Skip to main content
CIV.IQ

Methodology

How CIV.IQ handles campaign finance data, what the research says, where our data comes from, and what we do not claim.

How CIV.IQ presents campaign finance data

CIV.IQ shows campaign donations alongside voting records. We do this so you can see the full picture. We do not claim that donations caused any vote.

Party membership and personal beliefs predict how a legislator votes far better than campaign money. Ansolabehere, de Figueiredo, and Snyder (2003) found that contributions explain a “minuscule fraction” of voting after you account for party and ideology.1

The strongest finding about campaign money is about access, not votes. Kalla and Broockman (2016) found that donors get three to four times more meetings with elected officials than non-donors. Money opens doors. It does not necessarily change minds.2

There is also a chicken-and-egg problem. Donors tend to give money to legislators who already agree with them. This makes it look like money changed a vote, when the legislator was already going to vote that way. Researchers call this strategic giving, and it is the biggest challenge in studying campaign finance.

CIV.IQ presents donations alongside votes for transparency. We do not claim that donations caused votes.

Five ways campaign money relates to legislation

Researchers have found five ways campaign money connects to what legislators do. We rank them by how strong the evidence is, from strongest to weakest.

  1. 01

    Access

    STRONGEST

    Donors get more meetings with legislators. Kalla and Broockman (2016) ran a controlled experiment and found donors were three to four times more likely to get a meeting.2 This is the best-supported finding in campaign finance research.

  2. 02

    Committee work

    STRONG

    Donations match up with how hard a member works on bills in committee that affect donors. Hall and Wayman (1990) found this pattern.3 Hojnacki and Kimball (2001) found the same thing.4 Money may affect effort in committee, not how someone votes on the floor.

  3. 03

    Agenda-setting

    MODERATE

    Lobbying groups spend money to get certain issues in front of Congress. Furnas et al. (2023) studied how lobbying shapes which issues get attention.5 McKay (2018) looked at how fundraising connects to legislative outcomes.6 This is harder to measure than votes because you cannot see what did not make it onto the agenda.

  4. 04

    Strategic giving

    THE CATCH

    Donors give to legislators who already agree with them. When a legislator votes the way a donor wants, you can’t always tell why. Did the money change the vote? Or did the donor just pick someone who already agreed? This chicken-and-egg problem is the biggest challenge in campaign finance research.

  5. 05

    Direct vote-buying

    WEAKEST

    The idea that a donation flips a specific vote has very little support. Most studies find no real effect after accounting for party and ideology.1 A few narrow exceptions exist in specific industries, but the overall evidence is weak.

Data sources and methodology

CIV.IQ pulls data from 29 official government and public data sources. We never make up data or use estimates. When a source is unavailable, we tell you and show the date of the last available data.

CORE SOURCES

  • —Congress.gov API — Bills, members, committees, votes, hearings (Library of Congress)
  • —FEC API — Campaign contributions, expenditures, PAC filings (Federal Election Commission)
  • —Senate LDA — Lobbying disclosure filings (U.S. Senate)
  • —Census Geocoder — Address-to-district lookup, demographics (U.S. Census Bureau)

25 MORE SOURCES BY CATEGORY

Regulations and policy(2)
Financial disclosure(3)
Spending and economy(4)
Environment, energy, and safety(6)
  • —EPA ECHO — Environmental enforcement actions, facility violations
  • —OSHA Enforcement — Workplace safety inspections, violations
  • —EIA API — State energy production and consumption
  • —NOAA Climate API — Climate normals, severe weather events
  • —NHTSA API — Vehicle recalls, safety complaints
  • —FEMA OpenAPI — Disaster declarations, assistance data
Health, education, and housing(4)
Consumer protection, justice, and crime(4)
State government and biographical(2)
  • —Open States API — State legislators, state bills, state votes
  • —Wikidata — Biographical data, state executives, judiciary

How we analyze data

Every analysis starts with math, not AI. The numbers come first.

  • —Every insight has a confidence score from 0 to 1. We hide anything below 0.6.
  • —We require minimum sample sizes before showing results: at least 10 votes per sector, 4 quarters for trends, and 3 trades for stock analysis.
  • —We always compare a legislator to their peers in the same chamber, party, or committee. This shows patterns in context, not in isolation.
  • —AI-generated summaries use plain language at an 8th-grade reading level or below.
  • —We match records across sources by linking names, IDs, and organizations across FEC filings, Congress.gov, Senate lobbying filings, and SEC records. This lets you see the connections between money, votes, and lobbying in one place.

What CIV.IQ does not claim

We do not claim donations caused votes.

We show donations and votes together so you can see the full picture. We use words like “received,” “associated with,” and “correlated with.” We do not use “influenced,” “caused,” or “resulted in” when describing money and votes.

We do not claim lobbying is improper.

Lobbying is protected by the First Amendment. We show lobbying filings because they are public records, not because lobbying is wrong.

We do not take sides.

CIV.IQ presents data without opinion. We do not rate legislators as good or bad. We do not tell you how to vote. We show public records and let you draw your own conclusions.

We do not cherry-pick.

We do not select time periods or comparisons to make any legislator look better or worse. Every comparison follows the same rules for all members.

Data freshness and monitoring

CIV.IQ continuously monitors all 29 data sources for availability and freshness. Every analysis displays a “Data through” timestamp showing when the underlying data was last updated.

When a data source becomes unavailable, CIV.IQ continues to display the last available data with its date. We never silently show stale data as if it were current. If a source has been down long enough that its data is no longer meaningful, the affected section shows a clear “Data unavailable” message instead.

Core data sources (Congress.gov, FEC, Senate lobbying filings, Census Geocoder) are checked on every health probe. Supplementary sources are checked on a rotating schedule to stay within rate limits. Source availability and response times are logged for operational visibility.

References

  1. Ansolabehere, S., de Figueiredo, J. M., & Snyder, J. M. (2003). “Why Is There So Little Money in U.S. Politics?.” Journal of Economic Perspectives, 17(1), 105-130.
  2. Kalla, J. L. & Broockman, D. E. (2016). “Campaign Contributions Facilitate Access to Congressional Officials: A Randomized Field Experiment.” American Journal of Political Science, 60(3), 545-558.
  3. Hall, R. L. & Wayman, F. W. (1990). “Buying Time: Moneyed Interests and the Mobilization of Bias in Congressional Committees.” American Political Science Review, 84(3), 797-820.
  4. Hojnacki, M. & Kimball, D. C. (2001). “PAC Contributions and Lobbying Contacts in Congressional Committees.” Political Research Quarterly, 54(1), 161-180.
  5. Furnas, A. C., LaPira, T. M., Hertel-Fernandez, A., Drutman, L., & Kosar, K. R. (2023). “More than Mere Access: An Experiment on Moneyed Interests, Information Provision, and Legislative Action in Congress.” Political Research Quarterly, 76(1), 348-364.
  6. McKay, A. (2018). “Fundraising for Favors? Linking Lobbyist-Bundled Campaign Contributions to Legislative Outcomes.” Political Research Quarterly, 71(2), 379-391.

This page was last updated April 2026. CIV.IQ is open-source civic infrastructure. Our methods improve as research advances. For questions about our methodology, see our source code.