Skip to main content

Table 3 The parameter estimates, standard errors, t statistics and confidence level of different models of non-injury crashes, deaths and injuries on the Addis Ababa - Adama/Hawassa main road from 2002-2011

From: Effectiveness of an improved road safety policy in Ethiopia: an interrupted time series study

Models and variables Coefficient Standard error tstatistics Confidence interval
Non-injury crashes: Prais-     
Winston regression (Model 1)     
 Intercept β0 295.1577 33.8195 8.73*** (228.17, 362.14)
 Baseline trend β1 3.248131 .8461874 3.84*** (1.57, 4.92)
 Level change β2 -40.19842 49.2873 -.82 (-137.81, 57.42)
 Trend change β3 -5.096253 1.538459 -3.31** (-8.14, -2.05)
Death: Segmented regression (Model 2)     
 Intercept β0 150.1638 15.06675 9.97*** (120.32, 1.10)
 Baseline trend β1 .3492265 .3795861 .92 (-.40, 1.10)
 Level change β2 3.599756 22.71791 .16 (-41.39, 48.59)
 Trend change β3 -1.961251 .6828936 -2.87** (-3.31, -.60)
Injuries: Segmented regression (Model 3)     
 Intercept β0 96.99802 22.13401 4.38*** (53.16, 140.83)
 Baseline trend β1 1.862395 .557636 3.34** (.75, 2.96)
 Level change β2 -43.5782 33.37405 -1.31 (-109.68, 22.52)
 Trend change β3 -1.489333 1.003214 -1.48 (-3.47, .49)
  1. ***p < 0.0001, **p < 0.01, p < 0.05.