Speakers and the Spotlight

Abstract

Previous scholarship has described how members of Congress struggle to attract media attention. The Speaker of the House, however, is uniquely positioned to drive media agendas. That position may be conditioned though by the framework in which the Speaker operates, shocks to the system of leadership in the House, and the process of legislating. We inquire into the conditions under which the Speaker of the House receives increased or decreased attention from newspapers, network newscasts, and cable newscasts. We also examine the contraction and expansion of the gap between levels of presidential and Speaker coverage. We find that Speaker coverage is often shaped by increases in House polarization, changes in which party holds the position, Speaker scandal, and turning bills into laws.

Keywords

media coverage Speaker of the House polarization leadership change scandal

The public’s understanding of the process of governing is largely shaped by what is presented to them through the media. Americans have neither the time nor the desire to follow each and every development taking place in the legislative, executive, and judicial branches. They therefore delegate information gathering to newspapers, television networks, and increasingly the Internet. We know though that newspapers have only so many inches of column space and that television news programs have only so many minutes to which they devote attention to what is taking place in Washington. Some issues will get covered and some will not. Some individuals will get covered and some will not. On a daily basis, Americans will receive information on a small fraction of the developments taking place in the federal government, and what comprises that fraction will be shaped not by randomness, but instead by the choices journalists make and the norms journalists have developed over time concerning what is newsworthy and what is not (Graber, 2010). One such norm is that the President gets covered with great intensity, while members of Congress struggle to gain attention.

The President is a single individual with a single agenda, while Congress is made up of 535 individuals with at some times similar and at other times wildly cross-cutting agendas. Executive action is seen as contained, decisive, and often dealing with outputs, while congressional action is often sprawling, drowning in detailed processes, dealmaking, and dealbreaking. It is no surprise to many media scholars, therefore, that the President gains the attention of the media (Edwards & Wood, 1999). Although, to some extent, increases in media fragmentation and negativity in coverage may have weakened the President’s grasp on setting the agenda of journalists (Cohen, 2008, 2009), the President still receives more coverage than Congress does, longer coverage than Congress does, and more prominently featured stories than Congress does (Graber, 2010). When Congress is covered, said coverage often lacks coherent narratives over time (Hess, 1991). The legislative branch is often only able to garner media attention when members of Congress are campaigning (Robinson & Appel, 1979). Additionally, Congress is often portrayed much more negatively than the executive branch (Robinson & Appell, 1979; Rozell, 1996). Individual members do receive coverage (Patterson & Caldeira, 1990), but said members are forced, as Jamieson (1988) describes, to compete for what is left over after the President has received his share of attention from the media. Members compete with those in their chamber and they compete with those across chambers (a battle, as Hess, 1986, notes, that benefits a handful of “insider” senators more often than not). Individual members are also forced to find a way to place themselves amidst the incoherent narrative (Hess, 1986), to be a media entrepreneur advancing their own narrative (Kedrowski, 1996), or to fill a role (preferably a powerful one) toward which the media might gravitate (Hess, 1986; Kuklinski & Sigelman, 1992; Squire, 1988). Not every member of Congress is created equal though in this battle over scraps of attention from journalists. Some members of Congress are, as Arnold (2004) puts it, “consequential” (p. 2). Clearly, one of the most consequential individuals in Congress is the Speaker of the House.

Since the reforms of the 1970s that weakened the committee system and strengthened leadership powers, the Speaker of the House has become increasingly meaningful in terms of shaping the legislative process and the outcomes that flow forth (Rohde, 1991; Sinclair, 1998). The Speaker makes decisions as to what bills are heard in committees and what bills make it to the floor for votes. The Speaker ensures that the day to day operations of the House run as smoothly as possible (and is called upon when said business runs into roadblocks). Moreover, the Speaker represents the philosophy and actions of the majority party in his or her chamber. In essence, therefore, Speakers balance legislative, institutional, and partisan agendas in a way few other individuals within Congress do (Green, 2010). This balancing act gives Speakers an advantage in gaining a place within the aforementioned congressional narrative. Such power and responsibility should be especially attractive to journalists given their tendency to seek out the most authoritative sources possible (Sigal, 1986). We should expect the Speaker to face and attract the national media spotlight in a way that the typical member of Congress does not (Cook, 1989; Hess, 1986).

We also know that Speakers, like many other leaders, attempt to pull the spotlight towards themselves. For the past several decades, Speakers have become increasingly cognizant of the role the media might play in setting the stage for legislative victory and political popularity. Winning the competition for coverage (against fellow members of the House and Senate as well as against the President) is essential for Speakers to hold onto and wield power in their chamber, as well as win ideological and legislative battles across chambers and branches. Winning the competition might also, as Sellers (2010) describes in examining motivations to shape coverage and agendas, help “affect the progression and outcome” (p. 3) of policy debates taking place not just within government, but in the public at large. Leaders like the Speaker want coverage because it allows them to set agendas within Congress, increase issue salience amongst fellow members of Congress as well as the electorate, and move the debate forward once items are on the agenda (Sellers, 2010). Such media victories not only potentially get policy ideas turned into laws in the short term, but they may also have long-term benefits in terms of positive evaluations for both the politician shaping the coverage as well as said politician’s party (Sellers, 2010). As Peters (1990) notes, Speaker Tip O’Neill felt from the beginning of his speakership that nurturing his image was an important task alongside managing the House or winning votes, and that completing such a task might set him apart from previous Speakers. Following the elections of 1980 that turned Democrats out of power in both the White House and the Senate, O’Neill embarked on a strategy to make himself more public than before (Cook, 1989). Speaker Wright was known for holding regular briefings in his office that allowed journalists access not just to him but also to other key Democrats in his caucus (Cook, 1989). Even after taking power in the House, Speaker Gingrich continued the public message building and disseminating strategies that helped him rise to the speakership (Sinclair, 2000), so much so that he made himself “a symbol of the Republican program and a natural object of Democratic attack” (Peters, 1999, p. 59), perhaps hastening his downfall. Speaker Pelosi recognized a changing media landscape and as such, attempted not only to promote her efforts in traditional locations such as newspapers and network television stations, but also to manage her message in more modern fora like cable networks, talk radio, the blogosphere, and late night comedy programming (Peters & Rosenthal, 2010). Speakers are not just solo actors in the struggle to gain publicity for their efforts; they often also play a key role in broader party promotional campaigns (Sellers, 2010).

These efforts, however, are not guaranteed to draw a constant spotlight. The speakership appears to be, as Harris (1998) and Highton (2002) note, more public a position than it once was. As Cook (1989) and Kedrowski and Gower (2009) illustrate, this has translated into more media attention for leaders in general when compared to rank and file members and for specific Speakers like Gingrich and Pelosi when compared to similar Speakers in the past. Speakers may, like other party leaders, attempt to “win” media attention (Sellers, 2002), but such attempts do not guarantee victories. The ability of this more public speakership to translate into attention from traditional sources of information such as newspapers and television newscasts may be conditioned by the situations in which Speakers operate and the norms that shape what journalists are more or less likely to cover. The purpose of this research is to determine under what circumstances the media focus more or less of their attention on the Speaker of the House. More broadly, we are also asking what about the Speaker might capture the attention of journalists. Is coverage shaped by the potentially partisan and polarized framework in which a Speaker operates within government? Is coverage shaped by shocks to the system of governing in the House, such as changes in who holds power and who participates in scandalous behavior? Or, is coverage driven by legislative outcomes within government and potential flashpoints within the process of turning bills into laws? Moreover, given the competition for coverage amongst politicians, we inquire into how the framework, shocks to the system, and the legislative process might explain the ever-expanding and contracting gap between presidential coverage and Speaker coverage over time, allowing for a characterization of moments at which the Speaker appears to be more of a rival to the President for the media’s attention.

