Overcoming the scoring problem with SF-36 component summary scores: a method that works

Abstract

In this thesis I will discuss the shortcomings of the statistical methods used to derive scoring coefficients for the physical and mental health component summary scores of the Medical Outcomes Study SF-36 and SF-12 health status scales. I will propose an alternative statistical method for generating scoring coefficients for these scales, and produce scoring coefficients for both version 1 and version 2 of the SF-36 and SF-12. I will then demonstrate the superior measurement properties of summary scores generated using my method compared to the proprietary scoring, and discuss the limitations of the SF-36 author’s contention regarding international comparisons using this instrument. The study is articulated through several international peer reviewed publications, which provide a progressive story of the body of research. The papers themselves follow on from a wider discussion of the failure of the methods recommended for scoring the physical and mental component summary scores of the SF-36 and SF-12 and provide the argument for alternative scoring methods.