Abstract:
This paper uses a combination of the 2006 Catalan Health and Health & Examination Surveys to compute the size of weight and height self-reporting biases. The underlying determinants of these self-reporting biases are also analysed, placing special emphasis on examining the role played by social norms. Our findings show that social norms have a dual impact on these self-reporting biases: the higher the distance between individuals’ measured weight (height) and that of their average reference group, the higher (lower) the weight (height) bias or more (less) inclined they are to misreport their weight (height). This evidence confirms the influence of group-specific factors or social norms on self-reports of attributes. Finally, the relationship found between the measured and self-reported anthropometric data in our merged database was applied to the Spanish National Health Survey (NHS) to correct its self-reported information. Interestingly, after correcting for self-reporting errors, both the BMI and the prevalence of obesity were significantly underestimated; the increase was higher among women.
New Economics Papers: this item is included in nep-lab Date: 2009-01