To answer such questions and explain what conditions the ebb and flow of media attention paid to the Speaker of the House, we proceed in the following manner. We begin with a discussion of the lessons existing research has taught us about political coverage in general, namely the norms that journalists utilize to determine what ends up in print or on air and what does not. We speculate as to which elements of the speakership might meld with these coverage norms, causing media attention toward the Speaker to rise and fall over time. We look at national newspaper, network television coverage, and cable television coverage of the Speaker over three-plus decades. Through a series of error correction models examining the volume of print and broadcast stories of Speakers from 1977 through 2010, we find that the media spotlight is best explained by polarization within the House, changes in which party holds power, Speaker scandal, and the extent to which bills become laws.¹ We also use our battery of predictors to examine the ever-intense but often fleeting (at least in terms of volumes of coverage) rivalry for attention between the President and the Speaker as well. We again discover that polarization, changes in power, Speaker scandal, and lawmaking are essential to explaining when the Speaker fares better or worse in competition for the spotlight.

Explaining Speaker Coverage: Framework, Shocks, and Legislating

It is no surprise, given the aforementioned benefits in terms of policy, party, and popularity, that Speakers would attempt to be media entrepreneurs and win coverage. Given the power inherent to the position, Speakers should be successful at this entrepreneurism above and beyond the average member’s attempts (Kedrowski, 1996). Such attempts do not always dovetail nicely with a world in which those covering Congress may have a wholly different conception of what is newsworthy and what is not. Politicians seeking to get their messages across to the general public might hope that the media act, as Groeling (2010) describes, like a “conveyer belt faithfully delivering the views of elites” (p. 45), but such hopes are often confounded by the beats journalists keep and the norms journalists adopt to determine what is newsworthy and what is not (Althaus, 2003). Research on the congressional beat notes, that like many other political beats, coverage of Congress as a whole is generally driven by “exciting, novel, or controversial topics” and not “recurrent complex and mundane problems” on the agenda (Graber, 2010, p. 249). This type of coverage of Congress makes sense given broader journalistic tendencies to focus on novelty, familiarity, and conflict, while simultaneously often neglecting to follow major problems specifically and public policy in general (Graber, 2010; Groeling, 2010; Patterson, 2000). Speakers tirelessly work to make themselves, their positions, and their party fit the norms journalists follow; this work may be undermined or enhanced, we argue, by three factors: the presence or absence of partisan and ideological conflict within the House or the broader political system (a battery of variables we call framework), shocks to the system of governing in the House and/or by the Speaker (variables we call shocks), and legislative action on the part of the House and Congress more broadly (variables we call legislating). We now explain more specifically why and how this might be the case.

Framework and the Speaker

We know that Speakers have a wide variety of tools at their disposal to move legislation (and their own interests) forward, but we also know that to some extent, their ability to move strategically is either enhanced or undermined by the state of debate within their caucus and the opposing caucus, as well as the length of the leash the public has given Speakers to govern and wield power. As discussed earlier, we know that Speakers attempt to draw the spotlight toward them, but such efforts run into roadblocks like those discussed by Patterson (1994) and Weaver (1972), who argue that the media are increasingly interested in the strategic (and adversarial) “game” taking place between politicians attempting to advance their agendas. In essence, Speaker efforts to get attention and get things done should be affected (for better or worse) by similar efforts undertaken by others in power. This clash of efforts fits well with established media norms of seeking to cover conflict and subsequent negativity (especially when it comes to Congress, according to authors like Hibbing & Theiss-Morse, 1998; Rozell, 1994; and Tidmarch & Pitney, 1985). We should expect this push and pull between Speakers and the broader framework to potentially serve as fodder for a media interested in how key players like the Speaker act and interact. We anticipate the framework potentially driving Speaker coverage in three ways: through polarization within the House over which the Speaker runs, through divided government between control of the speakership and the President, and through divided government between control of the House and Senate.

One element of the strategic framework in which Speakers operate that should be of interest to the media is the potential for greater ideological extremity in the House leading to deeper divides (and as such, interesting and newsworthy conflict). Fiorina, Abrams, and Pope (2006) illustrate how the media consistently express interest in covering polarizing individuals writ large. We also know that the media have a distinct interest in extremism within Congress; as Kuklinski and Sigelman (1992) show, extremists within the Senate get more attention than their more moderate counterparts. We argue that moments of conflict and moments of extremism have the potential to draw Speakers into the discussion. Speakers will be forced to manage extremism or carry the flag for their side against the ever-polarizing opposition. Given the media’s continued and increasing attention to political gamesmanship and the potential for Speakers to either stoke the flames of ideological contention or serve as a force to bring legislators with differing viewpoints together to solve problems, we might expect situations in which parties and ideologies are prone to clash to be moments in which Speakers are the subject of newspaper and television coverage. One way we might measure this is to capture the extent to which parties in a chamber act differently (i.e., vote differently). When parties act together, the potential story being told might not spark coverage of Speakers in the way that we might expect if the parties were strongly opposed and combative. The further apart the parties are ideologically, the easier it might be to cover Speakers in a way that fits existing narratives about conflict, gamesmanship between partisan spokesmen (one of whom often is the Speaker), the general negativity in politics, and the inability for Congress to accomplish goals in the face of partisan bickering. We expect polarization to bring with it added attention to the Speaker. We include in our models a variable capturing polarization between the two parties; House Polarization is the distance between the average House Republican and the average House Democrat on Poole and Rosenthal’s DW-Nominate spectrum. Data on party averages is available through Poole and Rosenthal’s database.²

We are also interested in how what is going on between the House and the rest of the federal government might shape media attention paid to Speakers. One path by which other branches might drive coverage of the Speaker is through the presence of divided government. Harris (1998) notes that divided government in the late 1960s and early 1970s may have spurred media opportunities for Speakers past. These opportunities might be further exacerbated in times when control of the House, the Senate, and the White House seems increasingly fluid (or much more fluid than it was when Democrats had a stranglehold on Congress). When parts or the whole of Congress are controlled by a different party than the White House, the potential exists for highly visible (and highly interesting from a media perspective) conflict in the process of lawmaking. We know that coverage of Congress in general is often laden with discussion of interbranch and intrabranch sparring (Groeling, 2010; Morris & Clawson, 2005). Those individuals driving that conflict (or, at times, coming together across partisan divides to compromise) might be drawn into the spotlight as a result. They have the potential to become natural foils against hostile and powerful partisan opposition, be it on the other side of Congress or in the White House (Groeling, 2010). The media, whose interest in Congress is often driven by the negativity inherent in conflict, should pick up on the flashpoints between parties within and across branches. In times where the House and the White House or the House and the Senate are controlled by the same party, such potential for conflict is quelled to a larger extent. In such situations, Speakers might, as Groeling (2010) puts it in his discussion of congressional members of the presidential party, spend “life in the shadows” (p. 97). We conceive of divided government in two distinct ways in our models. We include a variable (House/Presidency Division) to account for quarters in which a Speaker’s party affiliation was different from that of the President’s, expecting that these time periods will be ones in which Speakers will receive higher levels of attention. We also include a variable (House/Senate Division) to capture quarters in which a Speaker’s party affiliation was different from that of the Senate Majority Leader’s. We expect that this type of division may create newsworthy tension between the House and the Senate in the process of creating legislation passable in both chambers.

