Skip to main content

Table 1 Results from the Bayesian changepoint analysis

From: Timing and effect of a safe routes to school program on child pedestrian injury risk during school travel hours: Bayesian changepoint and difference-in-differences analysis

Variable Posterior mean 95% CrI
SRTS Intervention
Changepoint (δ), number of quarters from January 2001 30.5 (30.02, 30.98)
β0 (Pre-changepoint intercept) 1.25 (0.90, 1.58)
β1 (Pre-changepoint slope) 0.03 (0.02, 0.05)
β2 (Pre- and post intercept difference) −1.68 (−2.59, −0.84)
β3 (Pre- and post slope difference) 0.04 (−0.1, 0.17)
β0 + β2 (Post-changepoint intercept) −0.43 (−1.45, 0.52)
β1 + β3 (Post-changepoint slope) 0.07 (−0.06, 0.21)
Non-SRTS Intervention
Changepoint (δ), number of quarters from January 2001 10.49 (10.02, 10.97)
β0 (Pre-changepoint intercept) 4.79 (4.74, 4.85)
β2 (Pre- and post intercept difference) −0.29 (−0.36, −0.27)
β0 + β2 (Post-changepoint intercept) 4.50 (4.46, 4.53)
  1. Posterior means and 95% Credible Intervals (CrI) on log scale. Safe Routes to School (SRTS) intervention census tracts and non-SRTS intervention census tracts, New York City, 2001–2010.