This page is part of an undergraduate assignment at Davidson College.
| Works Cited |
Literature Review
Endorsements are “almost never decisive in a modern presidential campaign.” “The
only time an endorsement might affect a campaign's fortune is if the candidate's
mother failed to issue one” (Baker, 2004). And what about this, “Very
few studies have examined the effect of endorsements, seemingly because political
scientists have generally assumed their impact to be minimal” (Hill 3,
2003). Perhaps they are both trying to be funny (well we know Baker is just
trying to be funny), but in actuality, endorsements of political candidates
are important to political campaigns from local government to the most recent
Presidential Democratic primaries. Many political scientists do not share this
deflated opinion of endorsements, and after the great importance that this
year’s candidates gave to the endorsements; it is hard to make that statement.
There is one study that shows the effect of actual politicians endorsing candidates.
Roshwalb and Resnicoff examined the 1970 New York Senatorial election in which
William Buckley received a last minute endorsement from the President and Vice
President. They conclude that “this study suggests that when there is
a substantial degree of indecision, endorsements by leading political figures
and published polls may influence the ultimate decisions of appreciable numbers
of voters” (Roshwalb and Resnicoff 414, 1971). And that’s basically
the literature on the topic…. Other than this measly paper, I was hard
pressed to find a study that directly looks at the effect of politicians or
party officials endorsement of candidates, but other endorsement research can
provide many of the underlying themes despite specifically looking at interest
groups, newspaper and political party endorsements.
In Timothy Hill’s study on interest groups, he found data that was not
statistically significant, but only suggested a relationship that a given participant
would vote for the endorsed candidate. However, he did a particularly good
job of getting inside the endorsement and understanding the finer minutia.
According to Hill, there are two possible voting decisions that endorsements
might influence: the decision of which candidate to support, and the decision
to vote (Hill 6, 2003). Endorsements typically have a larger effect earlier
in a race because it serves to “prime” the voter before information
is available (Hill 28, 2003). Endorsements will have less of an effect in more
visible races like the president because voters will know the president outside
of an endorsement context. A similar issue is usually at play with favored
candidates or incumbents. Oftentimes, they are expected to win and have already
hit the “glass ceiling,” but this typically sets up the “underdog” for
a heads start (Hill 177, 2003).
Lupia has a unique way of looking at endorsements. Typically, people will be
skeptical and dismiss endorsements as a ploy of some kind, but professor Lupia
looks at endorsements differently by developing a model in which he proves
that “voters can use endorsements to make more accurate inferences (i.e.,
the types of inferences that “informed” respondents make) and increase
the likelihood that they cast the same vote they would have case if the had
possessed better or complete information” (Lupia 397, 1992). She sees
an endorsement as a resource, but the problem is that most see it as a danger
snake because “Conventional wisdom has it that an influential endorsement
will sway a quantum of votes for those who acknowledge the influence” (Hollander
407, 1979). “The motivation for introducing the endorsement comes from
the fact that voters (collective decision makers in general) may not have an
incentive to acquire costly political information. These decision makers may
have greater incentive to use reliable low-cost information cues if they are
available (Lupia 393, 1992).
In their article, “Do Endorsement’s Matter? Group Influence in
the 1984 Democratic Caucuses,” Ronald Rapoport, Walter Stone and Alan
Abramowitz analyze candidate support and prenomination activity for Walter
Mondale by the members’ of endorsing teachers’, women’s and
labor groups, and they concluded that there is a substantial impact from the
teachers’ and labor groups but not from the women’s group.
Rapoport et al. echo the theme of the “instrumental responsiveness” –“…massaging
opinion to make it serve one’s own purposes” (Ornstein and Mann
32, 2000)—“Since candidates seek to build momentum early in order
to generate credibility and support in the round robin of caucuses and primaries,
groups can be key players in a process no longer dominated by organization
leaders of the parties” (Rapoport 194, 1991) Endorsements have completely
changed politics. Research about Newspaper endorsements says almost exactly
the same thing: “The decline of party and the rise of candidate-oriented
election campaigns have underscored the importance of the question of what
impact campaigns events have on election outcomes.” (Scarrow and Borman
388, 1979) But there is a bad side to this idea of candidate centered campaigns
((Jacobsen 1997) as referred to in Sellers 133), and that is that fact that “groups
can have direct influence on the nomination process, they have the ability
to make demands on candidates directly and circumvent ongoing party organizations.” (Rapoport
201, 1991)
Endorsements, however, do nothing on their own, they are bound to the media,
and often cases are a chance to get that “line of the day” just
as Nixon did in his campaign in 1970 (Cook 138). The key to deriving impact
from endorsements was usually to advertise them. One example that the New York
Times insinuated how one reason Gore won his primary was because Gore advertised
that he had Tom Harkin’s support whereas Bradley didn’t advertise
that he Senator Paul Wellstone’s support, so therefore no one knew and
he couldn’t benefit from it. (Clymer 2000).
The same ideas are at work behind endorsements and the “promotional efforts” that
produced “more favorable coverage for political parties” that Professor
Sellers wrote about in “Winning Media Coverage in the U.S. Congress” (Sellers
150): “Press coverage may produce electoral benefits among constituents” (Sellers
135).
Even looking at an example back in 1983 before the age of 24 hour news, a study
showed that a faithful viewer of CBS Evening News from January to October was
exposed to at total of 15.5 hrs of presidential campaign coverage, of which
10 hrs were devoted to the “horse race” and about 5.5 to candidate
information and policy issues” (1983, 149) (Bartels 271, 1993).
Despite endorsements naturally simulating a “horse race,” the “horse
race” can still be severely slowed depending on the format that it is
presented in, and that is what is unique about my model, which looks at newspaper
coverage. Other newspapers like the U.S.A. Today or The New York Post better
lend itself to reporting a “rat race,” but the papers that I draw
from in my analysis (New York Times, LA Times, the Telegraph Herald of Iowa,
Union Leader of New Hampshire, and Post Courier of South Carolina) do not.
There has been a “shift from hard news to soft news that has surely been
more dramatic in local than in national television news, yet greater in national
television news than in the New York Times in the past 20 years” (Livingston
and Bennett, 367, 19)
“ The organizational news-gathering routines that establish the working
relations between reporters and sources” (Livingston and Bennett 368) are
still like the used to be more or less with newspapers. “Newspapers are
still very institutional” (Bennett and Livingston 376)” The difference
between print and broadcast media is further evidence that news coverage is negotiated
between candidates and journalists. Television’s tremendous appetite for
visuals may make television more vulnerable to candidate control than newspapers” (Just
et al. 106, 1996)
© Davidson College, 2004, Department of Political
Science, Davidson College, Davidson, NC 28035
Send comments, questions, and suggestions to Dr.
Patrick Sellers, Professor of Political Science.
Created: 4/29/2004. Last updated: 5/03/2004.