Shocks and the Speaker

The decisions the media make in separating what to cover and what to ignore are driven in part by “economic pressures to generate large audiences” (Graber, 2010, p. 97). The conflict in polarization and divided government (and the strategy the Speaker utilizes in dealing with this framework) may be one such way to draw in the public, but it is not necessarily the only way. Novelty surrounding the speakership may provide another path for journalists to pursue in their effort to build an audience. New situations that shock the system in which the House (and, in turn the Speaker) operates may move the media’s attention toward the Speaker. We conceive of three types of shocks that the media might find novel: changes in party control, Speaker scandals, and scandals involving multiple members of the House.

Changes in which party holds the speakership might bring with them changes in levels of Speaker coverage. We should expect the idea of change to be inherently interesting to a political media looking for new angles on old stories. Whereas a Speaker of one party stepping aside for a new Speaker of the same party might often be a situation in which continuity in leadership style or partisan agenda reigns, a change in which party runs the House is guaranteed to be historic in nature and assuredly brings with it the novelty journalists seek to present. Journalists, pundits, and politicians alike will wonder what a new Speaker representing control of the House by a different party will bring to the position, to policy development, and to the House itself. They will speculate as to how a new type of control and leadership will affect other major players within government. We also might expect a certain amount of biographical and historical coverage capturing who this new Speaker is, helping familiarize the public with an individual who has been elevated to become a major player that will be sure to be a part of future coverage, building a foundation of narratives on which coverage of Speakers over subsequent quarters could be justified. As such, we would expect these historic changes in the speakership to drive coverage upward. We include a dichotomous variable (Change In Control) in our models that account for quarters in which control of the House changes hands from one party to the other.

We are also interested in how potentially scandalous actions of both Speakers and other legislators might shape coverage of Speakers. Scandals fit well with the search for novelty, conflict, and negativity intrinsic to the media’s approach to politics in modern times. We know that, in general, scandal coverage has increased over time (Lichter & Amundson, 1994). We also know that ethics issues might drive additional coverage of those specifically caught up in said issues (Arnold, 2004). Our models examine links between scandal and Speaker coverage in two ways, looking for effects of scandals directly involving the Speaker and scandals involving fellow House members. We expect scandals in general to push the spotlight onto leaders like the Speaker. When Speakers themselves are caught up in scandals, we should expect the media to pounce on the story, questioning how the scandal will affect the Speaker’s ability to lead and if the Speaker will be able to continue to hold onto his or her position. When members of Congress are caught up in scandals, Speakers are often times forced to react, be it through speaking out or taking steps to reprimand those scandalized. We include a dichotomous variable (Speaker Scandal) in each model that measures quarters in which a Speaker is the subject of a House Ethics Committee investigation.³ Also included in each model is a dichotomous variable (House Scandal) measuring the presence or absence of a scandal that affects the careers of multiple members within the House simultaneously.⁴

Legislating and the Speaker

As Patterson (1994) argues, in the wake of the media’s shift to covering strategy over substance, Americans still do want to know about issues and want to know how outcomes will affect them. While this strain of coverage may have declined, it has not disappeared. We know it has not disappeared on other realms, such as campaign coverage (Patterson, 1994). We also know it has not disappeared from the attention afforded to Congress in general by journalists; Hess (1991), Morris and Clawson (2005), Robinson and Appel (1979), and Tidmarch and Pitney (1985) show that parliamentary action does get attention from the mass media. Introduction of legislation (including sponsorship and cosponsorship), committee hearings and testimony, debates on the House and Senate floor, final votes, and conference committees all draw coverage in varying amounts. Media coverage is unlikely to take the form of a “play-by-play” mentality describing legislative developments as they happen; instead, the media seem interested in legislative activity that brings with it a sense of finality, explaining what has been done (Graber, 2010). The extent to which Speakers take action and bring about results (feeding the appetite for coverage of legislative outcomes, as it were) might shape the extent to which they receive attention from print and television media. Results, we argue, might be measured in two ways: turning bills into laws and standing up for decisions made by overriding vetoes.

One set of actions on the part of fellow legislators that potentially affects media attention afforded the Speaker is the movement of legislation through the legislative process to completion. A productive Congress might garner attention for its leaders, like the Speaker, in ways that an unproductive Congress might not. Legislative outputs could provide a steady stream of fodder that the media might dissect, as well as fodder for assessing the performance of a Speaker. We therefore include a variable (Bills Become Laws) that captures the number of pieces of legislation that were signed into law during each quarter. Archival searches of the Thomas legislative database were utilized to count quarterly levels of bill passage.⁵

Some rare parliamentary actions might also spur the media to pay attention to the role leaders like the Speaker are playing, while also fitting neatly into the journalistic norms of conflict and novelty discussed earlier. One such type of action we feel might draw the spotlight of newspapers and television onto the Speaker is successful attempts to counteract the power of the presidency through veto overrides. Overrides allow Congress to strike back against the President and reaffirm the power of the legislative branch in the legislative process. Overrides are rare, but when they do occur they not only create a new law, but they also send a signal as to the ability of a leader such as the Speaker to marshal his or her chamber to act decisively. We might expect the infrequent and conflict-laden nature of these successful attempts to be newsworthy. We include a measure (Overrides) to account for moments per quarter when Congress is able to override a veto.⁶

What Speakership Coverage Looks Like

We begin our inquiry into explaining the extent to which the media spotlight is on the Speaker with a quick overview of what media coverage of the Speaker of the House looks like across the past 30-plus years. For our purposes here (and to be able to speak about attention from the media in the broadest sense possible), we conceive of media coverage taking three primary forms in recent history: newspaper coverage, network television coverage, and cable television coverage. Internet coverage of Speakers may have taken off in recent years, but newspaper and television attention have been a constant presence across multiple decades. Focusing on traditional media allows us to characterize the predictors of coverage in a more comprehensive manner over time. We start with a look at how print media has covered the Speaker since the late 1970s. We have chosen to examine coverage from both The Washington Post and The Associated Press, allowing us to examine attention afforded the Speaker in a major national publication as well as by a news agency that provides copy that reaches national, state, and local publications. To gather the quarterly volume of this newspaper coverage of Speakers, we use the Lexis-Nexis media database.⁷ For Post coverage, stories in which the full name of a Speaker was included in the list of the Post’s story subjects (via the Lexis-Nexis story subject search) were counted as stories about said leaders. For AP coverage, stories in which the Speaker’s name appears in the headline or the lead paragraph (via the Lexis-Nexis HLead search) were counted as stories about said leaders.⁸

Our series of newspaper stories on the Speaker begin in 1977 and run through 2010. This choice is driven by the depth of the Lexis-Nexis database when it comes to The Washington Post, which dates back to 1977 online. This also explains the choice of the Post over comparable national papers like the New York Times, which does not have stories on Lexis-Nexis going back further than the early 1980s. We collect AP stories back to 1977 as well for the sake of comparability between series. Figure 1 captures coverage of the Speaker of the House in The Washington Post and The Associated Press.⁹

Figure 1.

Note: x axis = time; y axis = number of newspaper stories.

We see in general in Figure 1 that a quarter in which the number of stories rises over 100 is relatively rare, happening several times between early 1982 and early 1985 during the O’Neill speakership, in late 1987 and early 1988 during the O’Neill and Wright speakerships, during the second quarter of 1989 at the end of the Wright and beginning of the Foley speakerships, every quarter between 1995 and 1998 during the Gingrich speakership, during the second quarter of 1999 and 2000 during the Hastert speakership, during the first two quarters of 2007 at the start of the Pelosi speakership, and a handful of times between 2009 and 2010 near the end of the Pelosi speakership. In The Washington Post, Speaker coverage is at its highest during the first quarter of 1995 (where it increases to nearly 6 times the quarterly average of 87). As for The Associated Press, Speaker coverage is also at its highest in the first quarter of 1995 (where it peaks at 7 times the quarterly average of 65).

