Spatial Analysis Series Archive

The creation of electoral boundary maps has been traditionally the province of experts. Most often in the US, redistricting maps are created in “smoke filled rooms.” Even in places where public commissions seek wider input, the crafting of electoral lines has been limited to a select group of commission members. In the last few years, however, advances in information and communication technologies have opened new opportunities for participation in political mapping. These new technologies and algorithms have also made possible extensive public dissemination of data, feasible analysis of hundreds of districting criteria, and the creation of large numbers of automated and crowdsourced redistricting maps. The result has been that for the first time, thousands of members of the general public are participating directly in redistricting, and have created hundreds of legal electoral maps. In this talk I discuss the trends in participative electoral mapping, and describe in detail some results from the Public Mapping Project (http://www.publicmapping.org), which successfully increased participation in redistricting in several states. I also discuss types of inferences to which electoral maps are relevant, common misinterpretations, and how to draw correct inferences from these maps.

Today’s antipoverty safety net is dramatically different from the one in place two decades ago when welfare reform was enacted. Rather than a safety net primarily dependent on cash assistance programs, as is the common perception, the current system is highly reliant on social service programs funded by government and delivered through community-based nonprofits. Annual public and private expenditures for social service programs today exceed total federal outlays for cash assistance programs like welfare, food stamps, and the Earned Income Tax Credit (EITC). Yet, little attention has been paid to the implications of this transformation in antipoverty assistance and the challenges a service-based safety net faces in the wake of the Great Recession. Drawing on unique survey data from over 1,500 social service providers in Chicago, Los Angeles, and Washington, D.C., the author will discuss the new geography of the American safety net and the presence of spatial mismatches that undermine the system’s ability to respond to rising poverty. The author also will discuss implications of shifts in safety net assistance for the study of public policy and the welfare state moving forward.

The researchers discuss the merits of developing a new set of tools for matching electoral and administrative data. A bottleneck for researchers is caused by the fact that the shape and extent of electoral constituencies typically do not correspond precisely with administrative units. Data on unemployment, poverty, and ethnicity, for instance, are available at the administrative district level, while election results are at the constituency level. How do we marry these data to study questions such as the relationship between elections and poverty? They describe plans for developing a large database of geocoded electoral constituencies from most of the world’s democracies and including several election cycles, a set of data files with areal and population weights that will enable researchers to match data from electoral constituencies to data from other standard subnational units of analysis, a set of guidelines for researchers to use when matching geocoded data across different units of subnational analysis, and programming syntax that make it easy for people to use the data files with the areal and population weights.

This paper draws upon the data compilation efforts of several Human Rights NGO’s and Rwandan government data sources in an effort to understand how many people were killed in Rwanda, when, where, and by whom. To conduct this analysis, one needs information on victims, times, locations, and perpetrators. Based on an examination of a new database on Rwandan political violence (Davenport and Stam), our GIS animations shows that three distinct zones of political violence existed during the 1994 civil war: 1) those killings that occurred under government jurisdiction, 2) those where government and rebels are engaged in fighting each other (i.e., the battle-fronts or frontlines) and 3) those under rebel jurisdiction. The majority of killings take place in the zone under FAR/Rwandan government control (accounting for approximately 200,000-900,000 deaths). They are the ones proximately responsible for almost all of the political violence though culpability varies tremendously for any individual episode of killing. Given the wide range of types of killing and violence, absent eye-witness accounts regarding any individual’s participation in specific acts of violence it is impossible to claim with any certainty what specific crime an individual may or may not have engaged in.

Existing studies of welfare diffusion suffer from three shortcomings: (1) they do not model state welfare competition formally; (2) they use “nearest-neighbor” lags to detect the influence of external developments on internal policymaking decisions; and (3) they assume that AFDC and TANF are independent programs whose implementation patterns cannot be compared with one another. Here, I address these shortcomings by devising a formal model to describe state welfare competition and determining whether dimensions other than geographic adjacency can influence state policymaker decisions. I compare diffusion patterns in AFDC and TANF programs to determine whether state policymakers are influenced by the same sets of external influences in both cases.

