Skip to main content
. 2024 Mar 18;24:265. doi: 10.1186/s12877-024-04783-y

Table 6.

Heterogeneity analysis: income level

Income level Panel A:Dependent variable: consciously controlling sugar intake Panel B:Dependent variable: consciously controlling salt intake
Pension Optimal bandwidth N Pension Optimal bandwidth N
(1) Household income Low-income 0.630** 4.156 293;324 0.451* 4.458 293;324
(0.287) (0.266)
Medium-income 0.059 5.734 373;307 0.170 5.717 373;307
(0.312) (0.329)
High-income 0.811 4.499 291;219 0.488 4.990 291;219
(1.643) (1.169)
(2) Income satisfaction Satisfied 0.113 4.336 476;471 0.038 4.284 476;471
(0.401) (0.399)
Commonly 0.622* 5.210 241;205 0.538* 5.175 241;205
(0.323) (0.318)
Dissatisfied 1.026 3.700 136;134 1.198 3.358 136;134
(0.899) (1.021)
Income level Panel C:Dependent variable-Consciously controlling edible oil intake Panel D:Dependent variable-Learning health knowledge or wellness knowledge
Pension Optimal bandwidth N Pension Optimal bandwidth N
(1) Household income Low-income 0.410 4.267 293;324 0.264 4.569 293;323
(0.264) (0.241)
Medium-income 0.518 5.404 373;307 0.493 5.096 373;307
(0.416) (0.417)
High-income 0.218 4.973 291;219 −3.146 2.871 101;136
(1.166) (23.737)
(2) Income satisfaction Satisfied 0.132 4.530 476;471 0.065 4.230 476;470
(0.401) (0.424)
Commonly 0.409 5.170 241;205 0.422 4.994 188;159
(0.326) (0.321)
Dissatisfied 1.052 4.229 199;166 0.155 4.965 199;166
(0.766) (0.476)

The robust standard error of clustering to the township level is reported in parentheses, with * p < 0.1, * * p < 0.05, * * * p < 0.01. The default minimum mean square error bandwidth is used in the table