We also note the incredible similarity between The Washington Post and The Associated Press coverage of the Speaker of the House. The series are correlated at .92, suggesting patterns of reporting on the Speaker are incredibly similar for a major national publication as they are for a news agency with reach across nearly 2,000 papers of all sizes. We do see some points at which the Post and AP diverge, however. Coverage of Speaker Tip O’Neill’s earliest days in office was much more prevalent in The Washington Post than it was by The Associated Press. Coverage of Speaker Wright’s final quarter and Speaker Foley’s first quarter peaked more highly with the AP than it did with the Post. Attention paid to the start of the Gingrich and Pelosi speakerships was also generally higher in the Post than it was with AP. We now move to an examination of Speaker television coverage.

When it comes to television coverage of the Speaker of the House, we rely on the Vanderbilt Television News Archive.¹⁰ Overall, the Archive includes abstracts of news programs run by ABC, CBS, NBC, CNN, FOX, PBS, MSNBC, CSPAN, CNBC, and Univision. Examining Speaker coverage on all networks in total across decades proves somewhat problematic however, given that the Vanderbilt archive begins only in the mid 1990s to include cable networks (well after the birth of CNN) and does not cover these networks above and beyond CNN until the mid 2000s. We therefore create two television coverage variables: Speaker Network Coverage and Speaker Cable Coverage. To capture Speaker Network Coverage, we chose one longstanding network (ABC) at random. Our series of ABC newscasts with stories on Speakers begins in 1977 and runs through 2010. We choose to begin with 1977 to create a series of comparable length to the newspaper series discussed earlier. When it comes to Speaker Cable Coverage, CNN was chosen both because of its more evenhanded reputation (especially in comparison to the reputations MSNBC and FOX News have taken on over time) and because over 15 years of CNN story topics exist in the Archive (giving us a noticeably longer series than we would have had we examined other cable news options). Vanderbilt’s archive chronicles CNN’s newscast running from 6:00 to 7:00 p.m. Eastern time. Our series here runs from late 1995 through the end of 2010 due to the aforementioned data constraints.

In conducting both of these searches, we sought out mentions of the last name of Speakers in news abstracts.¹¹ Human coders validated the data collection process, taking steps to weed out any stories that may not have dealt with Speakers but instead with other individuals with the same last names. The abstracts searched include those for news segments, program introductions, specials, anchor “good night” segments, and evening news; these categories are options offered by the Vanderbilt Television News Archive itself. The two series of television coverage of the Speaker in Figure 2 track the raw number of ABC and CNN newscasts with stories in which the Speaker appeared, was discussed, or was quoted per quarter.

Figure 2.

Note: x axis = time; y axis = number of television stories.

We see that network news coverage of the Speaker remains near or slightly below average (average for ABC being 9 newscasts per quarter) for several quarters in the late 1970s during the O’Neill speakership. Across much of the early to mid 1980s, we then consistently see quarters in which Speaker O’Neill was covered in above average fashion, typically at a rate of 10 to 20 stories per quarter. Between 1987 and 1989 (which saw the departure of O’Neill and the beginning and end of the Wright speakership), we see several quarters in which ABC coverage of the Speaker was well above average; such trends die down in the early 1990s during the Foley speakership. The mid 1990s and the Gingrich speakership are particularly fruitful years for network news coverage of the Speaker, but by mid-1997 attention for the Speaker declines once more and remains well below average until late in the 2000s (with the beginning of the Pelosi speakership). Speaker Television Coverage is at its highest in the first quarter of 1995, when ABC ran 67 newscasts in which Speaker Gingrich was discussed; this period saw over 7 times the quarterly average of attention afforded the Speaker).

Examining the Speaker Cable Coverage series, we see quarters well above average (6 stories per quarter being average) in 1995, 1996, 1997, and 1998 during the Gingrich speakership. Similarly to our series of ABC News coverage, we see several years of below average attention paid to the Speaker between 1999 and 2006 during the Hastert speakership before the late 2000s and the Pelosi speakership brought with them a handful of quarters in which the Speaker once again becomes a topic of CNN’s attention. The series peaks in late 1995 with 35 CNN stories on Speaker Gingrich; this level of attention amounts to nearly 6 times the quarterly average across the series. As with our two newspaper series, we see a remarkable similarity in the extent to which ABC and CNN cover the Speaker; across the 15 years in which examine both series, said series are correlated at .85.

Across both figures capturing our four media coverage variables, we see a handful of strong commonalities. Both of the newspaper series as well as our ABC News series see several points across the early to mid 1980s where Speaker O’Neill receives coverage at rates slightly above the quarterly average. These three series also have well above average peaks in the late 1980s during the Wright speakership. All four series rise noticeably in the mid 1990s during the Gingrich speakership and in the late 2000s during the Pelosi speakership. It is also clear that the Post, AP, and ABC series see little attention paid to the Speaker early in the 1990s during the Foley speakership, and all four series decline sharply early in the 2000s during the Hastert speakership.

Modeling Speaker Coverage

To test the relationship between the framework of governing, shocks to the system, legislative outcomes, and levels of newspaper and television coverage over time, we utilize error correction modeling. Error correction models allow us to determine both short-run (i.e., contemporaneous) and long-run impacts of our independent variables on levels of media coverage over time. As illustrated by authors like DeBoef and Keele (2008) and Kelly (2005), such a modeling choice can be utilized in a variety of situations. In fact, as these authors show, choosing error correction over, for example, distributed lag modeling or other modeling choices, allows for a more general model that does not impose unnecessary and potentially unfounded restrictions on the relationships being examined.¹² Error correction modeling also allows insight into potential relationships beyond simply examining a unit change in an independent variable driving change in a dependent variable during the same time period; we can determine, as Kelly and Enns (2010) put it, the potential for “nearly immediate impact” as well as “effects . . . distributed over time such that the full effect is not felt all at once” (p. 863). We can determine not only if one variable significantly impacts another contemporaneously, but also to what extent across future quarters. Table 1 displays the findings of our initial inquiry into what drives newspaper coverage of the Speaker.

Table 1.

Explaining Newspaper Coverage of the Speaker of the House: An Error-Correction Model.

	Post coverage		AP coverage
	Long run	Short run	Long run	Short run
Predictor	Effects	Effects	Effects	Effects
House polarization	5.05	1656.08*	−19.58	1765.53*
	(23.46)	(369.18)	(24.87)	(395.54)
House/presidency division	9.03	8.59	13.02	14.23
	(7.83)	(20.48)	(8.58)	(21.96)
House/senate division	2.73	−5.03	0.93	−20.33
	(8.13)	(18.75)	(8.70)	(20.12)
Change in control	201.83*	272.26*	196.25*	190.12*
	(50.04)	(38.82)	(53.87)	(41.87)
Speaker scandal	20.99	92.17*	64.44*	122.37*
	(13.86)	(18.91)	(15.76)	(20.29)
House scandal	−2.53	6.76	−5.44	8.26
	(8.69)	(13.30)	(9.37)	(14.28)
Bills become laws	−0.21*	0.02	−0.31*	−0.04
	(0.09)	(0.06)	(0.10)	(0.06)
Overrides	6.26	5.03	4.60	2.65
	(11.56)	(8.27)	(12.42)	(8.90)
Story volume_t-1	−0.35*		−0.47*
	(0.06)		(0.07)
Constant		23.41		44.42
		(22.56)		(24.39)
N	135		135
R-Squared	.71		.64

Note: DVs = number of Speaker newspaper stories per quarter. Both newspaper series run from 1977:1 to 2010:4. Standard errors are in parentheses.

= p < .05.

