Cuyamaca College Typical Anatomy of a Journal Article in Political Science Download Article 2: Box-Steffensmeier, J. M., et al. (2019). “Cue-Taking in Congress: Interest Group Signals from Dear Colleague Letters.” American Journal of Political Science 63(1): 163-180.
Read Article 2
Optional: Schedule a tutoring appointment with the Writing Center or meet with a NetTutor (if available)
Identify the 10 parts of the article, as described in the Anatomy of a Journal Article
To demonstrate identification, hand write or electronically write in the margins of the article
Anatomy of a Journal Article
What does Journal Article Analysis Consist of?
Journal Article Analysis consists of reading journal articles and analyzing them. You are responsible for identifying ten parts of a journal article: title, main point and question, puzzle, debate, theory, hypotheses, research design, empirical analysis, policy implications, and contribution to the discipline and future research.
Journal Articles vary in their organization and inclusion of these ten parts. Many articles explicitly describe all or most of these parts; however, other articles may not state a part, or may omit it entirely.
You are responsible for analyzing 2 journal articles during the course.
Typical Anatomy of a Journal Article in Political Science
Journal Articles, especially in the field of political science, typically have ten parts.
The Title of an article appears on the first page of the article. The Title is brief, typically no more than 5-10 words, and identifies for the reader the subject of the article.
The Main Point and Question of an article are typically found in the Abstract. An Abstract is a summary of the article which is located on the first page, after the Title. The main point and question may be in the Introduction of the article.
The Puzzle is a missing piece of knowledge that the article seeks to fulfill.
The Debate is how scholars currently argue the subject of the article. Debates have at least two sides, and the two sides we are most familiar with are “pro” and “con”. However, debates can be more complex.
The Theory is how the author thinks something works. For example, we may have a theory about how campaigns influence voters. Theories consists of constants, variables, and the relationships between variables.
The Hypotheses are derived from the Theory. A hypothesis is the expectation that one variable effects another variable in a specific way.
The Research Design is how the author compares the effect of the explanatory variable (X) on the outcome variable (O) in a group (G) or set of groups.
The Empirical Analysis is the use of quantitative or qualitative evidence to explore whether the hypothesized relationship between two variables does indeed occur in the world.
The Policy Implications are how the findings of the article should influence the behavior of individuals, groups, organizations, or governments.
Finally, the Contribution to the Discipline and Future Research are how the article helps fill the missing Puzzle piece, as well as offer suggestions for future research that build on the findings from the article. Cue-Taking in Congress: Interest Group Signals from
Dear Colleague Letters
Janet M. Box-Steffensmeier The Ohio State University
Dino P. Christenson Boston University
Alison W. Craig University of Texas at Austin
Abstract: Why do some pieces of legislation move forward while others languish? We address this fundamental question by
examining the role of interest groups in Congress, specifically the effect of their legislative endorsements in Dear Colleague
letters. These letters provide insights into the information that members use to both influence and make policy decisions. We
demonstrate that endorsements from particularly well-connected interest groups are a strong cue for members with limited
information early in the legislative process and help grow the list of bill cosponsors. As bills progress, such groups have less
direct weight, while legislation supported by a larger number of organizations and a larger number of cosponsors is more
likely to pass. Thus, we illuminate the usage of Dear Colleague letters in Congress, demonstrate how members use interest
groups in the legislative process, and shed new light on the varying impact of groups on public policy.
Replication Materials: The data, code, and any additional materials required to replicate all analyses in this arti-
cle are available on the American Journal of Political Science Dataverse within the Harvard Dataverse Network, at:
https://doi.org/10.7910/DVN/JNEQYJ.
M
embers of Congress are routinely asked to
make complex policy decisions with limited
time and information. They want to make the
decisions that will be best for their electoral prospects
and result in good public policy. However, the challenges
of doing so with imperfect information have been well
documented (Fenno 1973; Mayhew 1974). Thousands
of bills are introduced each congress, and in an average week, a member may be expected to take positions
on legislation in a wide array of policy issues, from foreign policy to water rights. Members work to overcome
these challenges by specializing on the issues before their
committees and relying on cues to understand the political and electoral implications of a bill (Fenno 1978;
Kingdon 1981; Krehbiel 1991). However, some cues are
stronger than others. Thus, exploring how members of
Congress respond to signals from different groups and at
different stages of the legislative process provides a greater
understanding of how the legislative branch lives up to the
ideals of pluralism and democratic representation.
We examine the role of interest groups in providing members of Congress with informative signals that
help them decide which policy proposals are worth supporting. While we may wish for members to read and
understand every piece of legislation that comes across
their desk, in reality they simply do not have the time.
