Polls and elections: firing back: out-party responses to presidential State of the Union addresses, 1966-2006.
Panagopoulos, Costas
The U.S. Constitution requires American presidents to deliver
regular updates about the "State of the Union" to Congress.
Article II, Section 3, stipulates that the president, "shall from
time to time give to the Congress information about the State of the
Union, and recommend to their consideration such measures as he shall
judge necessary and expedient." This constitutional mandate
establishes the formal basis upon which presidents have advanced a
tradition of yearly reports to Congress about the state of affairs in
the nation known formally as the "annual message" between 1790
and 1934. President George Washington combined his inaugural address and
his annual message, and delivered it to a joint session of Congress in
New York City on April 30, 1789. He delivered his first regular annual
message on January 8, 1790.
Over time, the content and focus of the annual message has changed
considerably. In the nineteenth century, annual messages were typically
lengthy and technical administrative reports on the executive branch.
But after 1913, when Woodrow Wilson revived the practice of presenting
the message to Congress in person, it became a platform for the
president to rally support for his agenda. Advancement in communications
technology further enabled presidents to use the message as a forum to
speak directly to the American public. The address has been known
generally as the "State of the Union" since 1947, the year in
which President Harry Truman delivered the first televised broadcast of
the message. (1)
Since 1966, television networks have provided airtime for the
opposition party to deliver a response to the president's State of
the Union address. Senator Everett Dirksen (R-IL) and Congressman Gerald
Ford (D-MI) delivered the first rebuttal to President Johnson's
State of the Union message on behalf of the Republican Party. The
tradition of televised opposition responses to the president's
annual message to Congress by the out-party continues.
This study examines patterns in opposition parties' decisions
about which members deliver responses to the president's State of
the Union address over the past four decades. I investigate and compare
the major parties' selection of rebutters with respect to their
political backgrounds, institutional affiliation, and demographic and
political characteristics. I also develop and estimate an empirical
model to examine the impact of opposition responses on the effectiveness
of a president's State of the Union address to Congress and to
explore how differences in the selection of out-party representatives to
deliver rebuttals helps to explain the effectiveness of the response.
Strategy of "Going Public"
U.S. presidents typically attract greater media attention than any
other singular political actors. In his seminal work on the presidency,
Neustadt (1960) argued that part of presidents' "power to
persuade" can be derived from leveraging publicity to advance their
agenda. Kernell (1997) subsequently argued that presidents increasingly
employ a strategy of "going public"--appealing directly to
citizens with the help of the media--to strengthen support for their
policy proposals. Several studies suggest presidents may indeed be
capable of influencing public views by going public. Cohen (1997) has
shown, for example, that presidential communications can influence the
public's agenda, at least in the short term. Most studies, however,
detect only modest effects of presidential leadership on public opinion
and presidential approval. Baum and Kernell (2001), for example, find
that Roosevelt's radio addresses only lifted his approval ratings
by 1 percentage point on average. Similarly, Ragsdale (1984) reports
increases of 3 percentage points in approval in her study of
presidential speeches between 1949 and 1980. Brace and Hinkley (1992)
show that major addresses add 6 percentage points to presidential
approval ratings. Simon and Ostrom (1989) conclude that presidents'
televised speeches do not affect presidential approval at all. In the
most comprehensive, recent treatment of the topic, Edwards (2003)
contends that presidential efforts to engender greater public support
for their policy preferences by going public are generally unsuccessful.
Edwards observes, for example, that presidential approval before and
after major televised speeches, including State of the Union addresses
to Congress, rarely shifts significantly.
Edwards (2003) considers a host of explanations to account for why
presidents are unable to move public opinion by going public. He argues
that variation in characteristics of the messenger, message content, and
dwindling audience sizes may explain presidents' inability to
penetrate the public mindset. One possibility not considered is that
presidential communications do not occur in a vacuum. Even if they
attract less media attention, opposition party communications may
constrain the impact of the president's efforts by articulating
alternative viewpoints and criticism.
