Evidence particularly from developing countries that

Info icon This preview shows pages 17–19. Sign up to view the full content.

evidence, particularly from developing countries, that it is correlated with variables, such as income and education, which are potential determinants of true health (Strauss and Thomas, 1998; Sadana et al, 2000). Therefore, the use of subjective health measures like reporting of illness and self-assessed health level remains problematic. 36. Anthropometrics are basically measures of height and weight standardised for age and sex and compared to an international standard for normal child growth. 17 There is good evidence of negative correlation between child anthropometric measures and indicators of ill health (World Health Organisation, 1995). Certain caveats are warranted, however, with respect to the suitability of anthropometrics in examination of the health effects of child work activity. The indicators mainly reflect current or past nutritional status and so, if they are used as health outcomes, a crucial control variable is current or past calorie intake. If this is not available, then omitted variable bias will be a problem if calorie intake is correlated with child work activity, as seems likely. A second caveat is that the appropriate indicator must be selected depending upon whether the relationship under examination is short or long run. Weight-for- height is mainly an indicator if acute malnutrition and is not particularly relevant to examination of the health impact of child labour. Height-for-age is a better indicator of long-term health experience but mainly reflects health and nutritional exposure in early childhood and is of limited use in estimating the health effects of child work. A particular problem with the use of anthropometrics in the context of child labour is that they are better measures of nutrition and health experience at younger ages. As the child ages, stature is more likely to be a reflection of genetic factors. Many studies using height-for-age and weight-for-height restrict attention to children no older than 10 years, excluding the age range in which child labour is most prevalent. 37. Within the constraints imposed, the body mass index (BMI) and measures of self- reported morbidity appear to be the most promising measures of health. Each has its limitation and, at a minimum, experimentation with a number of health measures is advisable. The best strategy might be to explicitly recognise the measurement 17 There are four indicators height-for-age, weight-for-age, weight-for-height and the body mass index. The World Health Organisation recommends standardisation on the US average (de Onis and Habicht, 1996).
Image of page 17

Info icon This preview has intentionally blurred sections. Sign up to view the full version.