Skip to main content
. 2020 Nov 16;8(4):687. doi: 10.3390/vaccines8040687

Table 5.

Results of mixed logit model with main effects and interactions.

Attributes β SE p-Value 95% CI
Non-vaccination −3.112 0.638 <0.001 −4.361 −1.862
Protection rate prevented by a vaccine (ref: 65%)
80% 0.501 0.075 <0.001 0.355 0.647
95% 1.386 0.115 <0.001 1.160 1.611
Risk of severe side effect event (ref: high)
moderate 0.794 0.077 <0.001 0.643 0.944
low 1.609 0.201 <0.001 1.215 2.003
Location of vaccine manufacturer (ref: domestic)
Imported −0.403 0.091 <0.001 −0.582 −0.224
Duration of vaccine-induced protection (ref: 1 year)
5 years 0.315 0.072 <0.001 0.174 0.457
10 years 0.517 0.096 <0.001 0.329 0.706
Out-of-pocket Cost −0.001 0.000 <0.001 −0.002 −0.001
Severity of diseases prevented by vaccines (per 1%) 0.003 0.011 0.799 −0.018 0.024
Interaction terms
Non-vaccination * working 0.683 0.622 0.272 −0.536 1.902
Non-vaccination * age (>30 years old) 0.584 0.458 0.203 −0.315 1.482
Non-vaccination * father −1.576 0.585 0.007 −2.723 −0.428
Non-vaccination * rural −1.283 0.500 0.010 −2.263 −0.304
Non-vaccination * education level (college and above) 0.280 0.501 0.576 −0.701 1.261
Non-vaccination * health state (fair/poor) −1.899 0.854 0.026 −3.573 −0.224
Lowest risk of severe side effect * education level (college and above) 0.445 0.193 0.021 0.068 0.823
Lowest risk of severe side effect * rural −0.299 0.190 0.116 −0.671 0.074
Lowest risk of severe side effect * age (>30 years old) −0.155 0.187 0.406 −0.521 0.211
90% protection rate * age (>30 years old) −0.362 0.143 0.012 −0.642 −0.081
Protection duration of 10 years * father 0.433 0.204 0.034 0.033 0.833
Imported vaccine * education level (college and above) 0.468 0.124 <0.001 0.225 0.712
Log likelihood −3234.7672
AIC 6100.574
BIC 6339.304
Respondents, n 428
Observations, n 12840

Note: 1. CI—confidence interval. * — the multiplicative relationship which represents the interaction effect of two variables. All attributes except for cost and severity of diseases prevented by vaccines were coded as dummy variables. 2. Interaction terms were treated as fixed effect variables, and the others as random effect variables.