Even if they read every bill, members cannot be expected
to effectively comprehend the implications of complicated legislative text across a broad array of policy issues.
When legislation is considered in committee, there are extensive discussions in which members hear from interest
groups and other interested parties to gain an understanding of the arguments for and against a particular
policy (Arnold 1990). However, members are frequently
asked by colleagues, constituents, and interest groups
to take positions on legislation outside of their area of
Janet M. Box-Steffensmeier is Vernal Riffe Professor, Department of Political Science, The Ohio State University, 2140 Derby Hall, 154
N. Oval Mall, Columbus, Ohio 43210 (steffensmeier.2@osu.edu). Dino P. Christenson is Associate Professor, Department of Political
Science, Boston University, 232 Bay State Road, Boston, MA 02215 (dinopc@bu.edu). Alison W. Craig is Assistant Professor, Department
of Government, University of Texas at Austin, 158 W. 21st Street, STOP A1800, Austin, TX 78712 (acraig@austin.utexas.edu).
The authors are listed alphabetically. We would like to thank Matt Grossman, Geoff Lorenz, Eleanor Powell, and our anonymous reviewers
for comments as well as Andy Duthie and Joe Rafalak for research assistance. The research reported here is supported by the National
Science Foundation’s Law and Social Science Program and Political Science Program (grants #1124386, #1124369, #1627422, #1627358),
the National Science Foundation Graduate Research Fellowship (DGE-1343012), the Dirksen Congressional Center (grant #00031432),
and the Institute for the Study of Democracy at the Ohio State University.
American Journal of Political Science, Vol. 63, No. 1, January 2019, Pp. 163–180
C 2018,
Midwest Political Science Association
DOI: 10.1111/ajps.12399
163
164
JANET M. BOX-STEFFENSMEIER, DINO P. CHRISTENSON, AND ALISON W. CRAIG
expertise, before it is considered in committee, or sometimes before it is even introduced. We argue that one of
the ways they are able to do so is by relying on endorsements from interest groups, which serve as informative
signals for members with limited time and resources early
on, just as their committee testimony and lobbying efforts do in later stages of the legislative process (Arnold
1990; Grossmann and Pyle 2013; Hall and Deardorff
2006).
Using a unique data set of nearly 100,000 electronic
Dear Colleague letters sent between members of Congress
from 1999 to 2010, we identify the interest groups who
have endorsed legislation and determine the impact of
those endorsements at multiple stages of the legislative
process. We combine this novel source of data on signals
in Congress with a measure of the social power of interest
groups based on their relative locations within a network
of collaborative and purposive behavior in which organizations choose to coordinate their advocacy activities
(Box-Steffensmeier and Christenson 2014). Doing so provides a new way to document participation in American
democracy by interest groups and allows us to consider
how members of Congress rely on signals from interest groups to guide their position taking, and which signals are most effective. We demonstrate that members of
Congress both utilize and respond to endorsements from
interest groups mentioned in Dear Colleague letters. Endorsements from groups that are more well connected
within the advocacy community send stronger signals in
the earliest stages of the legislative process when there are
a large number of bills competing for a member’s attention and limited information about their likely impact. By
contrast, endorsements from a larger number of groups
are better indicators of success later in the process when
members need to amass a winning coalition of 218 votes.
Signals as Subsidy
Members of Congress are expected to be knowledgeable
about and take positions on a wide array of issues. At
the same time, they are limited in both time and resources. Members have always had to make choices about
where to allocate their time, choosing which issues to
prioritize and relying on their colleagues and interest
groups for guidance in other policy domains (e.g., Hall
1998; Kingdon 1981; Matthews 1960). However, the gap
between a member’s workload and available resources
has grown considerably in recent years, as members are
elected to represent larger districts, devote more of their
limited time to fundraising, and have fewer policy staff to
support them.
The expenses of running for office have reduced the
congressional capacity for policy analysis as members
spend more time fundraising, a necessity not only for
members facing a close election, but also for any who
aspire to leadership positions (Heberlig 2003; Heberlig,
Hetherington, and Larson 2006). In 2013, a presentation
for newly elected members by the Democratic Congressional Campaign Committee suggested the “Model Daily
Schedule” for a member consisted of one to two hours of
meetings with constituents, two hours for committee and
floor proceedings, one hour for strategic outreach and
press, one hour of “recharge time,” and four hours of call
time, in which members place fundraising calls (Grim
and Siddiqui 2013). Nowhere in this allocation is time
for reviewing legislation introduced by other members or
any sort of policy analysis.
