Andrew M. Jones 579(ECHP). They explore its consequences for dynamic models of the association
between socioeconomic status and self-assessed health (SAH). Descriptive evidence
shows that there is health-related non-response in the data, with those in very poor
initial health more likely to drop out, and variable addition tests provide evidence
of non-response bias in the panel data models of SAH. Nevertheless a comparison
of estimates – based on the balanced sample, the unbalanced sample and corrected
for non-response using inverse probability weights – shows that, on the whole,
there are not substantive differences in the average partial effects of the variables
of interest.
Inverse probablity weights are used to attempt to control for attrition: this works
by estimating separate probit equations for whether an individual responds or does
not respond at each of the waves of the panel. Then the inverse of the predicted
probabilities of response from these models are used to weight the contributions to
the log likelihood function in the pooled probit models for SAH. The rationale for
this approach is that a type of individual who has a low probability of responding
represents more individuals in the underlying population and therefore should be
given a higher weight. The appropriateness of this approach relies on the assump-
tion that non-response is ignorable conditional on the variables that are included
in the models for non-response (“selection on observables”). If this assumption
holds then inverse probability estimates give consistent estimates. The findings in
Joneset al.(2006) and the earlier work by Contoyanniset al.(2004b) suggest that,
while health-related non-response clearly exists, on the whole it does not appear
to distort the magnitudes of the estimated dynamics of SAH and the relationship
between socioeconomic status and SAH. Similar findings have been reported con-
cerning the limited influence of non-response bias in models of income dynamics
and various labor market outcomes and on measures of social exclusion, such as
poverty rates and income inequality indices.
12.3.2 Health outcomes
12.3.2.1 Self-reported data
Self-assessed health is often included in general social surveys. For example, in
the BHPS, SAH is an ordered categorical variable based on the question: “Please
think back over the last 12 months about how your health has been. Compared
to people of your own age, would you say that your health has on the whole
been excellent/good/fair/poor/very poor?” The validity of self-reported measures
of health has caused considerable debate. As a self-reported subjective measure of
health, SAH may be prone to measurement error. General evidence of non-random
measurement error in self-reported health is reviewed in Currie and Madrian (1999)
and Lindeboom (2006).
Self-assessed health is not the only source of concern with self-reported data.
Bakeret al.(2004) use careful record linkage to check for flaws in self-reported data
on specific chronic conditions. Survey data from the Canadian National Popula-
tion Health Survey for 1996–97 are linked to ICD-9 (International Classification of
Diseases – ninth revision) codes for Ontario residents from administrative data on