Turning Out the Base or Appealing to the Periphery? An Analysis of County-Level Candidate Appearances in the 2008 Presidential Campaign

• Date: 01 May 2011
• Author: Andrew Reeves,Lanhee J. Chen
• Published date: 01 May 2011
• DOI: 10.1177/1532673X10385286
• Subject Matter: Articles

American Politics Research
39(3) 534 –556
Turning Out the Base
© The Author(s) 2011
Reprints and permission: http://www.
or Appealing to the
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1532673X10385286
http://apr.sagepub.com
Periphery? An Analysis of
County-Level Candidate
Appearances in the 2008
Presidential Campaign
Lanhee J. Chen1 and Andrew Reeves2
Abstract
We examine county-level campaign appearances by the Republican and
Democratic tickets during the 2008 general election. Our analysis reveals
that the McCain-Palin ticket campaigned in a way that was quite different
from the Obama-Biden ticket. McCain-Palin pursued a “base” strategy that
was focused on counties where Bush-Cheney performed well in 2004. They
also stayed away from counties that showed vote swings from 2000 to 2004
or population growth. On the other hand, the performance of the Kerry-
Edwards ticket in 2004 was a very weak predictor of where Obama-Biden
campaigned in 2008. They pursued a “peripheral” strategy that targeted
counties that had experienced significant population growth. Their efforts
to target peripheral, rather than base constituencies, have significant implica-
tions for our understanding of presidential campaign strategy.
Keywords
campaigns, elections, presidential candidates, campaign strategy
1University of California, Berkeley
2Boston University
Corresponding Author:
Lanhee J. Chen, University of California, Berkeley, Institute of Governmental Studies,
109 Moses Hall, #2370, Berkeley, CA 94720-2370. An online appendix and replication
materials available at http://people.bu.edu/areeves.
Email: launch@post.harvard.edu

---

Chen and Reeves
535
It comes as no surprise to anyone who follows presidential politics that the
major-party candidates during the 2008 presidential election spent a dispro-
portionate amount of time in swing states like Ohio, Florida, Pennsylvania,
and Virginia. In fact, visits to these four states accounted for nearly half of all
the campaign appearances made by the candidates for President and Vice
President of the United States combined.1 Both the political science literature
and popular writings on recent presidential elections are replete with accounts
of how presidential campaigns disproportionately devote their resources—in
terms of both time and money—to the handful of swing states that generally
decide the outcome of the election (e.g., Shaw, 1999c; Todd & Gawiser,
2009). Although the identity of the swing states may change from election to
election, the fact that campaigns train their attention on these crucial states
does not.
States, however, can be big and varied places. A candidate appearance in
the panhandle of Florida reaches a very different audience—and results from
different political and strategic motivations—than an appearance in Miami.
Scholars measuring the impact of candidate appearances have, almost with-
out exception, based their analyses on data that examine these appearances at
the state level (Herr, 2002; Shaw, 1999a, 1999b, 1999c; West, 1983).2 As a
result, we have an excellent understanding of which states candidates visit
and why, but we know less about where within swing states presidential can-
didates choose to go.
Strategists and scholars alike suggest two competing strategies for where,
within electorally important states, candidates and campaigns might spend
their time in an effort to maximize vote share. First is a “base” strategy where
candidates (and incumbent officeholders) target their core constituencies in
an effort to turnout known supporters. In this framework, candidates avoid
uncertainty and target those areas that have provided electoral reward in the
past. In contrast, a “peripheral” strategy suggests that politicians will do best
by targeting constituencies beyond their electoral base in an effort to win
over swing and new voters. Indeed, boosters of this strategy argue that what
matters most is finding new constituencies that can provide the necessary
support to win an election. We seek to test these competing hypotheses to
determine where, within swing states, candidates made appearances during
the 2008 presidential general election campaign.
We find that the Republican and Democratic tickets pursued divergent
strategies. John McCain and Sarah Palin pursued a base strategy by targeting
core partisan counties where the Bush–Cheney ticket performed well in
2004. Barack Obama and Joseph Biden, in contrast, pursued a peripheral
strategy. Indeed, high levels of Kerry–Edwards support in 2004 were not

