News Analytics

FastAPI-backed source, lens, workflow, and event diagnostics

The news surface reads from the backend contract and exposes exploratory comparisons, data quality checks, workflow health, and same-event analysis.

Index

Pages

All routes below stay inside the same shared analytics surface.

News Source Effects

Live FastAPI data

What this page does

Compare source effect sizes by lens with pooled, within-topic, and within-tag views.

Data Mode

Current Snapshot

Scope

Analysis Mode

Switch between pooled, within-topic, and within-tag source comparisons without changing the underlying data mode.

Pooled (topic-confounded)Within-topic Within-tag

Controls

Effect Filters

Control lens count, FDR threshold, and the detailed lens used for the source-mean comparison chart.

Reliability

Source Reliability Assessment

Heuristic readout for whether source-difference claims have enough support. This is not a quality score; it summarizes sample size, separability, FDR exposure, and controlled-slice availability.

Methodheuristic-v1

Pooled Labeltopic-confounded

Pooled Tierhigh

Pooled Score1.00

Event-Control Tiermoderate

Event-Control Score0.45

Topics OK1 / 2

Tags OK1 / 2

Pooled Flags0

Articles	729
Sources	11
Lenses	12
Classification accuracy	24.3%
Baseline accuracy	10.4%
Classification lift	13.9%
Best q-value	0.0476
FDR-significant lenses	12 / 12

Topic Slices

Topic Reliability

Per-topic support levels for within-topic source comparisons.

Topic	Status	Tier	Score	Articles	Sources	Lift	Best q	Flags	Reason
Untagged	ok	high	1.00	685	10	15.8%	0.0476	none
world	unavailable	unavailable	0.00	44	1	0.0%	0.0000	insufficient_source_count, source_differentiation_unavailable, source_lens_effects_unavailable	Source differentiation and lens effects are both unavailable for this slice.

Tag Slices

Tag Reliability

Per-tag support levels for within-tag source comparisons.

Tag	Status	Tier	Score	Articles	Sources	Lift	Best q	Flags	Reason
Unknown	unavailable	unavailable	n/a	0	0	0.0%	0.0000	none
Unknown	unavailable	unavailable	n/a	0	0	0.0%	0.0000	none

Lens Effects

Pooled Source Effects

Lens-level source differences ranked by effect size, with FDR-aware filtering and source mean context.

Label: topic-confounded

Rows shown: 10 | q-threshold: All | Reliability: high (1.00)

Rows10

Permutations20

Multiple Testingbenjamini-hochberg

Tests12

Best q-value (FDR)0.0476

Best raw p-value0.0476

Lens	n	Sources	F	eta²	p_perm_raw	p_perm_fdr	Source Gap	Top / Bottom
Objectivity vs Opinion Lens	729	11	24.530	0.2546	0.0476	0.0476	26.83	PBS NewsHour / CBS News
Entity-Level Sentiment Lens	729	11	24.489	0.2543	0.0476	0.0476	33.33	The Guardian / CBS News
Sentiment Clarity Lens	729	11	21.730	0.2323	0.0476	0.0476	28.55	The Guardian / CBS News
Credibility Lens	729	11	17.149	0.1928	0.0476	0.0476	27.41	The Guardian / CBS News
Linguistic Quality Lens	729	11	15.773	0.1801	0.0476	0.0476	24.36	BBC News / CBS News
Agency and Voice Lens	729	11	14.236	0.1655	0.0476	0.0476	19.14	The Guardian / CBS News
Authority and Source Positioning Lens	729	11	13.690	0.1601	0.0476	0.0476	28.75	DW / CBS News
Emotional Intensity Lens	729	11	12.476	0.1480	0.0476	0.0476	34.61	Fox News / CBS News
Epistemic Modality and Certainty Lens	729	11	12.271	0.1460	0.0476	0.0476	28.52	DW / CBS News
Hegemonic Common-Sense Lens	729	11	9.270	0.1143	0.0476	0.0476	26.57	PBS NewsHour / CBS News

Reference Table

Topic Slice Overview

Quick scan of per-topic availability and the strongest lens effect present in each topic slice.

Topic	Status	Lens Rows	Best Lens	Best eta²
Untagged	ok	12	Objectivity vs Opinion Lens	0.250
world	unavailable	0	n/a	n/a

Reference Table

Tag Slice Overview

Quick scan of per-tag availability and the strongest lens effect present in each tag slice.

Tag	Status	Lens Rows	Best Lens	Best eta²
Unknown	unavailable	0	n/a	n/a
Unknown	unavailable	0	n/a	n/a