Skip to content

File Drawer Problem

Recognize that studies with null results disproportionately go unpublished while significant ones enter the literature, so any synthesis treating the published record as the full population of conducted research systematically overestimates effect sizes toward the filter.

Core Idea

The file drawer problem, named by Rosenthal in 1979, is the meta-research phenomenon in which studies with null or unsurprising results disproportionately go unpublished while significant ones enter the literature, making the published record a biased sample of the conducted research. Its mechanism has three parts: a population of conducted studies, a publication filter correlated with effect size, and a downstream synthesis that mistakes the filtered record for the full population.

Scope of Application

The file drawer problem lives across the subfields of meta-research, evidence synthesis, and scientific reform — the substrate of science as a publication-mediated knowledge system.

  • Meta-analysis methodology — file-drawer correction as a routine step: funnel-plot asymmetry, trim-and-fill, p-curve, Egger regression, selection models.
  • Clinical-trial registries — registration before data collection instituted to make the conducted population visible regardless of outcome.
  • The replication crisis — the file drawer as one proximate mechanism behind over-stated, non-replicating effects.
  • Drug-trial regulation — mandates that sponsors report all trial results, targeting selective publication of favourable trials.
  • Pre-registration and registered reports — editorial models that decide publication before results exist, removing the filter for accepted work.

Clarity

Naming the file drawer problem moves the locus of suspicion outside the individual study to the record itself, treated as a measurement that can be biased independently of the quality of any study in it. It converts a vague unease — "can I trust this literature?" — into a sharp question: is the visible record a fair sample of the conducted research, or a filtered one?

Manages Complexity

The concept replaces an unbounded study-by-study audit with a three-part generative model and one latent quantity: the hidden mass of filed-away results and its zero-ward pull. The bias reads off a low-dimensional summary the record already exposes — effect-versus-precision asymmetry — and the remedy forks on timing relative to the filter.

Abstract Reasoning

The concept licenses a diagnostic move from a fingerprint the record exposes (funnel asymmetry implies a removed mass of nulls), an interventionist move forked on timing (eliminate the filter prospectively versus estimate and correct after the fact), and boundary-drawing on the locus of the defect, including a signed prediction that the pull is always toward zero.

Knowledge Transfer

Within meta-research the file drawer transfers as mechanism and is the connective tissue of a cluster of publication pathologies. Beyond that substrate the deeper pattern — a selection filter on visibility biases the visible record — is carried by parents selection_bias, survivorship_bias, and selection_effect, of which the file drawer is the publication-pipeline case. Its named machinery and pipeline framing stay home.

Relationships to Other Abstractions

Local relationship map for File Drawer ProblemParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.File Drawer ProblemDOMAINDomain-specific abstraction: Publication Bias — is a kind ofPublication BiasDOMAIN

Current abstraction File Drawer Problem Domain-specific

Parents (1) — more general patterns this builds on

  • File Drawer Problem is a kind of Publication Bias Domain-specific

    Drawer Problem is the whole-study null-suppression species of the broader results-dependent publication and reporting filter.

Hierarchy paths (12) — routes to 6 parentless roots

Neighborhood in Abstraction Space

File Drawer Problem sits in a moderately populated region (59th percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.

Family — Publication Bias & Research Artifacts (5 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-07-12