Craig Baumrucker’s Blog

First, most dietary milk and dairy product links to disease (cancer, etc) are derived from correlations studies utilizing rating scales (see below).  These studies rely on survey data obtain from individuals who list their diets (mostly from memory).  Most current surveys are enhanced by the fact that they are obtained from “professional nurses” who are better at recording or remembering their diets, but recent information suggests that such information is around ~60% reliable.  However, the accumulation of thousands of survey results from numerous countries provides statistical power to diet and correlations.

Statistical Correlations:  This is a statistical technique which can indicate whether and how strongly pairs of variables are related.  Several different correlation techniques may be utilized, but like all statistical techniques, correlation is only appropriate for certain kinds of data.  Correlation works for data in which numbers are quantities of some type and cannot be utilized for categorical data.  Another problem that arises with correlations is the analysis with rating scales (0 to 1; 2 to 4; 5 to 7, etc.).  While these scales represent quantity, their meaning in not precise.  Most statisticians avoid correlations with rating scales because the technique assumes the differences between numbers are exactly equal.

The summation of all of the available diet/survey data from nurses studies indicates a wide variety of slight positive (more incidence) and negative (protective) correlations with selective diseases without providing any possible mechanism of such contradictory action (positive and negative).  Furthermore, the studies do not take into account the highly nutritious nature of milk and dairy products when compared to other foods.  Thus, the correlation data is only a gross indicator that suggests that milk and other dairy products have something different than other foods.  It is logical to suspect nutritional quality.