Obesity and its relation to employment incomes: does the bias in self-reported BMI matter?

Abstract

This study explores what difference, if any, the bias in self-reported body mass index (BMI) has on our understanding of the relationship between body size and income attainment. To accomplish this, aggregated data from Cycle 1 and Cycle 2 of the Canada Health Measures Survey, in which information on both self-reported and measured BMI was collected, are used. Based on subsamples of female and male employees, OLS regression analyses contrasting the effect of self-reported and measured BMI on income show that for women, self-reported BMI leads to underestimates of a negative body size effect, whereas for men, self-reported BMI leads to overestimates of a positive body size effect. Additional analyses examining the appropriateness of correction factors to improve the accuracy of self-reported BMI effect estimates suggest correction factors do little to reduce these systematic errors.