Table 1 examines the effect of framework, shock, and legislating variables on The Washington Post and The Associated Press coverage of Speakers of the House. Looking first at what drives Post coverage, we find that polarization within the House itself matters. In the short run, the greater the distance between the average House Republican and House Democrat, the higher the level of coverage the Speaker will receive. Given that the DW-Nominate scale on which the variable is based ranges from -1 to 1, a shift upward of one in the distance between average party members would be massive and unlikely; as such, considering how smaller shifts might matter makes more intuitive sense.¹³ If the parties in the House moved 0.1 further apart on the DW-Nominate scale, we would expect the Speaker to receive approximately 165 additional stories in the period where movement occurs. We might take the relationship between polarization and Speaker newspaper coverage to be proof that potential conflict between partisan actors within the framework draws the spotlight toward Speakers expected to be at the core of said conflict, either leading the charge or managing tumult. Speakers must manage the widely different approaches to the political game being played by partisans in the House. We see no long-run component to the significant relationship; polarization only matters in an immediate sense and then disappears in terms of effects.¹⁴

Shocks to the system directly involving the Speaker himself or herself also clearly drive Post coverage of the Speaker. We see positive and significant effects on story volume of a change in party control of the speakership. New Speakers representing new party control (in this study, Speakers Gingrich and Pelosi fit this role) should be incredibly newsworthy; they may bring new personalities, new styles of leadership, and new relationships with fellow members and the media to the table. Change in leadership in general attracts the types of media norms Graber (2010) noted; these changes bring with them a historic element as well. Changes in party control of the speakership may also mean new strategic choices for the media to cover. They may also involve individuals less well known at first from previous Speakers, so a flood of biographical stories might be expected. In essence, newness of change in party control hand in hand with newness of the individual wielding that control wins coverage. This relationship has a contemporaneous component (the quarter in which control shifts sees a boost of approximately 272 Speaker stories) as well as a long-run component. Table 2 teases out that long-run component over the next three quarters.¹⁵

Table 2.

Long-Run Effects on Newspaper Coverage of the Speaker.

Predictor	Newspaper	t+1	t+2	t+3
Change in control	Post	202	141	92
Bills become laws	Post	−0.21	−0.14	−0.09
Change in control	AP	196	104	55
Speaker scandal	AP	64	34	18
Bills become laws	AP	−0.31	−0.16	−0.09

Note: Effects above 1 rounded off to the nearest whole number.

Table 2 illustrates significant long-run impacts on newspaper coverage of the Speaker of the House by The Washington Post and The Associated Press. When it comes to the long-run effects of a change in party control of the House, we see an increase of 202 stories in quarter t+1, 141 stories in quarter t+2, and 92 stories in quarter t+3. This long run significance is not surprising; after all, the impact on legislating and governing of a different party taking over control of the House is something that has the potential to extend longer than the first 3 months in power. Such transitions should continue to provide for interesting copy to journalists on the congressional beat. We should also note that the somewhat small error coefficient (estimated at -0.35) means this effect (and other significant long-run effects of predictors on Post coverage) should dissipate slowly across future quarters.

Another shock to the system of governing in the House that drove Post coverage upward was the presence of a Speaker scandal. Quarters in which Speakers were under ethics investigations saw just over 92 more stories on the Speaker than quarters without. Such a relationship suggests that newspapers were clearly (and unsurprisingly given the modern media’s focus on conflict, the negative, and the potentially salacious) interested in the troubles Speakers Wright and Gingrich faced in grappling with ethics charges. This relationship is strictly short-run in nature however. It has no lasting effect beyond the investigation.

One final relationship to note when it comes to Post coverage is the significant long-run relationship between the process of turning bills into laws and attention afforded the Speaker. This relationship is a negative one, signifying that that more productive Congress is, the less coverage the Speaker receives across future quarters. The long-run impact of this relationship (as seen in Table 2) across time periods t+1, t+2, and t+3 leads to approximately 0.21, 0.14, and 0.09 fewer Speaker stories respectively for every 1 bill that becomes law. Such findings suggest that the media may not be as interested in talking about a Speaker doing his or her job (or the Speaker’s role in legislative productivity) so much as they are interested in talking about periods of divisiveness or changes in the chamber (as seen in the significance of aforementioned predictors). This finding might fit with the previously discussed media norm of journalists being less willing to cover policy.

Moving to columns 3 and 4 of Table 1, we see an incredibly similar picture when it comes to explaining what drives coverage of the Speaker by The Associated Press as we did in looking at coverage of the Speaker by The Washington Post. We see a positive and significant short-run effect of polarization on AP coverage of the Speaker. The more polarized the House is, the more coverage the Speaker receives. An increase in distance between the two parties in the House of 0.1 on the DW-Nominate scale brings forth approximately 177 more stories on the Speaker by The Associated Press. Said effect is slightly larger for the AP than it is for the Post. As with the Post though, there appears to be no long run component to this relationship either.

As with the previous model gauging the effects of shock-based predictors on Speaker coverage in The Washington Post, we see that said predictors are essential to explaining The Associated Press’s focus on the Speaker as well. A change in which party controls the speakership has both short- and long-run positive and significant effects on AP coverage of the Speaker. Quarters in which control of the speakership shifts from one party to the other see an increase of just over 190 Speaker stories. This contemporaneous effect is slightly smaller when it comes to AP coverage in comparison to Post coverage. In the long-run (as illustrated in Table 2), we see an additional 196 Speaker stories in t+1, an additional 104 in t+2, and an additional 55 in t+3. Comparing the long-run dynamics between the Post and AP, we find that the impact of a change in party control of the speakership on coverage over future quarters is stronger with the Post than with The Associated Press. This can be explained by both the stronger effect in t+1 for the Post as well as the higher error coefficient (-0.47) for the AP.

Scandal involving the Speaker also shifts The Associated Press’s spotlight. We see a contemporaneous increase of approximately 122 stories in quarters where the Speaker is the subject of an ethics committee investigation. Unlike with The Washington Post’s approach to scandal, however, with The Associated Press we see a significant long-run component as well. This long-run effect is illustrated further in Table 2. Speaker scandal drives The Associated Press’s attention to the Speaker across future quarters; t+1 sees 64 more Speaker stories, t+2 sees 34 more Speaker stories, and t+3 sees 18 more Speaker stories.

One last predictor that shapes AP attention paid to the Speaker is the process of bills becoming laws. As with The Washington Post, we see a negative and significant long run relationship between lawmaking and Speaker coverage. The more bills that become laws, the less attention The Associated Press pays to the Speaker across future quarters. The last row of Table 2 shows t+1, t+2, and t+3 impacts of -0.31, -0.16, and -0.09 respectively. Comparing this long run effect with that of the Post, we see that in quarter t+1, legislating has a stronger effect on decreasing coverage of the Speaker for The Associated Press, but due to the differing error coefficients, by quarter t+3 the effects are exactly the same; were we to push even further into the future to t+4 and beyond, we would find the impact would be stronger for the Post than for the AP.

Across both newspaper models, we see some commonalities when it comes to what does not drive coverage of the Speaker of the House. Divided government in both forms (division in party affiliation between the Speaker and either the President or the Senate Majority Leader) does not shift the spotlight onto or away from the Speaker. Veto overrides do not lead to higher or lower levels of newspaper coverage of the Speaker. Finally, scandals involving multiple members of the House do not push The Washington Post and The Associated Press to focus more or less on the Speaker than in other quarters.

