June 1, 2009
This paper proposes that an individual's self-assessed health (SAH) does not only suffer from systematic reporting bias and adaptation bias but is also biased owing to confounding health norm effects. Using 13 waves of the British Household Panel Survey covering the period 1991-2005, I show that, while there is a negative and statistically significant correlation between SAH and individuals' own health problem index, this negative effect reduces with the average number of health problems per (other) family member. The relative health bias is small, however, which implies that measures of SAH may not suffer seriously from systematic health norm bias. This is an important finding for researchers working with SAH data as it indicates that we do not have to worry too much about controlling for confounding influences from the health of other household members when estimating SAH regression equations.
Social Science and Medicine
Volume: 68 (2):251-259
Previously 'In press, corrected proof' 6 Dec 2008
Albert Sloman Library Periodicals *restricted to Univ. Essex registered users*