Related Experiment Video
Updated: Aug 10, 2026

Assessment of Child Anthropometry in a Large Epidemiologic Study
Published on: February 2, 2017
Meditations on birth weight: is it better to reduce the variance or increase the mean?
1Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, MA 02138, USA. dhaig@oeb.harvard.edu
Abstract:
A conceptual model is presented here in which the birth weight distribution is decomposed into a distribution of target weights and a distribution of perturbations from the target. The target weight is the adaptive goal of fetal development. In the simplest model, perinatal mortality is independent of variation in target weight and determined solely by the magnitude of the perturbation of birth weight from the target. In this model, mortality risk is concentrated in the tails of the birth weight distribution. A difference between populations in their distributions of target weights will be associated with a corresponding shift in their curves of weight-specific risk, without any difference between the populations in overall risk. In this model, risk would be reduced by decreasing the variance of the distribution of perturbations. The model is discussed in the context of the so-called "paradoxes of low birth weight."
Related Concept Videos
Regression Toward the Mean
Variation: Normal Distribution, Range, and Standard Deviation
Weighted Mean
For example, consider the number of goals scored in the matches of a tournament. While computing the average number of goals scored in the tournament, it may be more important to...
Trimmed Mean
Although certain measures of central tendency are not sensitive to outliers, there are alternative versions of the mean that get around the...
Estimating Population Mean with Known Standard Deviation
The confidence interval estimate will have the form as follows:
(point estimate - error bound, point estimate + error bound)
The...
Testing a Claim about Standard Deviation
The hypothesis testing for the claim of population standard deviation (or variance) requires the data and samples to be random and unbiased. The population distribution also must be normal. There is no specific requirement on the sample size as the estimation is based on the chi-square distribution.
As a first step, the hypothesis (null and alternative) concerning the claim about...