We now move to a framework, shock, and legislating-based explanation of television coverage of the Speaker on the part of both network newscasts (ABC) and cable newscasts (CNN). We begin with Columns 1 and 2 of Table 3 that capture short- and long-run predictors of network news coverage of the Speaker from 1977 through 2010. Immediately, we see similarities to the previous models attempting to explain newspaper coverage of the Speaker of the House. Once again, polarization matters. The more polarized the House of Representatives is in terms of the distance between parties, the more attention the Speaker receives from network newscasts like those produced by ABC. Such a relationship is not unexpected; as argued earlier, polarization should bring with it conflict attractive to media and the Speaker must find ways to maneuver amidst said conflict. The state of the framework in which the Speaker wields power matters. This relationship has both short- and long-run components (marking the only instance where polarization is significant in the long run). Were Democrats and Republicans in the House to move 0.1 further apart on the DW-Nominate scale, it would lead to slightly over 29 more Speaker stories, an amount well above the quarterly average on ABC. Interestingly, however, the long-run relationship between polarization and ABC coverage of the Speaker is negative. Table 4 displays the nature of this relationship across subsequent quarters.

Table 3.

Explaining Television Coverage of the Speaker of the House: An Error-Correction Model.

	ABC coverage		CNN coverage
	Long run	Short run	Long run	Short run
Predictor	Effects	Effects	Effects	Effects
House polarization	−15.49*	294.33*	7.95	188.54
	(4.79)	(71.73)	(24.86)	(161.80)
House/presidency division	2.03	1.62	1.47	−3.25
	(1.49)	(3.97)	(2.80)	(6.04)
House/senate division	0.30	−0.71	−2.01	−2.29
	(1.57)	(3.64)	(2.79)	(4.75)
Change in control	32.08*	26.57*	13.19	9.91
	(9.84)	(7.59)	(15.31)	(13.60)
Speaker scandal	10.24*	18.91*	8.94*	2.47
	(2.71)	(3.69)	(3.58)	(5.88)
House scandal	−0.76	−0.09	1.27	−0.81
	(1.69)	(2.58)	(2.41)	(6.44)
Bills become laws	−0.06*	−0.01	−0.03	0.01
	(0.02)	(0.01)	(0.03)	(0.01)
Overrides	4.65*	2.48	0.02	−0.56
	(2.26)	(1.60)	(3.44)	(2.42)
Story volume_t-1	−0.82*		−0.88*
	(0.08)		(0.15)
Constant		18.58*		−3.44
		(4.80)		(23.99)
N	135		60
R-Squared	.66		.53

Note: DVs = number of Speaker television stories per quarter. ABC series runs from 1977:1 to 2010:4; CNN series runs from 1995:4 to 2010:4. Standard errors are in parentheses.

= p < .05.

Table 4.

Long-Run Effects on Television Coverage of the Speaker.

Predictor	Network	t+1	t+2	t+3
House polarization	ABC	−16	−3	−0.50
Change in control	ABC	32	6	1
Speaker scandal	ABC	10	2	0.34
Bills become laws	ABC	−0.06	−0.01	−0.002
Overrides	ABC	5	1	0.15
Speaker scandal	CNN	9	1	0.13

Note: Effects above 1 rounded off to the nearest whole number.

The first row of Table 4 illustrates the impact of polarization on ABC coverage of the Speaker across quarters t+1, t+2, and t+3. We see a 1-unit change in polarization leads to 16 fewer stories in t+1, 3 fewer stories in t+2, and 0.5 fewer stories in t+3. Given what has been previously stated about the nature of the polarization variable, however, the real world effects of polarization on ABC coverage of the Speaker are even smaller across future quarters. A more likely 0.1-unit increase would lead to 1.6, 0.3, and .05 fewer stories across the subsequent three quarters following the change. This effect is significant but incredibly small (and incredibly quick to dissipate given an error coefficient of -0.82). The direction of this effect could be explained by strong positive initial interest in changing polarization, followed by slightly decreasing interest when that change in polarization becomes the status quo.

We also see that a change in which party holds the position of Speaker also brings with it network news attention. This shock to the system of governing in the House, where control of the speakership changes hands from one party to the other, brings with it a contemporaneous boost of just under 27 stories. We see that this effect is not just a short-run effect, but that it also spurs coverage across future quarters. New Speakers from parties just taking control receive higher levels of network news attention well after the change in who holds the position has taken place. Impacts across the first year of a new Speaker’s rise to power are small but significant ones. Teasing out the long run effect of a change in Speaker over the next three quarters in Table 4, we would see a t+1 impact of 32 stories, a t+2 impact of 6 stories, and a t+3 impact of 1 story. Such effects are noticeably smaller than they were when it came to newspaper coverage, but the long-run significance of this relationship makes apparent the idea that the novelty and excitement of a change in which party holds this position can attract television news attention and hold it into the future. The newness and strategic elements of a different party controlling the speakership hold sway.

As with our previously discussed newspaper models, Speaker scandal has a positive and significant effect on network news coverage of the Speaker of the House. Both the Wright and Gingrich ethics woes moved network news as well as newspapers to pay added attention. In Table 3, we see that quarters in which the Speaker is the subject of an ethics investigation see a contemporaneous impact of approximately 19 stories with ABC News. These scandals had long-run effects as well. Table 4 illustrates that this impact is nearly halved in quarter t+1 (to 10 additional stories) and disappears rather quickly across the subsequent two quarters (with impacts of 2 and 0.34 stories in t+2 and t+3 respectively). It is clear that scandal involving high profile leaders like the Speaker of the House draws and holds the interest of network news.

Once more, we also see lawmaking have a negative and significant impact on Speaker coverage; once more as well, this impact is solely a long-run impact. The fourth row of Table 4 breaks down this effect across three future quarters and shows that with effects of -0.06, -0.01, and -0.002, such effects on network news are incredibly small and incredibly quick to dissipate. However, it should be noted that this is the third model in a row where we see long-run negative and significant effects of lawmaking on Speaker coverage, suggesting once more that there is something to be said for the idea that the media are less interested in the Speaker the more Congress is turning bills into laws.

Interestingly, however, the network news model delivers the first (and only) evidence that overrides might significantly shape when the media spotlight shines on the Speaker of the House. When the House of Representatives overrides a veto, the Speaker of the House receives significantly more attention than normal from ABC News. This coverage, surprisingly, is not contemporaneous, but happens instead across future quarters. As Table 4 illustrates, the Speaker sees an additional 5 stories in quarter t+1 following an override, an additional 1 story in quarter t+2, and an additional 0.15 stories in quarter t+3. Perhaps the presence of such legislative events slowly orients the media toward reassessing the Speaker’s place in the system of lawmaking; overrides may send a signal to the media that the Speaker is a stronger player worthy of more attention in the future.

Finally, we attempt to use our framework, shock, and legislative outcome-based predictors to explain CNN coverage of the Speaker. In columns 3 and 4 of Table 3, we track the short- and long-run effects on coverage from the 4th quarter of 1995 through the end of 2010. Interestingly, our findings here do not fit as well with the findings from models explaining Speaker coverage in The Washington Post, The Associated Press, and ABC News already discussed. Polarization, change in which party holds the speakership, and bills becoming laws, three variables that at times significantly predicted when Speakers would or would not receive the spotlight from print and network news, do not seem to drive the agenda of cable news during this time period. Previously insignificant predictors like House scandal and divided government situations also fail to shape when CNN covers the Speaker and when CNN does not. What does drive CNN’s coverage of the Speaker of the House is our Speaker scandal variable. It does so only in the long run, and this impact across future quarters is spelled out in the final row of Table 4. We see that scandal involving the Speaker leads to 9 more stories in t+1, 1 more story in t+2, and an impact less than 1 in quarter t+3; this impact declines sharply due to an error correction coefficient of -0.88. This is the third instance (out of 4 models overall) in which ethics investigations involving the Speaker have had long-run impacts on how the media treat the Speaker. Examining this CNN coverage model as a whole, we include these mostly null findings to tease out the idea that, when it comes to the Speaker, not all news sources apparently have the same definition of what is newsworthy. This model illustrates the idea that cable news coverage of Speakers appears to be driven much less by the types of considerations that drove coverage undertaken by sources like network news, newspapers, and news agencies like the AP.

