 will provide the first nationally representative test of complex multi-directional dynamics in the linkages between neighborhood characteristics and crime.