Abstract
In China, rising obesity has coincided with increasing affluence. Few studies have properly accounted for geographic variation, however, which may influence prior results. Using large data with biomarkers in China, we show body mass index (BMI) to be positively correlated with higher person-level education if estimated using ordinary least squares. In stark contrast, fitting the same data within a multilevel model gives the complete opposite result. We go on to show that the relationship between BMI and person-level education in China is dependent upon geography, underlining why multilevel modelling is crucial for revealing these types of people-place contingencies.
Original language | English |
---|---|
Pages (from-to) | 1-5 |
Number of pages | 5 |
Journal | Health and Place |
Volume | 46 |
DOIs | |
Publication status | Published - 1 Jul 2017 |
Bibliographical note
Publisher Copyright:© 2017
Keywords
- China
- body mass index
- education
- geography
- health literacy
- multilevel models (statistics)