The extensive use of weighing and measuring in the campaign against the malnutrition of children demands standards of mean weight for height, age and sex and standards of normal variation of the highest attainable reliability. The Baldwin-Wood tables are perhaps as satisfactory standards of mean weight for general use as we are likely to have; but the standards of normal variation—after all, the really vital part of the weight-height method—now in general use are far from satisfactory. Indeed it would appear that they have been selected somewhat arbitrarily on the basis of clinical impressions, modified by a desire for simplicity rather than from a mathematical appraisal of the observed variability.
The result of the application of these standards, the reliability of which is open to question and the interpretation of which has been far too rigid, is a rapidly growing suspicion by the medical profession and the public that a
FABER HK. VARIABILITY IN WEIGHT FOR HEIGHT IN CHILDREN OF SCHOOL AGE. Am J Dis Child. 1925;30(3):328–335. doi:10.1001/archpedi.1925.01920150044004