STTPN

Political-Leaning Methodology

STTPN describes the general political leaning of whole news publications, never individual articles. A publication's label is shown next to its articles in our Related Coverage lists so readers have context when comparing coverage.

Political leaning is not accuracy. A label says nothing about whether a particular article is true, and “Center” does not mean objective, impartial in every story or more reliable.

The five classifications

  • Left: strong, consistently observable left-leaning editorial tendencies.
  • Lean Left: noticeable but less pronounced or less consistent left-leaning tendencies.
  • Center: no consistently observable left- or right-leaning tendency in the evidence reviewed.
  • Lean Right: noticeable but less pronounced or less consistent right-leaning tendencies.
  • Right: strong, consistently observable right-leaning editorial tendencies.

What “Unrated” means

Unrated is not a sixth viewpoint and never means Center. It means STTPN has not completed a human-approved review of that publication, or the evidence was too limited or conflicting to support a defensible label.

Four review criteria

  1. Political framing: recurring language, emphasis, context and caveats across similar political stories.
  2. Story selection and prominence: which events are featured, omitted or prominently placed over a comparable period, using dated homepage and archive observations as supporting context.
  3. Source selection: the recurring mix of experts, officials, institutions and affected people, and how contrasting evidence is presented.
  4. Editorial positions: stated editorial opinions and endorsements, kept distinct from straight news reporting.

The criteria are weighed together and explained in words. We don't use numeric bias scores, percentages or confidence figures, and we don't base a label on audience stereotypes, ownership, a single columnist or another organization's ratings.

How a review works

  • A first review examines about 20 articles from the 90 days before the review, across at least three political topics, with news and opinion pieces recorded separately. Dated homepage or archive observations are used as supplementary context.
  • Twenty articles is a practical starting sample, not a statistically validated one. When the sample is thin, one-sided or contradictory, we expand it, or leave the publication Unrated.
  • Counterexamples are recorded alongside supporting evidence.
  • AI tools may help organize research and surface contrary examples. The editor independently checks key cited material, and only an explicit human approval can publish or change a label.
  • Rated publications have a public profile with the label, the date of the last completed review, a short evidence-based explanation and selected references.

Reassessment and challenges

Approved labels are reassessed about once a year, or sooner after a significant editorial change or a credible challenge. While a reassessment is under way, the last approved label and its true review date stay visible until a new decision is approved, or the label is withdrawn to Unrated if it is no longer defensible. Every decision is kept in a dated private history.

Anyone can challenge a classification through our contact page, with supporting links. No account is needed. Challenges are reviewed on their evidence; there is no public voting.

Limits

These are qualitative editorial judgments made by a small, founder-run publication. Human judgment and sampling limits can't be eliminated. STTPN controls its own methodology and labels; advertisers, sponsors and donors have no role in them.

Publications in our directory