Justifying Background Assumptions in the Internalist Frequentist Framework

Develop a systematic account that justifies the background assumptions defining the parameter space O used to evaluate inference methods in frequentist statistics within an agent’s context of inquiry, thereby explaining from a first-person internalist perspective how these assumptions can be warranted so that significance levels, power, and other reliability-based criteria assessed relative to O are themselves epistemically justified.

Background

The paper argues that frequentist statistics can be interpreted as an internalist form of reliabilism by evaluating inference methods over a parameter space O that represents the scenarios compatible with the background assumptions taken for granted in a given context of inquiry.

While this interpretation clarifies how reliability standards (e.g., significance level, power) can be assessed from the first-person perspective, it leaves open how those background assumptions themselves are justified within that same context. The author points to contextualist ideas (e.g., Annis 1978) as a possible direction, but the justification problem is explicitly left unresolved.