This page documents, in full, how VoteLog turns official government records into the scores and money figures shown across the site. The complete methodology covers fourteen sections; each is summarized below in the same order it appears on the page.
VoteLog is an independent public-accountability project. It records what members of Congress officially did: bills sponsored, cosponsored, and voted on, and campaign money reported to the FEC. It does not claim to know why. Relationships shown between money and votes are correlational, not causal.
The core of the database is legislation that directly addresses Israel, Palestine, or the conflict: aid and arms, sanctions, diplomacy and recognition, humanitarian policy, and commemorative measures. Three connected categories are deliberately included because they function as Israel/Palestine policy in practice: Iran legislation aimed at Iran's nuclear program or external threat posture, measures targeting the International Criminal Court in response to its Israel-related cases, and antisemitism legislation that incorporates Israel-related speech enforcement.
Bills are found by searching the official Congress.gov API for Israel/Palestine-related keywords (Israel, Palestine, Gaza, Hamas, and others). The net is deliberately wide, so every keyword match is reviewed; off-topic matches (for example, tribal water-rights settlements matching “settlement”) are excluded and listed publicly on the excluded-bills audit trail rather than silently dropped.
Every tracked bill gets one of four labels: pro-Israel, pro-Palestine, neutral, or bundled (packages such as NDAAs and emergency supplementals that mix Israel-related provisions with unrelated ones). Labels follow the published classification rules below, with a curated override list for edge cases; every label change is recorded in the public changelog.
The text decides: labels are assigned from what a bill's operative text does, and identical or companion text always carries the same label across chambers and congresses. Antisemitism bills are neutral by default and classified pro-Israel only when their text mandates the IHRA working definition, elevates the Special Envoy office, or builds enforcement machinery aimed at Israel-related speech; pure hate-crime, synagogue-security, Holocaust-education, and remembrance bills stay neutral. Commemorative resolutions honoring a state's officials or symbols carry that side's label, while resolutions honoring private individuals stay neutral unless their text does national-question work. Iran bills targeting the nuclear program, the IRGC, or Iran's proxy network are pro-Israel; Iran measures with no Israel nexus are neutral. Omnibus vehicles are always labeled bundled, never pro-Israel or pro-Palestine.
The most consequential editorial call: bills that adopt or mandate the IHRA working definition of antisemitism are classified pro-Israel, because seven of the definition's eleven contemporary examples concern criticism of the State of Israel, and its policy effect is to treat a defined set of Israel-related criticisms as discriminatory rather than political. Civil-liberties groups and the definition's own lead drafter have opposed codifying it into law. VoteLog makes this call explicitly and consistently rather than burying these bills under a civil-rights heading; the full reasoning is documented on this page.
A vote is scored by its effect, not just Yea or Nay. Voting Yea on a pro-Israel bill (or Nay on a pro-Palestine bill) counts as pro-Israel, and vice versa. Procedural votes such as motions to table flip the meaning accordingly. Present and Not-Voting are excluded, as are bills with no clear lean; bundled bills score zero.
Each member's weighted score adds 3 points for sponsoring a bill, 3 for authoring a classified amendment, 2 for cosponsoring, and 1 for a supportive roll-call vote (on a bill or an amendment), signed by the measure's classification. The net score shown on a member's profile is the difference between their pro-Israel and pro-Palestine totals.
A separate persuadability score estimates which members might be movable on future votes, built from mixed voting records, public statements, and evidence volume, and gated by a confidence threshold so thin records are not over-read.
Money figures come from official FEC records via a curated registry of Israel/Palestine-aligned committees, tracked in tiers (including a separate liberal pro-Israel tier for the J Street family). Three channels are tracked: direct PAC contributions, independent expenditures for and against candidates, and bundled individual donations earmarked through conduits. Money spent against a member is always shown separately and never summed into what they raised. All totals are documented lower bounds.
A large share of pro-Israel money moves as bundled individual donations rather than through PAC treasuries: donors giving maxed-out personal checks through an organization's conduit. These appear on FEC Schedule A as earmarked individual contributions and are recovered from conduit records, kept in their own category, and never double-counted with PAC money.
Each member profile adds context beyond the roll-call record: the member's own press releases (filtered by the same Israel/Palestine topic rules used for bill discovery), aggregated news coverage, and public video appearances.
All data comes from official sources: Congress.gov for legislation and votes, the FEC (openFEC API and bulk filings) for campaign finance, and the US Census for district context. The production site is served as a static snapshot; data refreshes when the site is republished, and each page shows when its data was last updated.
State and local legislation, closed-door committee work, classified briefings, and whip pressure are out of scope: VoteLog only shows what ends up in an official filing or on the public record, and its primary-source pipeline covers the federal Congress.
Anyone can flag an error via the contact form. Corrections are applied to the data, not annotated around it, and classification changes appear in the changelog.