Explaining The Shifting Spotlight: Speaker Coverage Versus Presidential Coverage

We now have a grasp on the types of predictors that explain when the Speaker of the House receives attention from newspapers like The Washington Post, news agencies like The Associated Press, network newscasts like those on ABC, and cable newscasts like those on CNN. But what of the Speaker and the broader media agenda? The Speaker may stand head and shoulders over the average member of Congress in terms of a share of the spotlight, but, as discussed earlier, that share pales in comparison to the share received by the President. Figure 3 illustrates just that.

Figure 3.

Note: x axis = time; y axis = number of Washington Post stories.

Figure 3 tracks presidential coverage and Speaker coverage by The Washington Post between 1977 and 2010.¹⁶ We see quite clearly that during most quarters, Speaker coverage in The Washington Post pales in comparison to presidential coverage. In the average quarter in fact, the President receives 11 times as much attention from the print media as does the Speaker. We also see, however, that the gap between the two is not constant but seemingly changes (and sometimes wildly) from one quarter to the next. In some quarters, this chasm between the spotlight shone on the two leaders is much greater; for example, in the first quarter of 1993, President Clinton received nearly 50 times as much print coverage as Speaker Foley did. Two years later, however, we see a handful of quarters in which President Clinton received less than twice as much coverage as Speaker Gingrich did, a massive shift in the media’s agenda. How well might the predictors discussed so far, the broader framework in which Speakers operate, the shocks to governing in the House, and the legislative outcomes from the House, explain the shifting disparity between consideration on the part of the media paid to the Speaker and to the President? Given our findings so far, we might expect that polarization, change in control, Speaker scandal, and lawmaking might be key.

Table 5 displays the results of an error correction model testing the effects of framework, shock, and legislative outcome predictors on the ratio of Speaker newspaper coverage to presidential newspaper coverage. We see clearly that some of the lessons learned in attempting to explain Speaker coverage in general are also applicable in explaining the extent to which the spotlight on the Speaker shines a bit more or less brightly when compared to the spotlight on the President. We see that polarization is a key predictor of the ratio between Speaker coverage and presidential coverage. The further apart that Democrats and Republicans are in terms of their party position on the DW-Nominate scale, the more the disparity between Speaker coverage and presidential coverage is diminished. If we consider the concept of a 0.1 unit increase in the distance between parties in the House, we see a world in which the ratio of Speaker coverage to presidential coverage increases by 0.16. The effects here are only short-run in nature. Such effects speak to the impacts of polarization discussed earlier; the gap between Speaker coverage and presidential coverage might shrink because of the boost polarization gives to the spotlight on the Speaker. When the House is more polarized, the Speaker must manage this potential and real conflict, elevating the role of the Speaker and putting the Speaker on a more equal footing with the President.

Table 5.

Explaining Speaker Coverage Versus Presidential Coverage, 1977-2010: An Error-Correction Model.

	Long run	Short run
Predictor	Effects	Effects
House polarization	0.009	1.601*
	(0.027)	(0.430)
House/presidency division	0.014	0.037
	(0.009)	(0.024)
House/senate division	−0.008	−0.022
	(0.009)	(.022)
Change in control	0.207*	0.277*
	(0.059)	(0.045)
Speaker scandal	0.011	0.087*
	(0.016)	(0.022)
House scandal	−0.004	0.005
	(0.010)	(0.015)
Bills become laws	−0.0002*	0.0001
	(0.0001)	(0.0001)
Overrides	0.011	0.009
	(0.014)	(0.010)
Story volume_t-1	−0.343*
	(0.061)
Constant		0.027
		(0.026)
N	135
R-Squared	.68

Note: DV = Ratio of speaker newspaper stories to presidential newspaper stories per quarter Ratio series runs from 1977:1 to 2010:4. Standard errors are in parentheses.

= p < .05.

We also find that a shock to the system like change in party control of the House of Representatives has an effect on the ratio of Speaker coverage to presidential coverage. Given the interest on the part of the media in novelty, how new players from different parties will not just affect legislation but also affect the way the game, as Patterson (1994) and Weaver (1972) put it, is played, we should not be surprised that quarters in which new Speakers from parties now in the majority rise to power are quarters in which the ratio between Speaker and presidential coverage might shift. In quarters in which this control changes, the ratio of Speaker coverage to presidential coverage increases by 0.28. Such impacts might be driven by the fact that new party control is an inherently new storyline for the media to follow. Given the timing of our changes in control in this study (1995, 2007), the change in the ratio might also be driven by the contrast of new story lines in the House (dealing with Speakers Gingrich and Pelosi) versus old story lines in the White House (given that both Presidents Clinton and George W. Bush were not new to their offices at the same time). Another possible explanation for the impact of change on this ratio might lie in the weakness of both presidencies during the time in which party change in the House occurred; both changes in control were seen as rebuking the administration and the President’s party to some extent. Speakers representing strong and different new ways of governing took hold of the gavel. Media interest in such changes in comparison to media interest in the presidency might be further heightened as a result. This contemporaneous effect is not the only effect of a change in party control on the ratio of coverage; we also see long-run implications for said ratio. Such long-run implications are displayed in Table 6.

Table 6.

Long-Run Effects on the Speaker-President Coverage Ratio.

Predictor	t+1	t+2	t+3
Change in control	0.21	0.14	0.09
Bills become laws	−0.0002	−0.0001	−0.00007

Here we see that a change in party control of the House affects the ratio of Speaker coverage to presidential coverage in noteworthy ways in at least the year following said change in control. The impact of this change on the ratio of coverage in t+1, t+2, and t+3 is 0.21, 0.14, and 0.09 respectively. The excitement and novelty of a change in leadership (and the anticipation of how said leadership might wield power and react to those in the House, in Congress as a whole, and in government more broadly speaking) makes the Speaker a more competitive player in the race for media attention, and this competition matters not just when the change occurs, but also across future quarters. An error correction coefficient of -0.34 ensures the effect will be somewhat slow to dissipate over time, suggesting small yet significant impact much longer into the future than t+3. New Speakers like Gingrich and Pelosi representing a change in control remained newsworthy in comparison to the President not just when they took office, but for many quarters afterward.

Speaker scandal also lessens the typical gulf between Speaker coverage and presidential coverage. We see in Table 5 that quarters in which Speakers are under ethics investigation like those faced by Speakers Wright and Gingrich are quarters in which the ratio of Speaker coverage to presidential coverage increases by 0.09. This effect of scandal involving the Speaker is solely a contemporaneous one. This once again sends the signal that scandal involving the Speaker drives media narratives. This may be the rare case, however, in which a Speaker does not want his or her profile to rise in comparison to that of the President in the eyes of the media and the minds of the general public. Here the shifting spotlight clearly works against the Speaker; the Speaker may be more of a media rival to the President in these time periods, but for all the wrong reasons and in a way that is fleeting and potentially weakens rather than consolidates power.

Finally, we once more see the lack of coverage-based benefits to turning bills into laws for the Speaker of the House. Table 5 reveals a long-run negative and significant effect of lawmaking on the ratio between Speaker coverage and presidential coverage. The more bills that become laws, the poorer the Speaker fares in comparison to the President in the battle for the media spotlight. Table 6 reveals, however, that the impact of this relationship across future quarters is an incredibly small one. Still, it becomes clearer looking across multiple models that legislative accomplishments in terms of raw numbers do not benefit the Speaker in terms of media attention; in fact, they seem to penalize the Speaker. Turning bills into laws shifts the spotlight away from the Speaker and decreases the Speaker’s ability to rival the President for said spotlight. Once more as in previous models attempting to explain coverage more generally, elements like divided government and broader House scandals have no effect whatsoever on the ratio between Speaker coverage and presidential coverage.

