Absolutely Zero Evidence

TL;DR

This paper critically examines the concept of measuring scientific evidence, arguing that common statistics do not conform to proper measurement scales and exploring the idea of an absolute zero point for evidence.

Contribution

It introduces a representational measurement perspective to evaluate evidence statistics and challenges the notion of an absolute zero evidence point.

Findings

Common evidence statistics do not conform to legitimate measurement scales

The concept of absolute zero evidence differs from intuitive expectations

A new perspective on evidence measurement scales is proposed

Abstract

Statistical analysis is often used to evaluate the evidence for or against scientific hypotheses, and various statistics (e.g., p-values, likelihood ratios, Bayes factors) are interpreted as measures of evidence strength. Here I consider evidence measurement from the point of view of representational measurement theory, and argue that familiar evidence statistics do not conform to any legitimate measurement scale type. I then consider the notion of an absolute scale for evidence measurement, in a sense to be defined, focusing particularly on the notion of absolute 0 evidence, which turns out to be something other than what one might have expected.