Methodology

Bridge Grades: Full Methodology

Bridge Grades scores members of Congress on how collaboratively or divisively they govern compared to their peers.

We use non-partisan 3rd party data sources to sort legislative “bridgers” from “dividers”.

• Bridgers build win-win consensus solutions through collaboration and coalitions for the benefit of cross-partisan interests

• Dividers pursue zero-sum game governance, engage in personal attacks, and demonstrate predictably partisan legislative records

Our systems compile multiple trusted 3rd-party data sources into a blended 0-100 index (the “Bridge Score”). We use data that measures what a person does (legislative record: authoring and sponsoring cross-partisan bills) and what a person says (rhetoric: public statements, speeches, social media).

The Bridge Grade grading rubric rewards:

Consensus Solutions

• Introduces bills (as a “sponsor”) in collaboration with cross-party co-sponsors
• Introduces bills (as a “co-sponsor”) in collaboration with a cross-party sponsor

Coalition Building

• Membership in cross-partisan Problem Solvers Caucus (House only)
• Joins previously introduced cross-partisan bills as a co-sponsor

Civil Discourse

• Makes fewer public personal attacks (Twitter/X, releases, floor statements) than their peers
• Makes more bipartisan public statements than their peers

Courage

• Engages in bridging behaviors even when representing a hard-leaning voter district
• Engages in bridging behaviors even with ideological positions far from the center

How we calculate Bridge Grades: The recipe

Think of the process of calculating Bridge Grades like following a recipe. The exact recipe for our process is detailed here. Below is the shorter version.

• Step 1: (collect refreshed data inputs)
  First, our systems proactively collect up-to-date data from each of our sources: govinfo.gov, Polarization Research Lab’s America’s Political Pulse, Cook Political PVI, Voteview, and house.gov.
• Step 2: (legislative record: sponsoring cross-partisan bills) [Source: govinfo.gov]
  We reward those who introduce (sponsor) bills with cross-partisan co-sponsors more often than their peers. We tally counts, make adjustments for bill type, normalize to a 100-point scale, and apply a 3.0x weight.
• Step 3: (legislative record: co-sponsoring cross-partisan bills) [Source: govinfo.gov]
  We reward those who co-sponsor bills that are introduced by the opposite party. We tally counts, adjust for timings, adjust for bill types, normalize to a 100-point scale, and apply a 2.0x weight.
• Step 4: (rhetoric analysis: bipartisan statements) [Source: America’s Political Pulse]
  We use tallies from America’s Political Pulse that to reward those who make bipartisan statements (“Collaboration and finding common ground across party lines”) more often than their peers. We weigh both the total counts, as well as the “share of voice” as some legislators simply make a higher volume of comments than others. We normalize each to a 100-point scale, and apply 1.0x weight for each.
• Step 5: (rhetoric analysis: personal attacks) [Source: America’s Political Pulse]
  We use tallies from America’s Political Pulse to punish those who engage in personal attacks (“Attacking or disrespecting a person or party”) more often than their peers. Again, we weigh both the total counts, as well as the “share of voice” as some legislators simply make a higher volume of comments than others. We normalize each to a 100-point scale, and apply 1.0x weight for each.
• Step 6: (extra credit bonuses - details here)
  We add bonuses for coalition building and courageous bridging behaviors (extra effort means extra credit):
  - Coalition building bonus for House members of the Problem Solvers Caucus. [Source: house.gov]
  - Courage bonus for bridging while representing hard-leaning partisan districts. [Source: Cook Political PVI]
  - Courage bonus for bridging while holding ideological positions far distances from the center. [Source: Voteview DW-NOMINATE]
• Step 7: We add up all of the totals, and normalize the results onto a 100-point scale which becomes their Bridge Score.
• Final Step: Finally, we translate the scores into corresponding Bridge Grades. All scores above the midpoint earn As and Bs, and all below the midpoint earn Cs and Fs. One standard deviation above the mean earns an A and one standard deviation below the mean earns an F.

Core principles of the Bridge Grades grading rubric

• Bridge Grades are assessed uniquely for each term served in Congress. Just because a Congress person deserved a “C” last term, doesn’t mean they can’t earn an “A” this term.
• Scores were reset to zero on January 3, 2025 at the beginning of the 119th Congress. The will reset once again in January 2027, for the 120th Congress.
• Data collected on legislative record and rhetoric is based on behaviors which have occurred only during the term of service.
• Scores and rankings are relative to their peers within each chaber. There will always be a top half and a bottom half of each class.
• What you do (legislative record) counts more than what you say (rhetoric analysis). But, both matter.
• Bridging difficulty is not equal. Extra credit for extra effort.
• No single data source tells the whole story. By combining several sources we can tune in the signal from the noise.
• Each chamber is scored only in relation to itself (Senate scale is different from House).
• Remove legislators who are no longer active, and don’t grade legislators that are too new to be fairly compared to their peers.

Notes

• Intersted in the data?
• Easter egg scatter plot chart here.