Out-party responses to the president's State of the Union
message provide an ideal opportunity to study the possibility that
rebuttals to presidential messages affect the impact of the
president's efforts. In accord with Edwards' (2003)
theoretical framework, I also consider the impact of the out-party
"messengers" on overall effectiveness. In my view,
characteristics of the out-party messengers selected to respond to the
presidential address can be consequential. I investigate these claims by
examining the rebutter selections of out-party between 1966 and 2006. I
then develop and estimate an empirical model to explain the impact of
strategic decisions about rebutter selection on the effectiveness of the
president's State of the Union message as measured by public
approval.
Who Fires Back? Out-Party Rebuttals 1966-2006
The out-party has delivered 35 televised responses to the
president's State of the Union addresses over the past forty years.
(2) Details about the out-party representatives selected to present the
opposition response appear in the appendix. A summary of these details
appear in Table 1.
Analyses of these data reveal there has been considerable variation
in the number of out-party members to deliver a response to the State of
the Union between 1966 and 2006. The first two responses were delivered
by only two out-party members, representing both chambers of Congress.
In both 1966 and 1967, Senator Dirksen and Congressman Ford delivered
the response. Several responses, especially between the late 1960s and
the mid-1980s, brought together relatively sizable groups of elected
officials to deliver the response. In 1968, for example, the Republican
Party recruited 16 members to respond to President Lyndon Johnson's
message, the largest group to ever deliver a rebuttal. On average, the
out-party response has been delivered by three representatives of the
opposition party.
Table 1 also shows there is considerable variation in terms of the
institutional affiliations of opposition party members selected to
deliver responses to the State of the Union. Overall, two U.S. senators
and two U.S. House of Representative members on average are included in
the group of rebutters. Occasionally, the out-party will ask one of its
governors to respond to the president's address, and, in one
instance (1986), the Democrats asked Harriett Woods, Missouri's
lieutenant governor to respond. Most of the opposition party rebutters
during the period 1966-2006 (48% on average) are U.S. senators, although
40% of the rebutters are House members. The average group also consists
of 12% governors and 11% women.
Table 2 presents details about partisan differences in the
composition of the groups of out-party representatives selected to
deliver responses to the president's State of the Union address.
Data presented in Table 2 suggest Democratic rebuttals are delivered by
larger groups of rebutters: four on average compared to three for
Republicans. The typical group of Democratic rebutters also includes two
senators and two House members (compared to one senator and one House
member for Republicans). However, because Democratic rebuttal groups
tend to be larger, the proportion of senators in the overall group is
lower for Democrats than for Republicans (45% to 54%, respectively). The
proportion of governors included in Democratic rebuttal groups also
exceeds Republicans by nearly a two-to-one margin (14% to 8%,
respectively). The Republican Party appears to showcase its female
members more so than the Democratic party does in terms of members it
selects to present the State of the Union response: 16% of Republican
rebutters are women compared to 8% of Democratic rebutters. (3)
Table 3 reports partisan differences in the overall ideology of the
group of out-party responders. I measure ideology using the DW-Nominate
scores developed by Poole and Rosenthal (1997). (4) Measured on a scale
that ranges from -1 (most liberal) to +1 (most conservative), I average
the ideology of all representatives in the group of rebutters to
determine an overall estimate of the out-party response group. The data
presented in
Table 3 indicates that groups of Democratic rebuttals tend to be
more ideologically extreme than Republican rebutters are over the period
of this study. Moreover, I also examine the ideological distance between
the ideology of the out-party rebuttal group and the president who
delivers the State of the Union address. I find that Democrats tend to
designate rebutters or rebuttal groups that are ideologically more
distant from the president than are Republican groups of rebutters.
