Skip to main content

Table 4 Random effect probit regression and average marginal effect for determining the valuing of main attributes of patient preference

From: Effect of hospital attributes on patient preference among outpatient attendants in Wolaita Zone, Southern Ethiopia: discrete choice experiment study

Attributes Random effect probit regression Average Marginal Effect (Delta Method)
Coefficient SE 95% CI dy/dx SE 95% CI
Moderate HCP competence 0.27** 0.020 0.23 0.31 0.105 0.008 0.09 0.12
Good HCP competence 0.63** 0.021 0.59 0.68 0.24 0.0084 0.22 0.26
Partially available medical equipment and supply 0.13** 0.019 0.096 0.17 0.051 0.007 0.037 0.066
Fully available medical equipment and supply 0.57** 0.027 0.52 0.62 0.22 0.010 0.197 0.24
Near to the hospital 0.057** 0.017 0.024 0.09 0.022 0.0064 0.009 0.03
No information on hospital reputation 0.17** 0.02 0.14 0.23 0.066 0.0091 0.048 0.084
Moderate hospital reputation 0.087** 0.021 0.046 0.13 0.033 0.0082 0.017 0.049
Good hospital reputation 0.19** 0.026 0.14 0.24 0.072 0.01 0.053 0.091
Waiting time -0.03* 0.0095 -0.046 -0.0086 -0.01 0.0036 -0.017 -0.0033
Service cost -0.00025*** 0.00003 -0.0003 -0.0002 -0.000095 0.000011 -0.00012 -0.000074
Constant -0.43** 0.024 -0.47 -0.38     
Log of variance (lnsig2u) -.1.5674   -4.1846 -0.3149     
Standard deviation (Sigma_u) 0.4567   0.1234 0.8543     
rho (\(\rho )\) 0.1725**   0.1499 0.2219     
  1. Log likelihood = -19,533.95. Number of observations = 29,260. Number of groups = 1045. Wald x2(10) = 439.17. Prob > x2 = 0.0000. * Indicates p-value < 0.05 and ** if p-value < 0.001. dy/dx for factor levels is the discrete change from base level