Who Decides? Media, MAGA, Money, and Mentions in the 2022 Republican Primaries

• Published date: 01 December 2024
• DOI: http://doi.org/10.1177/10659129241268820
• Author: Rachel M. Blum,Mike Cowburn,Seth Masket
• Date: 01 December 2024

Article

Political Research Quarterly

2024, Vol. 77(4) 1314–1332

© The Author(s) 2024

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/10659129241268820

journals.sagepub.com/home/prq

Who Decides? Media, MAGA, Money, and

Mentions in the 2022 Republican Primaries

Rachel M. Blum

1

, Mike Cowburn

2

, and Seth Masket

3



Abstract

Political elites play an important role in determining who wins primaries, yet comparatively little is known aboutwhich

voices in party networks matter when different intra-party signals are sent. We examine this question using an original

dataset of Republican Senate and gubernatorial primaries in 2022, an election cycle with substantial intra-party conf‌lict in

primary elections. We demonstrate that Fox News appearances (media), Trump’s endorsements (MAGA),campaign

fundraising (money), and Twitter engagement (mentions) were all positively associated with vote share. We then assess

the state of primary f‌ields prior to Trump’s endorsements, showing that endorsed candidates were outperforming their

competitors prior to his involvement. Finally, we consider the state of primary f‌ields after Trump endorsed, dem-

onstrating that his support was associated with a thirteen percentage point increasein both fundraising share and polling

which lasted through to the primary. These f‌indings provide clarity on the relative weight of different signals in contested

party nominations.

Keywords

intra-party, Republican Party, Trump, primary elections, candidate selection

Primary elections offer a rare glimpse of intra-party com-

petition. Elite inf‌luence in primary elections is well estab-

lished (Cohen et al. 2008;Hassell 2018;Manento and Testa

2021), yet we know comparatively little about whose voices

matter in the context of conf‌licting elite signals. To better

understand the inf‌luence of disparate cues, we focus on non-

presidential primaries, which have become increasingly

important for U.S. party politics in the twenty-f‌irst century

(Cowburn 2022). By 2022, the Republican Party was in a

historically unusual position, where one individual—Donald

Trump—loomed large across nominations. We disentangle

the importance of varying elite cues by measuring the re-

lationship between Fox News coverage (media), Donald

Trump’s endorsement (MAGA faction), campaign fund-

raising (money), and social media attention (mentions), on

the outcomes of the highest prof‌ile, state-wide primary

contests: governor and U.S. Senate.

1

We f‌irst tackle the question of who decides, demon-

strating that all of these signals were, to varying degrees,

positively associated with vote share in Republican pri-

maries. Trump’s endorsements were strongly associated

with vote share and winning primaries. Candidates who

Trump endorsed received roughly ten additional per-

centage points of the vote, with a further six percentage

point penalty for candidates’whose opponents were

endorsed.

We further investigate Trump’s role in shaping the f‌ield by

focusing on the extent to which he served as a party

“kingmaker or a cheerleader”(Kousser et al. 2015). In a

second set of analyses, we focus on who Trump endorsed.

There, we f‌ind that when Trump endorsed, he did so in

support of candidates who were already leading their primary

f‌ields. Prior to receiving his support, Trump’s endorsees raised

more money, appeared on Fox News more often, received

more attention on Twitter, and were ahead in the polls. These

f‌indings highlight the endogeneity issue present when

studying intra-party signals that are likely interrelated and

empirically diff‌icult to isolate, suggesting two-way f‌lows with

Trump both leading and following candidates’performance.

Given that Trump’s candidates were already out-

performing their opponents prior to his endorsement, our

1

Department of Political Science, University of Oklahoma, Norman, OK,

USA

2

European University Viadrina, Frankfurt an der Oder, Germany

3

Department of Political Science, University of Denver, Denver, CO,

USA

Corresponding Author:

Mike Cowburn, European University Viadrina, Grroße Scharrnstraße 59,

Frankfurt an der Oder 15230, Germany.

Email: cowburn@europa-uni.de

third set of analyses consider the relationship between his

support and primary outcomes. Receiving an endorsement

by Trump was associated with a roughly thirteen per-

centage point boost in both campaign fundraising and

polling, and the improved polling positions lasted until the

primary election. Conversely, Trump’s endorsement bore

little relationship to the media landscape, with no asso-

ciated increase in Fox News appearances or social media

attention.

