Guess et al. (2023) null findings do not rule out algorithmic effects below the study's minimum detectable effect size or effects on non-volunteer subpopulations with different baseline polarization levels.

[REBUT] Guess et al. (2023) conducted a large field experiment with consenting volunteers, introducing volunteer bias that may reduce or mask treatment effects relative to the full population. Absence-of-evidence arguments require specification of the minimum detectable effect size and acknowledgment that non-detection does not prove absence, particularly when volunteer bias may suppress effects. The framing's ≥10% bar requires evidence that effects do not reach that threshold; null results in volunteer samples do not establish this across the full population. Therefore, Guess et al. (2023) null findings do not rule out algorithmic effects below the study's minimum detectable effect size or effects on non-volunteer subpopulations with different baseline polarization levels. (Warrant: Null results in volunteer samples cannot establish that effects are below the framing's ≥10% threshold across the full population without explicit analysis of volunteer bias and generalization.)

• Absence-of-evidence arguments require specification of the minimum detectable effect size and acknowledgment that non-detection does not prove absence, particularly when volunteer bias may suppress effects.
• Guess et al. (2023) conducted a large field experiment with consenting volunteers, introducing volunteer bias that may reduce or mask treatment effects relative to the full population.
• The framing's ≥10% bar requires evidence that effects do not reach that threshold; null results in volunteer samples do not establish this across the full population.

These are challenges this argument’s reasoning pattern must still withstand. Answering them on Isonomia strengthens the argument.

• Is the absence of positive evidence strong enough to justify concluding ¬A, or only to justify withholding belief in A?Open
• Is the absence of evidence due to absence of investigation rather than to A's being false?Open
• Could disconfirming or null findings have been suppressed, unpublished, or systematically under-reported (file-drawer / publication bias)?Open
• Has the investigative regime actually been adequate (well-funded, well-powered, well-designed) to detect E if A were true?Open
• Could A be true but produce only a weak signal that escapes detection at the prevailing statistical thresholds?Open

No one has cited this argument yet.

No arguments cite this one yet — no one has built on or contested it. That is an absence of engagement, not a finding of soundness. Build on or contest it on Isonomia to change that.