Ten of the 35 out-party rebuttals I examine are delivered during
presidential election years. Because it is reasonable to expect that the
parties' strategic considerations will differ during electoral
cycles, I assess differences in the composition of rebuttal groups in
presidential and non-presidential election years. These data are
presented in Table 4. The evidence reveals the total number of rebutters
is greater in presidential election years (five rebutters on average,
compared to three in nonpresidential election years). The data also show
that the out-party designates more senators to respond to the president
during presidential election years (three versus one, respectively).
Overall, 60% of rebutters are U.S. senators in presidential election
years, compared to 44% in nonpresidential years. Even as the number of
rebutters who are U.S. House members doubles in presidential election
years (from one to two, respectively), the overall proportion of House
members does not differ much between presidential and nonpresidential
election years. Table 4 also reveals that out-parties simply do not
designate governors to respond to the president's State of the
Union message during presidential election years.
I may also expect that out-parties will attempt to project
ideological moderation more so during presidential election cycles than
in other years by designating groups of rebutters that are ideologically
more moderate. Data presented in Table 5 provide some empirical evidence
along these lines. The details indicate that the Republican Party has,
in fact, designated more moderate rebutters in presidential years
compared to nonpresidential years (0.258 average group ideology versus
0.367, respectively), but Democratic responders' ideology is nearly
identical in both presidential and nonpresidential years (-0.388 and
-.0389, respectively).
Developments over Time
In this section, I analyze developments over time with respect to
out-party selections of State of the Union rebutters. In theory, parties
can incorporate information about the effectiveness of decisions about
the composition of rebuttal groups in order to be strategic and to
maximize impact. Assessments about patterns of compositional changes
over time may reveal key insights about lessons the out-parties may have
learned over the four decade period I examine. (5)
Perhaps the most evident trend over this period is the steady
decline (statistically significant at the p < .01 level) in the
overall number of responders designated by the out-party to rebut
presidential State of the Union addresses. Analysis of the data suggests
much of this erosion is fueled by developments in the Democratic Party,
which historically designated larger groups of rebutters. In the most
recent responses, both parties designate no more than two
representatives to deliver rebuttals. In fact, no more than two
rebutters have responded to the State of the Union since 1986.
The mix of the levels of elected officials tapped to deliver
out-party responses has also changed markedly over the period of my
study. For example, the overall number, as well as the proportion, of
U.S. senators designated to respond to the State of the Union generally
declined between 1966 and 2006. Even as rebuttals are often delivered
only by U.S. senators (1994, 1996, 1998), it is not unusual, especially
since the mid-1980s, to find no senators responding to the
president's State of the Union message. Much of this decline is
driven by developments in the Democratic Party. Democrats increasingly
overlook senators in designating rebutters to the annual message.
Similarly, the overall number of House members asked to respond to the
State of the Union has also dropped over the past four decades. Since
the mid-1980s, no more than one House member has participated in a
rebuttal (except in 1999 when two House members delivered the Republican
response), and no House members have responded in 7 of the past 12
responses. Despite this decline, the proportion of House members who
participate in a response group has remained relatively stable over this
period, at roughly 40% of the rebuttal group. I find no evidence of
significant, partisan differences in the proportion of House members who
deliver rebuttals over time.
Above I noted that governors are only occasionally designated to
respond to the president's State of the Union address, but my
evidence suggests clear differences over the period I examine.
Out-parties are increasingly turning to state executives to respond.
Both the overall number as well as the proportion of governors tapped to
respond to the president's message appear to be on the rise during
this period. This trend is most pronounced in the Democratic Party,
which has turned exclusively to governors to respond to the president in
two of the four most recent rebuttals. Republicans, conversely, have
only designated one governor to respond to Democratic presidents'
State of the Union message: in 1995 when New Jersey Governor Christine
Todd Whitman delivered the rebuttal.
