Quantifying information content in survey data by entropy

Abstract

We apply Shannon entropy as a measure of information content in survey data, and define information efficiency as the empirical entropy divided by the maximum attainable entropy. In a case study of the Norwegian Function Assessment Scale, entropy calculations show that the 5-point response version has higher information efficiency than the 4-point version. © 2010 by the authors; licensee Molecular Diversity Preservation International, Basel, Switzerland.