---

536
American Politics Research 39(3)
associated with more campaign visits by Obama and Biden in 2008. Instead,
the Obama–Biden campaign pursued a peripheral strategy by targeting
counties that had seen substantial population growth.
This article proceeds in five subsequent sections. We first review the
pertinent literature and discuss the theoretical underpinning of our analysis.
Next, we provide a general overview of candidate appearances during the
2008 presidential general election. We then present our empirical model
and results, as well as a discussion of our findings. Our final section
concludes.
The Places They Go: Theoretical Underpinnings
Presidential campaigns devote significant time (and therefore resources) to
sending their candidates to rallies, town hall meetings, fundraisers, and other
events across the country.3 Substantially, more ink has been spilled on other
topics related to presidential campaigns, including the efficacy of campaign
advertising (e.g., Ansolabehere & Iyengar, 1995; Freedman & Goldstein,
1999; Jamieson, 1996) or on news media coverage of the candidates (e.g.,
Kerbel, 1995; Lichter & Noyes, 1996; Patterson, 1993). The lack of attention
is also notable given that campaign appearances also generate “free media”
that has the potential to change electoral outcomes by reaching voters
beyond merely those present at the event itself (Althaus, Nardulli, & Shaw,
2002; Herr, 2002). Where presidential candidates decide to make appear-
ances also gives us an eye into party realignment trends and a party’s
attempts to broaden its voter base or coalition of support (Sundquist, 1973;
West, 1983).
Scholarship on candidate appearances tends to either focus on how appear-
ances influence electoral outcomes4 or develop explanations for how cam-
paigns decide where to send their candidates for appearances. Recent
scholarship suggests that candidate appearances can produce positive elec-
toral outcomes for presidential campaigns (Herr, 2002; Shaw, 1999a). The
studies are not, however, unequivocal in their conclusions. In Shaw’s words,
“Too much should not be made of the campaign effects discovered [in this
analysis]” (p. 357). Hillygus and Jackman (2003) rightly suggest that view-
ing an electoral outcome as the sine qua non of dependent variables in studies
on campaign effects may oversimplify the otherwise complex process of
voter contact and response that takes place during the course of a general
election campaign.
But where do candidates actually choose to spend their time and make
public appearances? There is broad agreement regarding the states—or at

---

Chen and Reeves
537
least the characteristics of the states—that presidential candidates visit during
the general election campaign. Scholars agree that presidential candidates
tend to disproportionately allocate their resources and time to states that
have the potential to yield the greatest reward in the electoral college
(e.g., Bartels, 1985; Brams & Davis, 1974). Kelley (1961) and Colantoni,
Levesque, and Ordeshook (1975) concluded that a state’s competitiveness, in
addition to its importance in the electoral college, determined the targeting of
campaign resources. More recently, Shaw (1999b) asserted that a predictable
electoral college strategy, which is formed by campaigns before the start of
the fall campaign, generally dictates where campaign resources, including
candidate appearances, are allocated.5 These studies provide insights into the
allocation of resources on the state level but tell us little about the strategy
within those states.
Our study examines whether, within states, candidates make appearances
based on a “base” or “peripheral” strategy of electoral mobilization. Whereas
scholarship in this area has focused on states, there is at least one exception
(the only such study we were able to find) that addressed the question of
where candidates make appearances within electorally crucial states. Althaus
et al. (2002) concluded that candidates tend to visit larger media markets with
greater concentrations of likely voters. They show that these appearances do
not diverge substantially according to candidate partisanship and that counties
in competitive states, as compared to those in secure states, draw the greatest
number of candidate appearances.6 These are all useful findings, but they fail
to move us toward a better explanation of why candidates choose to visit
certain counties or areas within pivotal states.
The existing literature on campaign effects and related areas begins to lead
us toward an explanation of the types of voters—and therefore the specific
places within swing states—that candidates visit during presidential cam-
paigns.7 There are two divergent schools of thought on the places within states
where candidates might focus their campaign appearances. Our article aims
to ascertain which of these two strategies, which we have termed the base
and...