Planning Examples

Role of the Arts in Neighborhood Change

The arts are widely credited with sparking neighborhood change resulting in both positive—neighborhood revitalization that benefits existing residents—and negative outcomes—gentrification and displacement of lower-income residents. While most studies about the impact of the arts on communities are based on case study methods, Carl Grodach, Nicole Foster, and James Murdoch III (2014) conducted a statistical test using linear regression for 100 large metropolitan areas in the United States.

Figure 12.34 Reading SPSS Data and Drawing a Scatterplot: R Script and Outputs

To be specific, they tested how two different types of arts activities—the fine arts and commercial arts industries—are associated with conditions indicative of urban revitalization and gentrification. The study areas were neighborhoods (zip-code-level) within ten miles of the central business district (CBD) in each metropolitan area.

The dependent variables measuring neighborhood change were based on a set of ten variables found in the literature (e.g., changes in income, workers, educational attainment, the White population, housing value, population density, etc.). To reduce the number of dependent variables to be modeled, they conducted an exploratory factor analysis (see Chapter 5 of Advanced Quantitative Research Methods for Urban Planners) and found three factors that capture different dimensions of neighborhood change: “neighborhood revitalization,” “neighborhood upscaling,” and “neighborhood build-out.” The latter two categories represent gentrification and potential displacement.

Then the authors estimated linear regression models for each of three factors, meaning three different linear regression models (Table 12.5). The key independent variables are two different sets of arts industries—fine arts (e.g., visual and performing arts and museums) and commercial arts (e.g., film, music, design)—and other control variables are a range of social, economic, and housing variables.

Figure 12.35 Running a Linear Regression: R Script and Outputs

The commercial arts have a significant association with both of the gentrification factors, but show no relationship to neighborhood revitalization. Conversely, the fine arts have a positive association with neighborhood revitalization, but correlate negatively with gentrifying neighborhoods. These findings support the claim of the role of the arts in revitalizing central city neighborhoods, but also clarify their role in the gentrification processes, especially by commercial arts industries.

The linear regression was not the only test used in this paper. Because neighborhoods that experience different levels of change may have varying associations with arts activity, they use quintile regression. It divides the dependent variable into five levels of change.

Their findings show that the arts are not necessarily connected with either gentrification or revitalization. Also, the fine arts and commercial arts have different impacts on communities. While fine arts are more associated with revitalization, commercial arts seem to be inducing gentrification (and hence displacement).

Figure 12.36 Durbin-Watson Test and Residual Plots: R Script and Outputs

Table 12.5 OLS Regression Results: The Relationship Between the Arts and Neighborhood Change

»p < .1; »«p < .05; »»*p < .01 Source: Grodach er al., 2014

Residential Yard Management and Crime

W hen it comes to crime, vegetation is a debatable element. It can afford criminals concealment and a place to stash stolen goods or weapons. On the other hand, well-designed trees and lawns can make spending time outdoors more appealing, leading to more “eyes on the street,” informal surveillance networks (Jacobs, 1961). Also, vegetation might discourage crime, as it can be seen as a territorial marker or a cue to care (the inverse of the “broken window" hypothesis) signaling that the residents are actively involved with their surroundings.

Troy, Nunery, and Grove (2016) examined the relationship between residential yard management and neighborhood crime in Baltimore City and County. The researchers surveyed more than 1,000 front yards and collected multiple indicators of yard management vegetation maintenance. Crime data were summarized for a 150-meter buffer around each parcel.

The crime counts were modeled as a function of yard characteristics and surrounding tree canopy, after controlling for socio-economic or housing-related variables. Three statistical approaches—ordinary least squares (linear regression), spatial error regression (see Chapter 9 of Advanced Quantitative Research Methods for Urban Planners) , and Poisson regression (see Chapter 4 of Advanced Quantitative Research Methods for Urban Planners)—were used comparatively to evaluate the stability and consistency of results. OLS was used for its simplicity and ease of interpretation, and this section will focus on the OLS regression results.

The OLS results show that indicators of yard management, from cut lawns to presence of yard trees, are inversely associated with crime, while indicators of neglect, such as litter or desiccation, are positively associated with crime (Figure 12.37). R-squared was 0.58 in the OLS. The results were similar to those in other two advanced models.

The authors admitted that the regression results do not imply causality but the possibility of mechanisms by which environment design may reduce crime. Their theories are cues to care and eyes on the street drawn by active involvement of neighbors and appealing landscape features, which in turn deter criminals.

Significant at 90% level.

Significant at 95% level.

Significant at 99% level

Figure 12.37 Regression Results

Source: Troy el al. (2016)