Instead, members rely on professional staff to do the
bulk of this work, supporting the member by studying
proposals and recommending which bills are worth supporting (Salisbury and Shepsle 1981). However, staff levels in Congress are declining. Although members represent 200,000 more constituents than they did 30 years ago,
the number of professional staff in the House of Representatives declined from 11,537 in 1985 to 7,703 in 2015.1
At the same time, the Government Accountability Office, Congressional Research Service, and Congressional
Budget Office, which were created to provide nonpartisan professional advice on technical issues, lost 37% of
their staff, and the Office of Technology Assessment was
eliminated (Roberts 2017).
As a result, there has been a steady decline in the
number of issues Congress can address and its capacity
for gathering the information necessary to create policy
(Baumgartner and Jones 2015). A 2016 survey of congressional staff found that while 67% of respondents felt
it was important for the functioning of Congress that
“members have adequate time and resources to understand, consider and deliberate policy and legislation,”
only 6% were satisfied with Congress’s performance in
that regard.2 Similarly, only 15% of respondents felt that
“staff knowledge, skills and abilities are adequate to support members’ official duties,” and 11% agreed that “the
chamber has adequate capacity and support (staff, research capability, infrastructure, etc.) to perform its role
in democracy” (Goldschmidt 2017, 9).
Members cope with these constraints by becoming
specialists, focusing their work on the issues before their
committees, and using informative signals to guide their
1
2
This figure includes committee, personal, and leadership staff.
Respondents were chiefs of staff, legislative directors, communications directors, and state/district directors.
CUE-TAKING IN CONGRESS
decisions in other areas. These signals may take a number
of forms, including memos from party leadership, committee reports, and colleagues’ voting decisions (Kingdon
1981). Members also rely on interest groups to provide
the information they need to make policy decisions. Interest groups have a long and well-documented history
of providing expert testimony in committee hearings,
dating back to the Progressive Era (e.g., Burstein and
Hirsh 2007; Evans 1996; Goss 2013; Tichenor and Harris
2002–03). They also provide legislators with information
on the views of their constituency (Hansen 1991), communicate issue salience (Ainsworth 1993), serve as brokers between other interest groups (Box-Steffensmeier
and Christenson 2014, 2015, 2017; Heaney 2006), and
help members intervene in agency decisions (Hall and
Miller 2008). Hall and Deardorff (2006) reconcile the
varying accounts of what interest groups do by arguing
that lobbyists form an extension of a member’s office,
using their expertise to provide policy information and
political intelligence to their allies in Congress. This information is invaluable to members making voting decisions.
In this article, we expand the understanding of when
and how interest groups influence the legislative process.
We argue that endorsements from organizations that are
well connected in the larger interest group community
provide a quick and highly informative signal to members as they review the thousands of bills introduced each
congress to determine which are worthy of support. A single endorsement from an organization with ties to other
well-connected interest groups can influence a member’s
decision to cosponsor a bill. At the same time, endorsements from a larger number of interest groups signal that
a bill has broad support, which translates into a larger
lobbying team and increased votes.
Members of Congress receive daily requests from colleagues, constituents, and interest groups to cosponsor
legislation, as cosponsorship is an important form of position taking and a key element of the winnowing process
that determines which bills will be among the approximately 14% that move onto the House floor (Koger 2003;
Krutz 2005). In the absence of a committee report or
recommendation from the leadership, members rely on
the information provided by the bill’s sponsor, who has
no incentive to warn a colleague of any potential political issues with the bill and little knowledge of his or her
colleague’s electoral landscape. With limitations on both
time and information, members seek out clear political
signals regarding the strength or opposition to a proposal
(Kingdon 1995; Wright 1995). At the same time, interest
groups seeking policy change work to improve the visibility of their issue among policy makers (Baumgartner et al.
2009). Endorsements fulfill the needs of both members
165
and interest groups, drawing a member’s attention and
providing a quick and easy-to-process indication that the
group supports the legislation in question. Therefore, a
member who is concerned with the approval of a particular interest group will know that a bill is “safe” to support
if it has their endorsement.
For example, if a member is asked by her constituents
to cosponsor a health care bill, doing so might earn her
goodwill with these voters. However, she may not know
how the bill will affect health care providers or broader
populations in her district and does not have the time or
expertise to study it in depth. If the bill is endorsed by
the American Medical Association or the American Hospital Association, that endorsement sends a signal to the
member that the bill has been vetted by a leading industry
expert outside the sponsor’s office and supporting it will
not cause problems with health care providers.
