Observer variability in anthropometry

Abstract

Sixteen trained field workers, woking in eight teams, measured eight children to find the observer variability in the anthropometric measurements. All measurements were found to be subjected to a significant observer error, stature being the one with least error and skinfold thickness the one with the highest error. It is recommended that the latter measurement should not be used unless obsolutely indicated. The importance of recognising the high degree of observer error is discussed and certain precautions to reduce it are indicated.