Questions

What are the spatial patterns of natural resource governance perceptions in relation to environmental condition?

 

Environmental condition was represented as: (1) Environmental restoration sites, and (2) aggregate ‘environmental effects’ scores by census tract on a scale from (1) low to (10) high that include lead risk from housing, proximity to hazardous waste treatment storage and disposal facilities, proximity to nation priorities list facilities (Superfund Sites), proximity risk management plan facilities, and wastewater discharge.

 

Sub-questions related to the primary question are:

a1. Does being near a restoration site associate with individual perception? a2. Does being near more restoration sites associate with individual perception?

b1. Do the environmental effects where an individual lives associate with individual perception? b2. Do the environmental effects around an individual associate with individual perception?

Tools and Approaches

1. Nearest Neighbor analysis in ArcGIS Pro and R studio
2. Geographically weighted regression in R studio
3. Neighborhood analysis in ArcGIS Pro

Analysis Steps

1. Nearest neighbor analysis was used to examine questions a1 and a2. First, a file of restoration sites was loaded into R. The sites were points in the same projection as my participant location points. This analysis required four libraries (as the points were from different files. The libraries were: nlem, rpart, spatstat, and sf. To use the tool I needed, I first had to turn my points into point objects in R (.ppp objects). First I used the convenxhull.xy() function to create the ‘window’ for my point object, then I used that to run the ppp() function. After doing this for both sets of points, I was able to use the nncross() function. This function produced min, max, average, and quartile distances from individuals to the nearest restoration site. I added the ‘distance’ from the nearest neighbor as a variable in a linear regression to determine an association for a1.

 

To examine a2, I used these distances (1^st quartile, median, and 3^rd quartile) to produce buffers in ArGIS Pro. After creating the buffers, I ran spatial joins between them and the restoration sites. This produced an attribute table that had a join count—the number of restoration sites within the buffer. I exported the three attribute tables from the three buffer distances back to R. In R I ran a linear regression with join count as an independent variable.

 

2. To test b1, I preformed geographically weighted regression. I used both ArcGIS Pro and R studio to run this analysis. Initially I used the GWR tool in Arc to run the regression, but wanted the flexibility of changing parameters and an easily readable output that R provides. First I joined individual points to my rank data, a shapfile at census tract level. This gave individuals a rank for environmental effects at their location. In R, I used two different functions to run GWR, gwr(), and gwr.basic(). The gwr() required creating a band using gwr.sel, and the gwr.basic required creating a band using bw.gwr. The difference between these functions is that gwr.basic produces p-values for the betas. I ran gwr on both my entire data set and a subset based on perceptions. The subset was the ‘most agreeable’ and ‘least agreeable’ individuals who I defined as those one standard deviation above and below the mean perception.

 

3. I preformed neighborhood analysis to test the final question, b2. First, I created a dataset that was just the upper and lower values of my governance perceptions (one standard deviation above and below the means. I then added buffers to these points at 1 and 5 miles. I then joined the buffers in ArcGIS Pro to the rank values to get an average rank within those buffers. I exported the files for each buffer to R. I R I created a density plot of average rank for the low governance values at each buffer, and for the high governance values at each buffer.

Results

1. The median distance for individuals from restoration site was 0.037 degrees, 1^st quartile was 0.020 degrees, and 3^rd quartile was 0.057 degrees.

The regression on whether distance correlates with individual perception was insignificant (p = 0.198). This led to the conclusion that distance from the nearest restoration site does not influence perceptions.

 

For each regression on the number of sites near an individual, all coefficients were negative. This implies that the more sites near an individual, the more disagreeable their perception was. All produced significant results, but the effect size of number of sites near individuals was very minimal (Table 1).

 

Table 1. Regression results for ‘nearest neighbor’ of individuals to restoration sites.

Buffer Size	b	p-value	Effect Size (rpb)
Buffer 1 (1st quartile)	-0.003	0.0181	.003
Buffer 2 (median)	-0.002	0.045	.002
Buffer 3 (3rd quartile)	-0.001	0.002	.003

 

 

 

 

2. For the geographically weighted regression, I am presenting the results from the gwr.basic() model. For this model, I included demographic variables to control for these factors. In the model, rank, life satisfaction, years lived in the Puget Sound, and Race are significant (Table 2). All other variables were insignificant, so I will not discuss their trends. For rank (the main variable of interest), the coefficient was positive. In this case higher rank values are worse environmental effects, so as agreeable perceptions increase, environmental condition decreases. Life satisfaction is a variable of how satisfied individuals are with their life overall, which correlated positively with perception (Table 1). Years lived in the Puget Sound correlated negatively, or perception decreased the longer someone lives there (Table 1). Race, a dummy variable of white or not-white, indicated higher perceptions were held by white individuals.

 

Overall, the effect size of this model on governance perceptions was small, explaining about 10% of the variance in the data (Table 1).

 

Table 1. Regression results for environmental effects at individuals’ locations.

Variable1	b	p-value2
Rank	0.077	0.007**
Life Satisfaction	0.478	<0.001**
Years Lived in the Puget Sound	-0.010	0.002**
Sex	-0.156	0.289
Area	-0.150	0.179
Education	-0.002	0.922
Income	-0.022	0.622
Race	0.006	0.034*
Ideology	0.103	0.533

¹R² = 0.094

² ** = significant at the 0.01 level, * = significant at the 0.05 level

 

3. The plot of high and low governance values at the two buffers is presented below. The black and red curves represent respondents from the survey that were at least one standard deviation lower than the mean (the mean was neutral). The black curve is average rank with a one mile buffer, and the red curve is average rank at the five mile buffer. The green and blue curves represent respondents from the survey that were at least one standard deviation higher than the mean. The green curve is the average rank at the one mile buffer, and the blue curve is the average rank at the five mile buffer.

What this figure indicates is that there are two peaks in average rank at low environmental effects (~1) and at mid-environmental effects (~4.75). Those with lower perceptions of environmental governance had higher peaks at low environmental effects for each buffer size. Those with higher perceptions of environmental governance had a bimodal distribution with peaks at low and mid-environmental effects. The bimodal nature of these density functions leads me to believe there is some variable moderating governance perception related to environmental effects.

 

4. Critique

The methods I used were all helpful in determining the spatial relation of environmental condition to perceptions of natural resource governance. However, switching between the two programs (ArcGIS and R) was a little bit of a hassle. R has greater flexibility when running analyses, as in running is easier. This meant it was ideal for the analyses that required me to tweak my models multiple times. I also ran into little problems with Arc in terms of loading the data which made everything run slower, but Arc is more intuitive in terms of finding and executing analyses/