How we score

Our goal is to help you form your conscience, not to tell you how to vote. Here is exactly how every score on this site is made, so you can check our work.

Our principles

• Non-partisan. We score votes and positions, never parties. Every candidate is measured with the same rubric.
• Transparent. Every score links to the specific votes and statements behind it.
• Grounded in teaching. Each category rests on the Catechism of the Catholic Church, the Compendium of the Social Doctrine of the Church, and the U.S. bishops' Forming Consciences for Faithful Citizenship.
• Humble. This is not an official Church document. Scores are a starting point for prayer and study, not a substitute for your own judgment.

The eight categories

Not all issues weigh the same

The Church teaches that some acts, such as abortion and euthanasia, are always wrong. These "intrinsically evil" acts carry special weight. Other issues, like how best to fight poverty, call for prudential judgment, and Catholics may disagree in good faith. We reflect this in two ways:

1. Within a category, each vote or statement is weighted by the gravity of the issue:
   • Intrinsic evil at stake: counts 3×
   • Serious moral matter: counts 2×
   • Prudential judgment: counts 1×
2. Across categories, Dignity of Life has the largest default weight, in line with the bishops' teaching that the threat of abortion remains a preeminent priority.

The bishops also warn against using this distinction to dismiss other serious threats to human life and dignity, such as racism, the death penalty, and the treatment of migrants. That is why every category counts.

How a candidate's score is calculated

1. Gather evidence. Roll-call votes, bills sponsored or co-sponsored, and bills signed or vetoed make up the voting record. Campaign platforms, questionnaires, debates, and public statements make up the platform.
2. Rate each item. We read the full text of each bill and list every part that touches a moral issue, since bills are often packed with several unrelated issues. For each part, we decide which vote is most consistent with Church teaching and explain why. A bill is scored on its worst part: a Yes vote on a bill that includes abortion funding counts against Church teaching, even if the rest of the bill is good. A vote that matches counts +1, and a vote against counts −1. Missed votes are shown but not counted. Statements are rated from −1 to +1.
3. Weight each item. Weight = issue gravity × type of evidence. Record weights: roll-call 1×, cosponsor 1×, sponsor 1.25×, executive 1.5×. Platform weights: platform 1×, questionnaire 1×, debate 0.75×, public statement 0.75×.
4. Score each category. The weighted average rating is mapped onto 0 to 100, where 50 is neutral. The record and platform are scored separately, then combined: 70% record and 30% platform. If a candidate has only one of the two, that one is used alone.
5. Overall score. A weighted average of the category scores. Categories with no evidence are left out, and the other weights are scaled up to fill the gap.
6. Confidence. High confidence needs a voting record and at least 8 counted items across 5 or more categories. Medium needs at least 4 items across 3 categories. Candidates with no voting record are always marked "Based on platform only, lower confidence."

Ballot measures

We read the full text of each measure and rate every part that touches a moral issue, from −1 to +1, citing the teaching that applies. Like a bill, a measure is scored on its worst part, and every part is shown. We also show any public position from the state Catholic Conference or local bishops.

When Church teaching doesn't settle a question

Many policy questions are left to the conscience of Catholics, who can reasonably disagree. We don't guess. When Church teaching doesn't clearly favor one side, the vote or statement is listed on the candidate's page with the label "Left to conscience" and not counted in any score.

We give our reviewers and the AI this guidance on issues where it matters most:

AI evaluated and human reviewed

Magisterium AI, an AI trained on Church documents, rates each vote, statement, and measure. Every rating must cite Scripture, the Catechism, and another Church document, or it isn't published. These ratings appear right away with an AI evaluated label. When a person on our team checks a rating, it changes to Human reviewed. Plain-language explanations are written with AI and labeled the same way.

Judges

We do not score judges. Judges must rule on the law and facts in front of them, and ethics rules limit what they may say about issues. We link to their records so you can review them yourself.

Score adjustments

Scores are calculated automatically from the evidence. A reviewer may adjust a category score when the formula misses important context. Every adjustment is shown on the page with the reason, and every change is logged.

Set your own weights

Data sources

• Federal votes: Congress.gov and GovTrack
• State legislatures: Open States
• Ballots and polling places: Google Civic Information API and Ballotpedia
• Candidate positions: Vote Smart, campaign websites, and candidate questionnaires
• Church positions: the USCCB and state Catholic Conferences