China Policy Journal Volume 1, Number 1, Fall 2018 | Page 104

Subjective and Objective Air Quality in Urban China
Constant 6.37*** 6.37*** 6.35*** 6.47*** 6.37*** 6.41*** 6.39*** 6.37*** 6.38*** 6.40*** 6.69*** 6.59*** 6.64***
(0.15) (0.14) (0.15) (0.14) (0.15) (0.18) (0.15) (0.13) (0.11) (0.12) (0.18) (0.16) (0.17)
Random effects
Variance 0.71 0.64* 0.55** 0.51*** 0.69 0.59** 0.60** 0.52*** 0.36*** 0.43*** 0.58** 0.45*** 0.50***
(L2) (0.18) (0.16) (0.14) (0.13) (0.17) (0.15) (0.15) (0.13) (0.09) (0.11) (0.15) (0.12) (0.13)
Variance 3.63*** 3.55*** 3.56*** 3.56*** 3.55*** 3.56*** 3.56*** 3.55*** 3.56*** 3.56*** 3.55*** 3.56*** 3.56***
(L1) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03)
AIC 103,938 95,487 92,775 92,773 95,489 92,777 92,778 95,480 92,762 92,767 95,484 92,769 92,772
BIC 103,962 95,552 92,855 92,853 95,554 92,858 92,858 95,545 92,843 92,847 95,548 92,849 92,853
N 25,139 23,221 22,543 22,543 23,221 22,543 22,543 23,221 22,543 22,543 23,221 22,543 22,543
Note: Standard errors in parentheses. ***p<0.01, **p<0.05, *p<0.1.
either AQTI or PITI as the moderator
(see Models 6 and7). In
terms of PM 10
, the moderating
effects of both AQTI (β=0.80,
p<0.01) and PITI (β=0.83,
p<0.05) are supported (see Models
9 and 10). In the case of AQI,
only the moderating effect of
AQTI is significant and positive
(β=0.03, p<0.05) (see Models 12
and 13). In sum, the substantially
similar results among the four air
pollutants corroborate Hypothesis
2.
To visually illustrate the
moderating effect of environmental
transparency on the relationship
between objective and
subjective air quality measures,
we draw the marginal changes
of simple slopes of air pollution
as the function of the two moderators
in Figure 2. In Panel A
of Figure 2, we reveal that air
pollution (measured by SO 2
) is
negatively associated with subjective
air quality when PITI is at
the −1 standard deviation (SD)
condition, but its sign turns to
be slightly positive at the +1 SD
condition. The result suggests the
negative effect of air pollution
(herein SO 2
) on subjective air
quality is attenuated with the increment
of environmental transparency
(PITI). When the value
of centered PITI (ranging from
−27.355 to 33.145) is larger than
the lower confidence bound of
region at the 0.05 level (−2.2323),
the slopes of air pollution turn to
be statistically insignificant. Pan-
101