Mapping child maltreatment risk: a 12-year spatio-temporal analysis of neighborhood influences

Table 2 Results of different spatial and spatio-temporal regression Bayesian models for child maltreatment risk. Posterior mean, standard deviation (SD) and the 95% credible interval (CI) of all parameters

	Model 1 (β)			Model 2 (β + spatial heterogeneity + spatial effect)			Model 3 (β + spatial heterogeneity + spatial effect + temporal heterogeneity)			Model 4 (spatio-temporal autoregressive model)
	Mean	SD	95% CI	Mean	SD	95% CI	Mean	SD	95% CI	Mean	SD	95% CI
Intercept	4.055	.289	3.500, 4.615	4.335	.516	3.320, 5.328	4.284	.520	3.916, 5.304	4.135	.500	3.274, 5.127
Economic status^a	−.021	.003	−.027, −.015	−.016	.004	−.024, −.009	−.016	.004	−.023, −.009	−.016	.004	−.023, −.008
Education level	−1.261	.091	−1.431, −1.083	−1.464	.161	−1.745, −1.123	−1.418	.164	−1.746, −1.096	−1.391	.157	−1.690, −1.122
Policing activity	.026	.006	.014, .038	.036	.011	.012, .053	.035	.012	.012, .057	.031	.011	.009, .053
Residential instability	.000	.001	−.001, .001	.000	.001	−.001, .001	.000	.001	−.001, .001	.000	.001	−.001, .001
Immigrant concentration	.003	.005	−.006, .013	.005	.006	−.005, .016	.005	.005	−.005, .016	.009	.006	−.003, .020
σ_θ				.329	.084	.136, .468	.320	.095	.063, .456	.234	.045	.162, .333
σ_ϕ				.781	.115	.541, .979	.787	.118	.552, .976	.257	.062	.149, .391
σ_α							.023	.019	.001, .070	.021	.019	.001, .070
ρ										.903	.031	.827, .946
DIC	8517.9			8164.9			8166.8			8126.1

SD standard deviation, CrI credible interval, DIC deviance information criterion
σ_θ standard deviation unstructured term
σ_ϕ standard deviation spatially structured term
σ_α Standard deviation temporally unstructured term
^aThis variable was included as the cadastral value divided by 1000 to solve computational problems with the prior distributions assigned to fixed effects

ISSN: 1476-072X