Next I examine patterns in the total number of women designated by
the out-party to deliver responses to the State of the Union between
1966 and 2006. The data indicate there were no female responders in 14
out the 35 rebuttals included in the analysis. One female rebutter was
designated to respond in 10 of the 35 rebuttals, and, in 1972, Democrats
selected 2 women to be part of the 11-member group that delivered the
rebuttal. Still, the number of female responders relative to the overall
size of the group appears to be on the rise over this period. Moreover,
both of the major parties are increasing the proportion of female
responders in rebuttal groups.
Evidence of ideological polarization in recent years is ample (Bond
and Fleisher 2000). I find the overall ideology (using available
DW-Nominate ideology scores) of out-party representatives selected to
respond to the presidents' State of the Union addresses has also
polarized between 1966 and 2006. Analysis of my data reveals both
parties' rebutters have become growingly ideologically polarized
over time. Between the late 1960s and the early 1980s, Republican
rebutters' overall ideology was increasingly moderate. The trend in
Republican rebutters' ideology began to reverse in a more
conservative direction in the early 1980s. Democratic responders'
ideology overall also exhibits signs of growing moderation until the
late 1980s when it began to shift toward a more liberal position. Taken
together, these two developments indicate the parties' State of the
Union rebutters have never been as ideologically polarized as they have
been in the most recent years included in the analysis.
Explaining the Effectiveness of Presidential State of the Union
Addresses: The Impact of Strategic Choices in Rebuttals
In the sections above, I have described the empirical evidence
about out-party selections of representatives to respond to the
president's State of the Union message. I have also identified
trends and developments over time in an effort to gauge the
parties' strategies and choices over a period of four decades. In
this section, I examine the extent to which rebuttal selection decisions
affect the effectiveness of out-party's message relative to the
president's agenda. The analyses that follow are designed to help
explain how choices about out-party responses impact the effectiveness
of presidential State of the Union addresses.
To estimate the impact of out-party rebuttals on the effectiveness
of a president's State of the Union address to Congress, I adopt a
measure proposed by Edwards (2003). I calculate the impact of the
president's State of the Union address as the difference in public
support as reflected by Gallup opinion surveys immediately before and
after the delivery of the State of the Union. I argue that the
president's effectiveness is mitigated by the effectiveness of the
out-party's response to his address. Therefore, effective outparty
rebuttals will curtail increases in presidential approval that can be
attributed to the delivery of his State of the Union message.
Contrary to the view that presidential delivery of public addresses
enhances support for the chief executive (Kernell 1997), presidential
approval generally drops following the State of the Union message.
Analysis of Gallup polls before and after the State of the Union suggest
presidential approval falls 0.11 percentage points (standard deviation =
3.79; n = 35) following the address. Over the period of my study,
President Johnson fared worst after his delivery of the 1968 State of
the Union: his approval dropped 8 percentage points. By contrast,
President Clinton registers the largest increase in approval (+7
percentage points) following his State of the Union address in 1996.
To what extent is the effectiveness of presidents' State of
the Union messages linked to the opposition party's selection of
representatives to respond to the address? I begin my analysis to
examine this question by comparing the average change in presidential
approval by isolating groups of rebutters comprised exclusively of
senators, governors, or members of the U.S. House. These data, presented
in Table 5, suggest House members most effectively rebut presidential
State of the Union messages. The average shift in presidential approval
that can be attributed to the delivery of the State of the Union is
-2.60 percentage points when the rebuttal group is comprised exclusively
by members of the U.S. House. Rebuttals delivered exclusively by
governors are associated with an average shift of +0.67 percentage
points in approval, and those delivered only by senators result in an
average shift of +1.38 percentage points in presidential approval. These
analyses suggest senators fail to effectively rebut presidential State
of the Union addresses. This may be one reason out-parties are turning
to Senators less frequently as responders.
For a more systematic treatment, I develop and estimate an
empirical model to explain changes in presidential approval following
the State of the Union address for the period 1966-2006. The results are
presented in Table 6. To minimize the effects of outliers, I employ
robust regression. The dependent variable in my analyses is the change
in presidential approval registered by the Gallup poll immediately
before and after delivery of the State of the Union. I examine the
impact of the following variables on changes in presidential approval
levels: total number of rebutters, the proportion of senators, governors
(House members are the excluded category), and women selected to rebut,
rebutter ideology and rebutters' ideological distance from the
incumbent president. I also control for the out-party.
