Self-rated health and schooling in 69 countries: what the 2002 World Health Survey showed
Three researchers used the 2002 World Health Survey to test whether one question about health gives sensible answers, by checking whether adults with less schooling reported poor health more often.
What the study did and found
The paper, “Self-reported health assessments in the 2002 World Health Survey: how do they correlate with education?”, appeared in the Bulletin of the World Health Organization in 2010 (Subramanian et al., 2010). It analysed 219,713 adults aged 25 or over in 69 countries.
Overall 9.8% reported poor health, and each extra year of schooling went with about 7% lower odds of doing so; the least-schooled fifth had 2.29 times the odds of the best-schooled fifth, in every income group and region (Subramanian et al., 2010). National rates, however, tracked neither median schooling nor life expectancy.
Why it is cited
The referring page is a citation entry in WHO’s Multi-Country Studies Data Archive, the organisation’s catalogue of survey microdata. It gives the journal details and the authors’ abstract, with the bare address of this record as its only link.
The use is neutral, since the catalogue adds no claims of its own, although it labels the datasets “World Health Survey 2003” and links only 31 of the 69 country datasets the paper analysed.
The study used a single survey round and adjusted only for age and sex (Subramanian et al., 2010). A gradient by schooling is what a valid measure should produce, but it cannot prove that answers are accurate, and less-educated people might still understate poor health.
Later work gives partial support: cohorts of older adults in rural Indonesia and India found that poor self-rated health predicted death (Ng et al., 2012; Hirve et al., 2012), while vignette studies found that education can shape how people use the answer scale.
The topic today
Five-point health questions remain standard in European surveys and in the census of England and Wales. In the EU in 2025, 56.1% of people with at most lower secondary education rated their health good or very good, against 79.1% of those with tertiary education (Eurostat, 2026).
For the current picture, see our explainer Self-Rated Health and Education: Findings from 69 Countries.