The number of studies examining racial and socioeconomic disparities in the geographic distribution of environmental hazards and locally unwanted land uses has grown considerably over the past two decades. Most studies have found statistically significant racial and socioeconomic disparities associated with hazardous sites. However, there is considerable variation in the magnitude of racial and socioeconomic disparities found; indeed, some studies have found none. Uncertainties also exist about the underlying causes of the disparities. Many of these uncertainties can be attributed to the failure of the most widely used method for assessing environmental disparities to adequately account for proximity between the hazard under investigation and nearby residential populations. In this presentation, Dr. Mohai will identify the reasons for and consequences of this failure and demonstrate ways of overcoming these shortcomings by using alternate methods facilitated by the application of geographic information systems (GIS). Through the application of such methods, he will discuss how assessments about the magnitude and causes of racial and socioeconomic disparities in the distribution of hazardous sites are changed. In addition to research on environmental inequality, he will discuss how GIS methods can be usefully applied to other areas of research that explore the effects of neighborhood environmental context on a range of outcomes, including health.

Voters care about public policies that affect them, yet the scope of their interests may not be well defined by legislative district boundaries. The constraints of the electoral system are particularly evident within metropolitan areas, where citizens may live and work in different districts. Although citizens may not be able to vote in all legislative races relevant to them, they may seek to express their political preferences by other means. In this paper we examine one such form of participation: campaign contributions. We hypothesize that citizens’ contribution decisions will be a function of the connectedness of their metropolitan region: citizens living in more interconnected regions will be more likely to donate to candidates in neighboring districts than those who live in less connected regions. To test this hypothesis, we geolocate all campaign contributions to U.S. Congressional candidates made by individual donors in the 2007-08 election cycle. Using GIS technology, we overlay maps of congressional district boundaries and metropolitan regions (MSAs) to identify the congressional district and MSA of each donor and each recipient. We use several measures to characterize the interconnectedness of the MSA. We find positive and significant support for our hypotheses. Our results call to question the conventional theory of representation with its underlying premise of a dyadic relationship between the constituent and her home-district representative.

Spatial interdependence--the dependence of outcomes in some units on those in others--is substantively and theoretically ubiquitous and central across the social sciences. Spatial association is also omnipresent empirically. However, spatial association may arise from three importantly distinct processes: common exposure of actors to exogenous external and internal stimuli, interdependence of outcomes/behaviors across actors (contagion), and/or the putative outcomes may affect the dimensions along which the clustering occurs (selection). Accurate inference about any of these processes generally requires an empirical strategy that addresses all three well. From a spatial-econometric perspective, this suggests spatiotemporal empirical models with exogenous covariates (common exposure) and spatial lags (contagion), with the spatial weights being endogenous (selection). From a longitudinal network-analytic perspective, the same three processes are identified as potential sources of network effects and network formation. From that perspective, actors' self-selection into networks (by, e.g., behavioral homophily) and actors' behavior that is contagious through those network connections likewise demands theoretical and empirical models in which networks and behavior coevolve over time. This paper begins building such models by, on the theoretical side, extending a Markov type-interaction model to allow endogenous tie-formation, and, on the empirical side, merging a simple spatial-lag logit model of contagious behavior with a simple p-star logit model of network formation, building this synthetic discrete-time empirical model from the theoretical base of the modified Markov type-interaction model. One interesting consequence of network-behavior coevolution--identically: endogenous patterns of spatial interdependence--emphasized here is how it can produce history-dependent political dynamics, including equilibrium phat and path dependence. The paper explores these implications, and then concludes with a demonstration of the strategy applied to alliance formation and conflict behavior among the great powers in the first half of the twentieth century.