In addition to signaling the policy implications of a
bill, there are electoral aspects to interest group endorsements. Groups are selective in her endorsements, providing their name to initiatives they strongly support rather
than bills they find merely unobjectionable. A member
looking to gain favor with an organization may sign onto
a bill that group has endorsed, hoping her support will
result in some benefit, whether in the form of a campaign contribution, checkmark on the organization’s annual scorecard, or mention in a newsletter.
However, not all endorsements carry the same weight.
Some organizations are more influential than others because they have more resources, greater name recognition,
or better connections within the interest group community. When legislators are relying on interest group signals to guide their decisions, some signals are inherently
stronger than others. An endorsement from the National
Association for Sport and Physical Education (NASPE)
will tell a member less about the political ramifications
of a bill than one from the American Farm Bureau, particularly if the member represents a district with a strong
agriculture industry. Furthermore, there is likely less of an
electoral benefit from supporting a bill backed by NASPE
than the American Farm Bureau, which has a larger network of voters in which to disseminate the information.
How do we determine which groups are likely to be
more influential? Scholars have used a number of different metrics for interest group influence, from campaign
contributions to hours spent lobbying (see Smith 1995).
We focus on the relationships between interest groups,
arguing that organizations that are well connected in
the interest group community will be more influential
in the policy arena. Interest groups frequently work in
coalitions, not only when lobbying Congress, but also
in other venues, such as the judicial and administrative
166
JANET M. BOX-STEFFENSMEIER, DINO P. CHRISTENSON, AND ALISON W. CRAIG
branches (Box-Steffensmeier, Christenson and Hitt 2013;
Box-Steffensmeier, Christenson and Leavitt 2017; Hula
1999; Nelson and Yackee 2012). Organizations that are
positioned at the core of a network of interest groups are
seen as more influential, and these “pivotal” organizations
are more likely to be recruited to a coalition and attract
other groups to join (Hojnacki 1997). The strength of
their relationships with other organizations means they
are likely to be well known and in a position to influence
other groups. Therefore, as a result of the social power
gained from their connections within the interest group
community, we expect organizations that are well connected will be more influential in the legislative process.
H1: Endorsements from well-connected interest
groups provide informative signals to members
of Congress reviewing legislation.
Members of Congress are not merely passive receivers of these signals, as legislators and lobbyists frequently coordinate their strategies and mobilization efforts (Ainsworth 1997; Baumgartner et al. 2009; Mahoney
and Baumgartner 2015). In interviews with congressional
staff,3 the importance of obtaining key interest group endorsements in the early stages of the legislative process
was repeatedly emphasized. Members of Congress work
to gain the support of stakeholders when they introduce
a bill because they expect it will increase the likelihood of
success. One staff member described how his boss “likes to
build a network of both local and national interest groups
because they bring outside expertise and lend credibility
and resources.” When asked what members see as the
value of interest group endorsements, multiple offices
provided answers similar to one legislative director who
said, “if you have an industry backing a bill, that’s the
easiest way to get cosponsors.” As a result, we expect that
well-connected and therefore well-known organizations
with high social power will be particularly influential in
this early stage. When there are thousands of proposals
competing for a member’s attention and limited available
information about a bill’s implications, the sponsors can
use an endorsement from a prominent organization to
make their legislation stand out.
H1a: The value of this signal is stronger in the early
stages of the legislative process.
The strength of an organization’s relationships within
the interest group network is not the only way we expect coalition strategy to influence the legislative process.
We also examine the relationship between the number
3
The authors conducted personal interviews with member staff in
the House in 2016.
of groups endorsing a bill and its success. Coalitions allow organizations to demonstrate broad support for a
policy, share resources, and overcome the challenges of
many organizations competing for limited attention on
the issue agenda (Baumgartner and Jones 1993; Esterling
2004; Holyoke 2009). Existing research has found limited empirical support for a direct connection between
membership in a lobbying coalition and policy outcomes
(Hojnacki 1997; Hula 1999). However, interest groups
that endorse the same legislation are not necessarily coordinating their efforts and engaging in traditional coalition
strategies, which allows us to examine the effectiveness of
lobbying by larger groups of organizations, regardless of
whether they coordinate their strategies. Put simply, we
expect that when a bill is endorsed by a large number of
organizations, the endorsements collectively demonstrate
broad support for the policy in question, and that support
will translate into greater legislative success regardless of
each organization’s social power.
Considering the American Farm Bureau again, in
some cases the well-connected overarching organization
may not endorse a bill because it involves picking sides
among their own members. A proposal may be beneficial
to farmers in one part of the country or in a particular
industry, but detrime…
Purchase answer to see full
attachment