This paper examines misclassification error in survey estimates of disability. The results suggest that a significant number of those with a disability fail to be recorded as such in the British Household Panel Survey. In addition, the probability of a false positive is estimated as being very close to zero in all socio-demographic groups. There is a strong bias in estimates of differences in rates of disability across groups but only a small effect on estimates of the difference in employment rates by disability status. Copyright © 2014 John Wiley & Sons, Ltd.