The results of the model suggest the size of the group of rebutters
does influence the effectiveness of the president's message.
Controlling for the other variables in the model, the coefficient
suggests the larger groups of out-party rebutters are likely to only
enhance presidential approval. Thus, smaller groups of rebutters dampen
the effectiveness of the president's address and strengthen the
out-party's response. It is conceivable that smaller groups of
respondents present a clearer, more cohesive rebuttal compared to a
larger and potentially more disparate group. The estimates suggest the
change in the president's approval grows by nearly half of a
percentage point for each additional rebutter included in the out-party
response team.
The results of the model show that as the proportion of U.S.
senators in the group of out-party rebutters increases, relative to the
proportion of U.S. House members in the group (the base category),
presidential approval is likely to rise. This finding supports my
initial observation (described above) that senators do not appear to
deliver effective rebuttals to the presidential State of the Union
address compared with governors or members of the U.S. House. The
proportion of women in the out-party rebuttal group does not appear to
be related to changes in presidential approval.
My empirical model also investigates the impact of the ideological
composition of the group of out-party rebutters (see Table 7). My
measure of rebutter ideology is the absolute value of the average
ideology of the group of rebutters. I use DW-Nominate scores to
determine ideology. (6) I also construct a measure of the distance
between the ideology of the president delivering the State of the Union
address and the ideology of the rebuttal group. Although the overall
ideology of the rebutters does not exert influence on the effectiveness
of the president's State of the Union address, my findings suggest
that out-parties benefit by designating respondents who are
ideologically distant from the president.
Conclusion
This study presents the first systematic examination of out-party
responses to the president's State of the Union message. I present
evidence to describe patterns in the selection of out-party respondents
to the annual message over a period of four decades. I also investigate
how choices about out-party response to the president can curtail
presidential efforts to boost public support via direct, televised
communications with Americans. My analyses suggest the out-party's
decisions about which members will represent the party to deliver a
response to the president's message affect the overall
effectiveness of the president's message. I argue that
incorporating elements of out-party response to models of presidential
leadership as it relates to moving public opinion would provide a more
complete view of the complex process by which the public reacts to
presidential messages. At the very least, out-party responses to the
president, which have largely escaped scholarly scrutiny, have the
potential to be consequential as mitigating factors in this process.
Appendix
Year Response
1966 Senator Everett Dirksen (R-IL)
Rep. Gerald Ford (R-MI)
1967 Sen. Everett Dirksen (R-IL)
Rep. Gerald Ford (R-MI)
1968 Senator Thomas Kuchel (R-CA)
Senator Charles Percy (R-IL)
Senator Howard Baker (R-TN)