This study examines the external neighborhood effects of Low-Income Housing Tax Credit (LIHTC) Projects built in Santa Clara County from 1987 to 2000. Three types of developers have built LIHTC projects in this area: nonprofit, for-profits, and a county public housing authority. Using a difference-in-difference hedonic regression approach, this study finds that almost all the LIHTC projects examined have generated significantly positive impacts on nearby property value. In particular, the study also finds that most nonprofit projects have delivered benefits similar to those of for-profit projects. Yet projects built by some of the largest nonprofits and the county housing authority have generated the greatest neighborhood impacts. Low-income neighborhoods have also benefited more from LIHTC developments than other types of neighborhoods.

Spatial non-stationarity may well be a rule rather than an exception in the study of many political phenomena. When relationships vary, empirical analyses may benefit from using a geographically weighted regression (GWR) for modeling the continuous spatial heterogeneity in data. The talk will provide background on this method of analysis and examine data on political participation to illustrate the unique insights that one can garner with attention to this spatial peculiarity. We will examine spatial patterns of campaign volunteering in a major statewide election campaign. While volunteerism is associated with characteristics of the political and socioeconomic characteristics of the precincts in which the participants reside, these statistical relationships are spatially non-stationary. By using GWR to specify local regression parameters, we are able to capture the heterogeneity of contextual processes that generate the geographically uneven flow of volunteers.

Do democracies provide more public goods than autocracies? Clear answers to this question have been hampered by inconsistent, unreliable, or missing data. To address the shortcomings of self-reported government data, I propose a novel method to generate unbiased estimates of the provision of electrical infrastructure across the entire globe using satellite imagery of nighttime lights. After demonstrating the validity of my measure, I show that democratization is associated with a substantial decrease in unelectrified populations, even after controlling for differences in per capita income, population density, and other factors. Complementing the cross-national results, I use a difference-in-differences estimator applied to the former Soviet Bloc to show that democratization has positive effects on electrification over time. The results affirm the power of electoral incentives in inducing democratic leaders to provide higher levels of public goods than in autocracies where leaders do not need to win elections.

Standard measures of poverty concentration based on census-tract percentages do not accurately reflect neighborhood conditions because they offer a weak link to the underlying geography of a neighborhood. Changes in the spatial configuration of land use within a census tract can have the effect of increasing or decreasing the density of poverty. This study uses a dasymetric mapping technique in a raster GIS environment to intersect population data in a block group layer with land-use categories from a land-use layer. The method produces poverty counts and rates at a much finer spatial resolution than a block group, while also offering an explicit spatial relationship between population and surrounding neighborhood characteristics. The paper illustrates the technique for the City of Detroit by measuring poverty change between 1990 and 2000. It finds that (a) a substantially larger share of poor people in 2000 lived in places that exceeded the standard 40-percent threshold than is calculated by an a spatial census-tract approach; (b) poverty became more concentrated in space during the 1990s, counter to widely publicized reports of diminishing poverty concentration that were based on the percentage of poor people in high-poverty tracts; and (c) poverty improved where neighboring land-use conditions also improved.

When one of the major parties in the United States wins a substantially larger share of the seats than its vote share would seem to warrant, the conventional explanation lies in manipulation of maps by the party that controls the redistricting process. This paper uses precinct-level data from Florida to demonstrate a common mechanism through which substantial partisan bias can emerge purely from residential patterns. When partisan preferences are spatially dependent and partisanship is highly correlated with population density, any districting scheme that generates relatively compact, contiguous districts will tend to produce bias against the urban party. In order to demonstrate this result empirically, we apply automated districting algorithms driven solely by compactness and contiguity parameters, building winner-take-all districts out of the precinct-level results of the tied Florida presidential election of 2000. The simulation results demonstrate that with 50 percent of the votes statewide, the Republicans can expect to win around 59 percent of the seats without any “intentional” gerrymandering. This result emerges because urban districts tend to be homogeneous and overwhelmingly Democratic while suburban and rural districts tend to be moderately Republican. Thus, in Florida and other states where Democrats are highly concentrated in cities, the seemingly apolitical practice of requiring compact, contiguous districts will produce systematic pro-Republican electoral bias.