One role of biostatisticians is to create and apply statistical methods that help us measure stuff related to public health. The better the method, the better the measurement. The better or more precise the measurement, the more confident we can be that we understand disease patterns or health care outcomes, for example. I, for one, want policymakers to decide on changes to Medicare based on the best possible data measurements.
One aspect of measurement is using the correct tool. For example, to measure the amount of clean water produced by a water filter, we have to use a cup or something to capture the volume of water. It would be really difficult to count the number of water molecules. As a comparison, we can count the number of people in a community who are sick from contaminated water. The result is a different type of measurement.
Here is a biostatistics research article demonstrating a new use of a particular statistical method for measuring a different kind of data.
Manuguerra, Maurizio and Heller, Gillian Z. (2010) “Ordinal Regression Models for Continuous Scales,” The International Journal of Biostatistics: Vol. 6: Iss. 1, Article 14.
Source: COBRA (Collection of Biostatistics Research Archive), under Categorical Data Analysis