Senator Hugh Scott (R-PA)
Senator John Tower (R-TX)
Senator Peter Dominick (R-CO)
Senator Robert R Griffin (R-MI)
Senator George Murphy (R-CA)
Rep. William Steiger (R-WI)
Rep. Gerald Ford (R-MI)
Rep. Richard Poff (R-VA)
Rep. George Bush (R-TX)
Rep. Robert Mathias (R-CA)
Rep. Charlotte Reid (R-IL)
Rep. Albert Quie (R-MN)
Rep. Melvin Laird (R-WI)
1970 Senator William Proxmire (D-WI)
Senator Mike Mansfield (D-MT)
Senator Henry "Scoop" Jackson (D-WA) *
Senator Edmund Muskie (D-ME) *
Rep. Patsy Mink (D-HI)
Rep. John McCormack (D-MA)
Rep. Donald Fraser (D-MN)
1971 Senator Mike Mansfield (D-MT)
1972 Senator William Proxmire (D-WI)
Senator Frank Church (D-ID)
Senator Thomas Eagleton (D-MO)
Senator Lloyd Bentsen (D-TX)
Rep. Leonor Sullivan (D-MO)
Rep. John Melcher (D-MT)
Rep. John Brademas (D-IN)
Rep. Martha Griffiths (D-MI)
Rep. Ralph Metcalfe (D-IL)
Rep. Carl Albert (D-OK)
Rep. Hale Boggs (D-LA)
1974 Senator Mike Mansfield (D-MT)
1975 Senator Hubert Humphrey (D-MN)
Rep. Carl Albert (D-OK)
1976 Senator Edmund Muskie (D-ME)
1978 Senator Howard Baker Jr. (R-TN) *
Rep. John Rhodes (R-AZ)
1979 Senator Howard Baker Jr. (R-TN)
Rep. John Rhodes (R-AZ)
1980 Senator Ted Stevens (R-AK)
Rep. John Rhodes (R-AZ)
1982 Senator Donald Riegle (D-MI)
Senator James Sasser (D-TN)
Senator Robert Byrd (D-W.Va)
Senator Edward Kennedy (D-MA)
Senator Gary Hart (D-CO) *
Senator Paul Sarbanes (D-MD)
Senator J. Bennett Johnston (D-LA)
Senator Alan Cranston (D-CA) *
House Speaker Thomas P O'Neill (D-MA)
Rep. Albert Gore Jr. (D-TN)
1983 Senator Robert Byrd (D-WV)
Senator Paul Tsongas (D-MA)
Senator Bill Bradley (D-NJ)
Senator Joe Biden (D-DE)
House Speaker Thomas P. O'Neill (D-MA)
Rep. Tom Daschle (D-SD)
Rep. Barbara Kennelly (D-CT)
Rep. George Miller (D-CA)
Rep. Les AuCoin (D-OR)
Rep. Paul Simon (D-IL)
Rep. Timothy Wirth (D-CO)
Rep. W.G. "Bill" Hefner (D-NC)
1984 Senator Joe Biden (D-DE)
Sen. David Boren (D-OK)
Senator Carl M. Levin (D-MI)
Senator Max S. Baucus (D-MT)
Senator Robert Byrd (D-WV)
Senator Clairborne Pell (D-RI)
Senator Walter Huddleston (D-KY)
House Speaker Thomas O'Neill (D-MA)
Rep. Dante B. Fascell (D-FL)
Rep. Tom Harkin (D-IA)
Rep. William Gray(D-PA)
Rep. Barbara Boxer (D-CA)
1985 Gov. Bill Clinton (D-AK)
Gov. Bob Graham (D-FL)
House Speaker Thomas P. O'Neill (D-MA).
1986 Senator George Mitchell (D-ME)
Rep. Thomas Daschle (D-SD)
Rep. William Gray (D-PA)
Gov. Charles Robb (D-VA)
Lt. Gov. Harriett Woods (D-MO)
1987 Senator Robert Byrd (D-WV)
House Speaker Jim Wright (D-TX)
1988 Senator Robert Byrd (D-WV)
House Speaker Jim Wright (D-TX)
1989 Senator Lloyd Bentsen (D-TX)
House Speaker Jim Wright (D-TX)
1990 House Speaker Tom Foley (D-WA)
1991 Senator George Mitchell (D-ME)
1992 House Speaker Tom Foley (D-WA)
1994 Senator Robert Dole (R-KS)
1995 Governor Christine Todd Whitman (R-NJ)
1996 Senator Robert Dole (R-KS)
1997 Rep. J.C. Watts (R-OK)
1998 Senator Trent Lott (R-MS)
1999 Rep. Jennifer Dunn (R-WA)
Rep. Steven Largent (R-OK)
2000 Senator Susan Collins (R-ME)
Senator William Frist (R-TN)
2002 Rep. Richard Gephardt (D-MO)
2003 Governor Gary Lock (D-WA)
2004 Senator Tom Daschle (D-SD)
Rep. Nancy Pelosi (D-CA)
2005 Senator Harry Reid (D-NV)
Rep. Nancy Pelosi (D-CA)
2006 Governor Timothy Kaine (D-VA)
Source: U.S. House of Representatives, Office of the Clerk.
http://clerk.house.gov/histHigh/Special_Exhibits/stateUnion.html.