The question we pursue here is whether Donald

Trump’s unusually prolif‌ic endorsement behavior offset

the fundamentals of elections (e.g., campaign f‌inance,

media attention, incumbency, and activist support) in a

substantial way. In answering this question, we focus on

two possibilities: that Trump selectively endorsed the

candidates most likely to be successful, or that his en-

dorsement made candidates successful. Our f‌indings

suggest that Trump’s endorsement mattered in that it

attracted more of the things that make primary candidates

successful in the f‌irst place, such as campaign fundraising.

Candidate quality and early performance also mattered, as

Trump disproportionately endorsed candidates who were

already leading their f‌ields.

We recognize that this study suffers in terms of gen-

eralizability. That is, Trump and Fox News’inf‌luence over

one of America’s two major political parties does not

export well to other eras or political actors. This may be an

unusual—perhaps even unique—moment in the nation’s

political history, though it is certainly conceivable that

Fox’s and Trump’sinf‌luence are just larger than most

other actors in the same category. Even if this is a unique

moment, the moment has been going on for nearly a

decade, is highly consequential for U.S. politics, and does

not appear likely to abate in the near future. We have a

sense of how primary election dynamics function in

“normal”times; we believe the current dynamics in the

Republican Party warrant further attention from scholars.

We proceed as follows. First, we consider how dif-

ferent party signals in a primary might determine nomi-

nation outcomes. Next, we introduce our original dataset

of Republican primary candidates. We then present the

results of our three analyses, before discussing their

implications in the Republican Party and beyond.

What Matters in Primary Elections?

In 2017, then-Representative Ron DeSantis (R-Florida)

embraced a singular strategy to defeat the better-known

and better-funded Florida Agriculture Commissioner

Adam Putnam for the Florida Republican gubernatorial

nomination. DeSantis’approach was to appear on Fox

News as much as possible. This “Fox First”campaign

def‌ied the notion that politics is local, with DeSantis

instead deliberately making himself available to a national

audience. His reasoning was that more than 70 percent of

Florida’s Republican primary voters regularly watch Fox,

as did President Donald Trump. The approach paid off,

dramatically increasing DeSantis’name recognition.

Shortly thereafter, he received Trump’s endorsement, and

DeSantis went from trailing Putnam in the polls to leading

him, ultimately defeating him in the July 2018 primary by

roughly twenty points (Caputo 2018).

This account suggests that our understanding of pri-

mary elections may require updating. Media coverage,

name recognition, and the pursuit of prominent en-

dorsements have long been staples of primary election

campaigns. However, the overwhelming prominence of

one media outlet and one endorsement in modern Re-

publican primary elections would be something novel

(Cassino 2016;Hyun and Moon 2016).

American political parties slowly embraced binding

primary elections and by the mid-twentieth century, pri-

maries had become prominent methods for parties to

nominate candidates for most state and national off‌ices

(Kamarck 2018). Unlike general elections, where partisan

cues guide voters, primary outcomes are more sensitive to

features like name recognition, campaign spending, and

endorsements (Aldrich 2011). In the absence of a partisan

signal, primary voters have little information to guide their

choices, instead relying on campaign advertising and

signals from party elites in making their decisions. We

examine the impact of both of these factors more

explicitly.

First, we look at the impact of campaign advertising,

which is often a function of campaign fundraising. Weknow

from previous studies that fundraising plays a vital role in

primary elections—likely an even more powerful role than

in general election contests, where increased campaign

spending is known to boost primary vote shares (Breaux and

Gierzynski 1991). Funds may matter more if they come from

the right places. Albert, Desmarais, and Raja (2015) f‌ind that

donor networks were key in explaining primary success.

Similarly, Hassell (2018) f‌inds that candidates tended to

prevail in primaries when their donations predominantly

came from people who also donated to the party’s organized

campaign committees. The loosening of campaign f‌inance

restrictions in the wake of the Citizens United v. Federal

Elections Commission (2010) case may have fractured some

of those party f‌inancing networks (Boatright 2013), in-

creasing the importance of non-party organized and net-

worked donor groups can play in boosting primary

candidates (Manento 2019).

We also know that endorsements from inf‌luential party

insiders can play a role in determining primary voters’

choices. Cohen et al. (2008) found that presidential

candidates who received the majority of insider en-

dorsements typically win the nomination, and Dominguez

(2011) and Masket (2009) f‌ind a similar dynamic in

Blum et al.1315