Abstract
Shmueli (1996) highlights the problems raised by the use of measures of current health, and by the omission of past health, in retrospective analysis of cross section data on the prevalence of smoking. In this note the findings in Jones (1994) are augmented by new estimates from the longitudinal follow-up to the Health and Lifestyle Survey (HALS2).
| Original language | English |
|---|---|
| Pages (from-to) | 755-759 |
| Number of pages | 5 |
| Journal | Journal of health economics |
| Volume | 15 |
| Issue number | 6 |
| Publication status | Published - Dec 1996 |
Keywords
- smoking
- instrumental variables
- Health and Lifestyle Survey