* Presidential candidate in cycle that follows.
References
Baum, Matthew, and Samuel Kernell. 2001. "Economic Class and
Popular Support for Franklin Roosevelt in War and Peace." Public
Opinion Quarterly 65: 218-23.
Bond, Jon, and Richard Fleisher, eds. 2000. Polarized Politics:
Congress and the President in a Partisan Era. Washington, DC: CQ Press.
Brace, Paul, and Barbara Hinkley. 1992. Follow the Leader. New
York: Basic Books.
Cohen, Jeffrey. 1997. Presidential Responsiveness and Public
Policy-Making. Ann Arbor: University of Michigan Press.
Edwards, George C. 2003. On Deaf Ears: The Limits of the Bully
Pulpit. New Haven, CT: Yale University Press.
Erikson, Robert, and Christopher Wlezien. 1999. "Presidential
Polls as a Time Series: The Case of 1996." Public Opinion Quarterly
63: 163-77.
Kernell, Samuel. 1997. Going Public, 3rd ed. Washington, DC: CQ
Press.
Neustadt, Richard. 1960. Presidential Power: The Politics of
Leadership. New York: Wiley.
Poole, Keith, and Howard Rosenthal. 1997. Congress: A
Political-Economic History of Roll-Call Voting. New York: Oxford
University Press.
Ragsdale, Lyn. 1984. "The Politics of Presidential
Speechmaking, 1949-1980." American Political Science Review 78:
971-84.
Simon, Dennis, and Charles Ostrom Jr. 1989. "The Impact of
Televised Speeches and Foreign Travel on Presidential Approval."
Public Opinion Quarterly 53: 58-82.
COSTAS PANAGOPOULOS
Fordham University
(1.) This summary is adapted from details available from the U.S.
House of Representatives: Office of the Clerk,
http://clerk.house.gov/histHigh/Special_Exhibits/stateUnion.html.
(2.) In the six years rebuttals were not offered (1969, 1973, 1977,
1981, 1993, and 2001), presidential inaugurations occurred. In 1981,
President Ronald Reagan offered a speech to Congress on economic
recovery, but no rebuttal was given. In 1989, President George H. W.
Bush was elected to office and offered a special message to Congress,
"Building a Better America;" House Speaker Jim Wright (D-TX)
and Senator Lloyd Bentsen (D-TX) responded on behalf of the Democratic
Party. In 1993, President Bill Clinton offered a budget address after
his inauguration. In 2001, President George W. Bush also delivered an
annual budget message, but no response was delivered. In 2005, after his
reelection Bush delivered a State of the Union address.
(3.) I note these partisan differences, but I underscore that they
do not achieve significance at conventional levels.
(4.) http://www.voteview.com.
(5.) I employ lowess smoothing procedures to detect patterns over
time. Lowess (locally weighted scatter plot smoothing) creates a new
value for each time point based on the results of regressions using a
designated number of surrounding data points. Predictions from these
regressions are weighted based on their temporal distance from the point
in question to generate the new value (Erikson and Wlezien 1999). For
simplicity, figures and additional details are not presented but
available upon request.
(6.) DW-Nominate scores for ideology are not available for
governors. These cases are excluded from the analysis.
Costas Panagopoulos is an assistant professor of political science
and director of the Center for Electoral Politics and Democracy at
Fordham University.
