Table 3.
Multilevel Logistic Regression Estimates: Fixed Effect (Odds Ratios) and Random Effects of Practice of Nonmedical Healthy Lifestyle (Lifestyle)
Variables | Model 1 | Model 2 | Model 3 | Model 4 |
---|---|---|---|---|
Fixed effect | ||||
Individual variables (level 1) | ||||
Age | 0.9966 (0.0021) | 0.9966 (0.0021) | ||
Female | 4.6333*** (0.3154) | 4.6301*** (0.3151) | ||
High education | 1.2162** (0.0980) | 1.2160** (0.0980) | ||
Urban residence | 2.2107*** (0.2987) | 1.9944*** (0.3049) | ||
Medical insurance | 1.1185 (0.0866) | 1.1091 (0.0863) | ||
Low household income pc | 0.7966*** (0.0670) | 0.8014*** (0.0675) | ||
High household income pc | 1.1798** (0.0932) | 1.1720** (0.0927) | ||
Distance to the nearest facility | 1.1965** (0.0907) | 1.2104**(0.0920) | ||
Level 2 (community variables) | ||||
Number of facilities | 0.9877 (0.0258) | 1.0346 (0.0516) | ||
Community per capita income | 0.9999 (0.0028) | 1.0001 (0.0003) | ||
Random effect | ||||
Intra-class correlation, ρ | 0.2006*** (0.0227) | 0.1673*** (0.0210) | 0.1633*** (0.0208) | 0.1618*** (0.0207) |
Log-likelihood | −4144.21 | −4133.80 | −3814.65 | −3816.23 |
Number of observations | 9733 | 9733 | 9733 | 9773 |
Notes: Odd ratios are reported, standard errors are in parentheses.
*p < 0.10,
**p < 0.05,
***p < 0.01.