Readings › For the checks

Robust Evidence and Secure Evidence Claims

Staley, 2004. §1, §2, §3, §4.

Card published · 11 underlinings · 0 claims drawn · last read 2026-11-18
DOI
10.1086/423748
Shelf
For the checks
Concepts
Severity, Convergent paths

Why it is hereRobust results are what the programme counts on when independent runs or coders agree. Staley says what that agreement buys: not stronger evidence, but a more secure claim about the evidence.

The underlinings across the pages

466467468469470471472

pp. 466–472 · 11 marks

structure (red in Zotero) definition (green) key passage (yellow) book pages · pages not read are folded

1 Introduction

pp. 467–468

Philosophers have long held that evidence is better when independent tests converge, without saying why. Staley argues the case outside Bayesianism and separates two questions: how strongly data indicate a hypothesis, and how exposed the claim is to the failure of an auxiliary assumption. Robustness serves the second.

  1. Strength and security Strength is how strongly the data indicate the hypothesis; security is how far the evidence claim survives if an assumption fails.
  2. First- and second-order evidence Data can be evidence for a hypothesis, or evidence that other data are evidence of the strength claimed.
  3. Two uses of convergence A second, independent result can check that the first is as strong as claimed, or back it up so that the claim is not wrong altogether. Combining sources into one stronger claim does not protect against overstating its strength.

2 The Error-Statistical Theory of Evidence

pp. 468–469

Staley works within Mayo’s account: a result is evidence for a hypothesis only if the hypothesis passes a severe test with it, that is, only if a result fitting this well would be very improbable were the hypothesis false. Discrimination from the alternatives counts, fit does not, and the assessment is often informal.

  1. Discrimination, not fit Agreement with the hypothesis is not enough; the result must be one the alternatives would rarely produce.

3 First- and Second-Order Evidence

pp. 469–470

Since whether a fact is evidence is itself an empirical question, data can be evidence for a hypothesis or evidence that other data are evidence for it. The difference is one of role, not of kind: the same fact can play either part, depending on the hypothesis of interest.

4 Strength of Evidence and Security of Evidence Claims

pp. 470–471 · an extract with a card of its own

How strongly a result indicates a hypothesis is one question; how exposed the claim “this result is evidence for H” is to the failure of an auxiliary assumption is another. Robustness may not answer the first, but it can answer the second.

The extract: Strength of evidence and security of evidence claims →

Claims drawn from this card

Each claimA sentence from a source, kept verbatim, that the programme stands on, uses or has to answer to. shows its ENTIAThe notation this site uses on itself: what each piece is for, read through five layers. functionWhat a claim does for the programme: a conceptual tool, a working method, corrective evidence, a technical reference.; open + for its stanceHow the programme holds the claim: assumed, used, or contested., claim typeWhat kind of statement it is, so that the right test can be applied. Proposed until checked by hand. and relationsHow one claim bears on another, in the CiTO vocabulary: refines, extends, presupposes, confirms, contests..

No claims drawn yet.

Paraphrase CC BY-NC 4.0, Claudio Cammarano. Quoted passages belong to their author and are kept short; the text itself is not distributed. Notes from the Zotero library, updated weekly.