Such estimates would not be helpful in the

Info icon This preview shows pages 16–18. Sign up to view the full content.

Such estimates would not be helpful in the identification of the most harmful forms of child labour. In order to take the analysis further, more detailed measures of child work activity, which provide information both on the intensity of work and the sector of employment, are required. But such detailed measurement must be combined with a large sample of child workers in order for there to be sufficient numbers of various types of child workers to facilitate estimation of heterogeneity in the impact of child labour on health. 34. Of course, large detailed surveys are expensive, a factor that constrains the measures of health available for analyses in relation to child work activity. Detailed clinical measures are unlikely to be available, leaving the researcher with a choice between anthropometric measures and self-reported indicators of morbidity. The latter can relate to acute sickness, chronic illness and assessments of general health status. Indicators of chronic conditions and general health status are preferable for analyses of the long-term health consequences of child work activity, acute sickness being a noisier indicator of the individual’s underlying health status and better suited to analyses of the short-run impact of child work on health. The most widely available morbidity indicator for children in the developing world refers to any illness 16 See discussion of Kassouf et al (2001) and Guiffrida et al (2001) above.
Image of page 16

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