Skip to main content

Table 4 Geographic variability of associations between explanatory variables and depression between the five NYC boroughs

From: The geography of post-disaster mental health: spatial patterning of psychological vulnerability and resilience factors in New York City after Hurricane Sandy

  Manhattan Bronx Brooklyn Queens Staten Island Chow test
  N = 125 N = 37 N = 96 N = 93 N = 67 N = 5
  Est. S.E. Est. S.E. Est. S.E. Est. S.E. Est. S.E. Est.
Age -.02 .02 -.09 .06 .04 .03 -.01 .03 .01 .03 6.08
Female -.31 .81 1.07 1.86 .50 1.21 −1.07 1.08 2.09* 1.02 5.36
Asian -.42 1.32 .28 1.43 -.96 .94 7.10** 2.48 −1.25 .97 10.48*
Non-Hispanic Black -.24 1.03 −2.11 2.20 2.01 1.62 1.82 1.36 −1.62 1.79 4.67
Hispanic 1.55 .92 −1.71 2.57 1.29 1.70 .74 1.69 .37 1.65 1.68
Other race/ethnicity -.39 1.54 3.28 2.44 1.80 2.33 -.07 2.61 3.39 2.00 3.28
High school education or less 1.08 1.10 4.74* 2.30 -.01 1.33 1.46 1.37 2.01 1.19 3.59
Employed −1.67 .95 .45 2.28 −1.68 1.01 -.26 1.12 .36 .89 3.79
Married or cohabiting -.40 .81 1.20 1.95 −1.07 .87 -.51 1.08 −2.74*** .80 6.56
Parent, living with child at time of Sandy -.60 .96 −5.61** 2.18 2.52* 1.18 1.56 1.31 -.10 .88 12.91*
Experienced or witnessed trauma in addition to Sandy 2.07** .79 2.14 1.85 .92 .95 2.11 (.07) 1.17 -.07 .80 4.62
Number of hurricane-related stressors .75 .79 .81 1.45 .76 .59 1.34** .41 0.95 (.07) .52 1.03
Number of hurricane-related trauma −1.38 .88 6.34* 2.95 .65 1.56 .04 1.27 8.63** 2.71 17.37**
Global spatial lag
Spatially lagged depression across all boroughs -.39 .7          
Model fit            
Pseudo R2 .25           
Spatial pseudo R2 .24           
Global structural stability test            
Chow test 103.82***           
Spatial dependence
Anselin-Kelejian test .64           
  1. N = Number of cases, Est. = Coefficient estimate, S.E. = Standard error, Significance level: <0.001***, <0.01**, <0.05*. P-value reported in brackets when marginal significant. Spatial neighborhood definition: 9 km, N total = 418.
\