TABLE 1
Descriptive Statistics: Mean Levels of Rebutter
Characteristics, 1966-2006
Mean Std. Dev. Min Max
Total 3.31 3.95 1 16
Senators 1.60 2.16 0 8
Senators (%) 0.48 0.36 0 1
House 1.51 2.16 0 8
House (%) 0.40 0.34 0 1
Governors 0.20 0.53 0 2
Governors (%) 0.12 0.30 0 1
Women 0.34 0.54 0 2
Women (%) 0.11 0.22 0 1
N = 35.
Note: In 1986, Lt. Governor Harriett Woods (D-MO)
was part of the Democratic group that delivered the
response to President Ronald Reagan's State of the
Union address. For the purposes of this study, Woods
is included in the analyses as a governor.
TABLE 2
Partisan Differences in Rebuttals: Mean Levels of Rebutter
Characteristics, 1966-2006
OUT-PARTY
Republican Democrat
Total 2.69 3.68
(4.03) (3.95)
Senators (Total) 1.39 1.73
(2.06) (2.25)
Senators (%) 0.54 0.45
(0.38) (0.36)
House (Total) 1.23 1.68
(2.13) (2.21)
House (176) 0.39 0.41
(0.36) (0.33)
Gov. (Total) 0.08 0.27
(0.28) (0.63)
Gov. (176) 0.08 0.14
(0.28) (0.32)
Women (Total) 0.31 0.36
(0.48) (0.58)
Women (176) 0.16 0.08
(0.31) (0.15)
N 13 22
Note: Standard deviations in parentheses. Differences
across categories are not significant at conventional levels.
TABLE 3
Partisan Differences in Rebutter Ideology (Mean), 1966-2006
OUT-PARTY
Republican Democrat
Rebutter Ideology (Abs. 0.331 0.389 **
value of Mean) (0.031) (0.017)
Distance from Pres. 0.781 0.897 ***
Ideology (0.029) (0.017)
N 12 20
Note: Standard deviations in parentheses. Differences across
categories significant at ** p <.05 and *** p <.01 levels.
TABLE 4
Differences in Rebuttals by Presidential Election Year: Mean
Levels of Rebutter Characteristics, 1966-2006
No Pres. Election Pres. Election
Total 2.64 5.00 **
Senators (Total) 1.20 2.60 **
Senators (%) 0.44 0.60
House (Total) 1.20 2.40 *
House (%) 0.40 0.41
Gov. (Total) 0.28 0.0 *
Gov. (%) 0.16 0.0 *
Women 0.24 0.60 **
Women (%) 0.10 0.13
N 25 10
Note: Differences across categories are significant at * p <.10 and
** p < .05 levels.
TABLE 5
Partisan Differences in Rebutter Ideology (Mean) by Election
Year
OUT-PARTY
Republican Democrat
Presidential 0.258 -0.388
(0.074) (0.072)
N=4 N=6
Nonpresidential 0.367 -0.389
(0.104) (0.080)
N=8 N=14
Note: Standard deviations in parentheses.
TABLE 6
Average Change in Approval by Rebuttal Group
Rebuttals Average Change N
Senator(s) Only +1.38 8
(4.66)
House Member(s) Only -2.60 5
(3.36)
Governor(s) Only +0.67 3
(1.52)
Note: Standard deviations in parentheses.
TABLE 7
Impact of Out-Party Responses on Change in Presidential
Approval Following State of the Union Address, 1966-2006
Dependent variable: Change in Presidential Approval
Before and After State of Union (Gallup)
Independent variable
Total Rebutters .48 ***
(.18)
Senators (%) .04 *
(.02)
Governors (% ) .03
(.05)
Female (%) .01
(.05)
Rebutter Ideology 19.20
(12.85)
Distance from Presidential Ideology -54.08 ***
(16.12)
Out-Party Democrat 1.22
(1.78)
Year .30 ***
(.10)
Constant -558.73 ***
(184.81)
32
Note: Robust regression. Standard errors in parentheses. Estimates
significant at *** p < .01, ** p < .05 and * p < .10 levels,
two-tailed.
