Page 359 - Contributed Paper Session (CPS) - Volume 4

CPS2277 Iris M H Yeung et al.
Edu (post- 0.193 0.406 1.213 -0.136 0.359 0.873 0.167 0.281 1.182
secondary)
Age -0.031*** 0.008 0.969 -0.011 0.008 0.990 -0.022*** 0.006 0.978
Household -0.099 0.088 0.906 -0.107 0.079 0.898 -0.071 0.061 0.931
size
House type 0.121 0.296 1.129 0.130 0.269 1.139 0.118 0.208 1.125
(HOS)
House type 0.286 0.245 1.331 0.331 0.222 1.393 0.184 0.171 1.202
(private)
Income 0.591*** 0.170 1.806 0.108 0.170 1.114 0.441*** 0.121 1.554
Notes:
1. ***, **, and * denote the significance levels of 1%, 5%, and 10%, respectively.
2. Overall model evaluation criteria
Multinomial: Likelihood ratio statistic = 216.04 (p value < 0.0001), Generalized R-square =
0.281
Ordinal: Likelihood ratio statistic = 194.305 (p value < 0.0001), Generalized R-square = 0.256

Table 2: Distribution of Three WTP Groups for Knowledge on Waste Charge
Method, Landfill Fullness and Construction Time for New Incinerator (n=753)

Sample Below HK$30 Exactly Above Test results p-value
size (n=272) HK$30 HK$30
(n=244) (n=237)
Know waste Yes 385 31.95% 32.47% 35.58% χ2 = 8.030** 0.018
charge No 368 40.49% 32.34% 27.17%

Know landfill Yes 171 26.90% 36.26% 36.84% χ2 = 8.248** 0.016
No 582 38.83% 31.27% 29.90%

Know Yes 55 41.82% 18.18% 40.00% χ2 = 5.607* 0.061
incinerator No 698 35.67% 33.52% 30.80%
Note: ***, **, and * denote the significance levels of 1%, 5%, and 10%, respectively.

4. Discussion and Conclusion
The current paper identifies the factors affecting residents’ WTP for waste
disposal in Hong Kong using multinomial and ordinal logit models. Both
models suggest that higher degree of support for waste charge and building
new incinerator policies, more daily waste disposal amount, younger age and
higher income increase the likelihood of WTP for the contrast between “above
HK$30” and “below HK$30”. Multinomial logit model further shows that only
higher degree of support for waste charge, knowledge on landfill fullness and
construction time for new incinerator affect the likelihood of WTP for the
contrast between “exactly HK$30” and “below HK$30”.
The multinomial and ordinal logit models provide more information on the
factors affecting residents’ WTP as compared with binary logit models (Yeung
and Chung 2018). So, the government might try to launch promotional and
educational programs on both waste charge and new incinerator policies, and
information on landfill fullness as they all have positive effects on WTP.

348 | I S I W S C 2 0 1 9