The above article was published by The New York Times recently about the correlation between low vitamin D levels and increase in risk of Type 1 diabetes. The aim was to prove the hypothesis that a deficiency of vitamin D results in Type 1 diabetes, specifically in the tested active-military personnel.
The study was not a random sample, as every participant was a military member and selected based on vitamin D levels, an assortment of low and high. Between 2002 and 2008, the 'nested case-control' study of 1000 subjects' blood samples were matched and analyzed. The following image explains and also verifies the reliability of the testing, as the serum reading dates were accurate across the board for all participants' samples (time-wise).
The results support the researchers' hypotheses though it cannot prove causation of diabetes by vitamin D deficiency. The results are measured and presented in numbers representing value to the discovered phenomena, giving the [high and low] number relativity to one another (i.e. 17-23 nanograms vs. 40+).