Skip to content
Sharif Amlani, PhD

Research · PhD, Political Science · UC Davis, 2023

Measured,
not guessed.

The rigor behind the AI work. A PhD from UC Davis, five peer-reviewed publications, 170+ citations, and research featured in The Economist, The New York Times and The Washington Post.

Fig. 01 — Donor ideology, 1980-2016 · Illustrative dataDemocratic donorsRepublican donors
Gap between party medians
0.83
CFscore units · Illustrative
Out-of-district share of House money
84% → 81%
1982 → 2016 · Roughly flat, so it holds still here
Source

Each point is a donor placed by ideology score. The real figure draws from the 13M-record DIME file behind the dissertation.

1980: Democratic and Republican donor medians sit 0.83 CFscore units apart. Illustrative data.
PhD
Political science
UC Davis · 2023
5
Peer-reviewed publications
PLOS ONE · APR · Pol. Behavior · ELJ · Elect. Studies
174
Citations
Google Scholar · Oct 2026
13M+
Campaign-finance records
18 election cycles
3
National outlets
The Economist · NYT · Washington Post

Most-cited research · Electoral Studies 73 (2021) · With Carlos Algara

America votes the party line.
Like it hasn't since the Civil War.

“Presidential and Senate elections are more partisan today than any period since the Civil War, while gubernatorial elections are as partisan today as they were during the late 1800s.”

Amlani & Algara, Electoral Studies (2021)
152 years
Of American elections
1868–2020 · 39 presidential
3,113
Counties in 2020
49 states + DC · Alaska not in the data
2,624
Unique dataset downloads
Harvard Dataverse · 7,383 unique views
89
Citations
Google Scholar · +15 more cite the dataset
APSR · AJPS
Where it is cited
And Political Behavior, SSQ, SPPQ
Fig. 03 — Every county, every presidential election, 1868–2020
Compare with
President
SENATE
Loading county returns…
2020
—
How closely county SENATE results track the presidential vote (r)

Drag the slider or press play to watch 152 years of county results.

1.0.50186819001932196419962020
Senate vs. presidentGovernor vs. presidentMore RepublicanMore Democratic
Each bright dot is a county with returns that year (modern centroid); the data hold 1,723 counties in 1868 and 3,113 in 2020. Faint marks: no race or not yet a county. Alaska is not in the data (its votes are not reported by county); Hawaii is in the data but off this map. Color = Democratic share of the two-party vote. r = correlation across counties between the down-ballot and presidential Democratic share, same year. Senators were chosen by state legislatures until the 17th Amendment (1913). Data: Amlani & Algara (2021), Harvard Dataverse.Read the paper ↗Get the data ↗

The proof,
peer-reviewed.

5 peer-reviewed publications170+ citationsGoogle Scholar ↗

Dissertation · Competing Principals · UC Davis, 2023

Do out-of-district donors distort representation between members and their constituents?

8 of every 10 dollars members of Congress receive originate outside their district.

Out of districtIn district
$8.7B
Raised in 2020
By congressional campaigns, the most in American history (OpenSecrets).
~200%
Growth, 2008-2020
Money raised by campaigns grew by almost 200 percent.
18
Election cycles
The largest number examined by any study of out-of-district donors.
13M+
Records
Campaign-finance records, scored with Bonica's CFscores.
01 · The argument

Two constituencies.

Members' reelection fortunes are no longer set only by the district. They also answer to a donor constituency: the broad network of out-of-district donors who finance congressional campaigns.

02 · The method

Redistricting as identification.

Redistricting is an as-if random reshuffling of donors and districts. It breaks fundraising ties, so the effect of money can be read without reverse causation. Ideology comes from donor CFscores, not the national donor base.

03 · The finding

Constituents lose.

Redistricting leads to greater reliance on out-of-district money and better representation for out-of-district donors relative to constituents: agency loss. It extends Mansbridge's "monetary surrogacy" theory.

Competing Principals: The Increasing Prominence of Out-of-District Contributors and their Effects on Representation in CongressCommittee · Scott MacKenzie, Erik Engstrom, Michael Rocca, Walter Stone

Data · Bonica's DIME · CFscores · 13M+ campaign-finance recordsRead the dissertation on eScholarship ↗

Original data.
Fielded, not borrowed.

2 national surveys · Summer 2021With Spencer Kiesel ↗

01 · UC Davis · Lucid

UC Davis COVID-19 Vaccine Hesitancy Survey Project

N = 4,300
Nationally representative U.S. adults · Opt-in online panel
What it measured
ARespondents’ own experience with COVID-19.
BTheir social network’s attitudes about COVID-19 and vaccines.
CA conjoint experiment testing pro-vaccination arguments.
Data embargoed · Ask for access →
02 · UC Davis · Lucid

UC Davis Partisan Affect Survey & Project

N = 2,300
Nationally representative U.S. adults · Opt-in online panel
What it measured
AParty identification.
BOne word describing members of their own party, and one for the opposing party.
CThe sentiment of that word, coded by the respondents themselves.
DA conjoint experiment measuring partisan dislike.
Data embargoed · Ask for access →

Two talks.
Slides included.

Conference presentations.pptx downloads

MPSA · 2021

Vote-By-Mail and 2020 Turnout

Midwest Political Science Association annual meetingAmlani & CollittDownload slides (.pptx)
PolNet · 2022

COVID-19 and Social Networks

Political Networks Conference (PolNet)Amlani, Butters & KieselDownload slides (.pptx)