Discussion

What broader lessons might we take away from our examination of media coverage of the Speaker of the House over time? Although no single predictor is significantly related in the exact same way across our Speaker Newspaper (as expressed through the Post and AP), Speaker Network Television, and Speaker Cable Television models, as well as the ratio between Speaker coverage and presidential coverage, we do see several trends across all attempts to explain coverage. The first is that polarization between parties in terms of actual voting often helps predict when the Speaker will and will not receive coverage from the media. Polarization shapes the agenda of Post and AP newspaper coverage, ABC coverage, and the ratio between Speaker and presidential coverage. The broader takeaway point from the significance of polarization is clear: when partisans become even more partisan and different in how they act, the Speaker takes on a greater amount of the spotlight as the figurehead who must manage (and at times even drives) the conflict inherent to parties acting in distinctly different ways. We might argue that Speakers do not necessarily only win coverage like Sellers (2002) discusses in his examination of similar attempts by party leaders, but that the framework helps them win it. The framework elevates the Speaker, forcing the Speaker to manage partisan conflict within the House. This in turn draws in a media consistently interested in conflict and consistently interested in how the political game is being played. Seeing as politicians in Congress are becoming more polarized, the effects of polarization on coverage here have the potential to benefit Speaker efforts to become more public into the future. How leaders like the Speaker of the House work within this framework might go a long way to determining to what extent the increasingly public speakership that Harris (1998) described remains increasingly public. Given the nature of our figures here, we might even want to further analyze how we measure the concept of a public speakership (or at least acknowledge that there are ebbs and flows in coverage of the Speaker in modern times rather than constant upward movement); such a finding fits with a broader theme argued for by authors like Malecha and Reagan (2004). The speakership may be more public in terms of action, but that does not necessarily guarantee a spotlight. The framework in which the Speaker operates conditions the spotlight, as do shocks to the system and outcomes.

We might also conclude, based on our five models in total, that speakership coverage is driven by shocks to the system of governing in the House. Changes in who holds the speakership position driven by a new party taking over power spur coverage in three of our four media models and close the gap between attention afforded the Speaker and the President. Such change brings a new (and historic) narrative to the table. It allows the media to tell a new story. It allows for a discussion of new strategy (and perhaps a comparison to old strategy). It meets and crosses the threshold for what the media deem newsworthy. It is clear that the rise of Speakers Gingrich and Pelosi changed the standard operating procedure of the mass media, and did so in ways that lasted not just for the few months after taking over, but for several subsequent quarters. So too do shocks to the system directly involving the Speaker affect the story lines the media produce. Scandals linked to ethics violations affected media coverage conducted by newspapers, networks, and cable, and also put the Speaker on a more even playing field with the President in terms of the coverage ratio. Scandals fit neatly into broader lessons about the mass media, namely that on the job and/or personal problems involving those in power should be and are fodder for the media to dissect (Sabato, 2000). This dissection is not only immediate, but also continues to develop over time in print and on television.

One last conclusion that becomes clear after an examination of each of our models in total is that, in terms of winning and holding the media’s attention, the creation of new laws does not appear to benefit the Speaker. Across almost every model, the more bills that become laws, the less the Speaker is in the spotlight. Perhaps this is the result of a modern media that does not find lawmaking or the Speaker’s role in lawmaking as interesting as it does the Speaker’s attachment to the conflictual or the novel in the world of politics. Perhaps the Speaker’s media presence grows not from seeing bills become laws, but instead from seeing bills merely passed by the House. On the other hand, perhaps the Speaker himself or herself is not maximizing the potential gains of helping create laws. This might represent a relatively untapped avenue for winning coverage. It is clear that the day-to-day process of legislative success does not seem to benefit the Speaker’s media profile.

More broadly speaking, our findings here also lend further credence to the idea posited by Graber (2010) that media norms drive how politics and politicians are covered. Individuals like the Speaker try to win coverage (and, based on the work of those like Peters, 1990, 1999; Peters & Rosenthal, 2010; Sellers, 2002, 2010; and Sinclair, 2000, we know such efforts are increasing and evolving), but the norms journalists utilize may stand as roadblocks in the path Speakers wish to take to accomplish their media goals and enjoy the personal, policy, and partisan fruits of increased attention. The framework, shocks, and legislating are three such types of roadblocks. Moving forward, the next step in examining what shapes speakership coverage over time might be to determine how we can better explain why some specific individual Speakers are better than others at navigating these roadblocks. Moving beyond looking at single Speakers or single moments in time, might these relationships be explained by things like broader patterns of media entrepreneurship? How might some of these Speakers fare better or worse at handling the media? Is some of this driven by differing personalities? Can differing long-run effects be explained by what Speakers do to hold onto the spotlight once it shines brightly upon them? Elements like these might be easy to discuss anecdotally but difficult to quantify over time. Ideally, subsequent iterations of research of this ilk will begin to inquire into how such elements might be better operationalized. Another path for research in this area might inquire more deeply into why similar patterns emerged between Speaker newspaper and network television coverage but different patterns emerged in examining cable television coverage. In other words, what does explain when CNN, MSNBC, and FOX cover the Speaker if the predictors used here fall short (as they did in the case of CNN)? Other issues related to the findings here that might prove interesting for future research endeavors include how different measures of polarization (such as intraparty differences) might draw the spotlight, how predictors in this model might drive coverage of other powerful players like the Senate Majority Leader, or how the President’s point of view might drive the ratio between levels of coverage across branches, asking how presidential actions lead to a stronger or weaker grip on the spotlight.

Until then though, our predictors linked to the political framework, shocks to the system, and legislative outcomes do an impressive job of explaining widely changing levels of Speaker coverage across multiple decades as well as the ratio between Speaker and presidential coverage. It is clear that to some extent, leaders like the Speaker can control their own media fate and pull attention (and probably additional power) in their direction, but that time in the spotlight is conditioned by factors like polarization, change in control, personal scandal, and lawmaking that, while undoubtedly affected by what Speakers do and how Speakers work, ultimately remain partly out of a Speaker’s grasp.

Footnotes

Appendix

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Tyler Johnson is an Assistant Professor of Political Science at The University of Oklahoma. His research interests lie at the intersection of American political behavior and institutions, using elite activity and norms to explain media coverage and aggregate public opinion.

Caitlin O’Grady is a PhD candidate at The University of Oklahoma who recently served as an American Political Science Association Congressional Fellow. Her research focuses on representation and policies affecting the underprivileged.

References

Althaus

S. L.

(2003). When news norms collide, follow the lead: New evidence for press independence. Political Communication, 20, 381-414.

Arnold

R. D.

(2004). Congress, the press, and political accountability. New York, NY: Russell Sage.

Cohen

J. E.

(2008). The presidency in the era of 24-hour news. Princeton, NJ: Princeton University Press.

Cohen

J. E.

(2009). Going local: Presidential leadership in the post-broadcast age. New York, NY: Cambridge University Press.

Cook

T. E.

(1989). Making laws and making news: Media strategies in the U.S. House of Representatives. Washington, DC: The Brookings Institution.

DeBoef

Keele

(2008). Taking time seriously. American Journal of Political Science, 52, 184-200.

Edwards

G. C.

Wood

B. D.

(1999). Who influences whom? The president, Congress, and the media. American Political Science Review, 93, 327-344.

Fiorina

M. P.

Abrams

S. J.

Pope

J. C.

(2006). Culture war? The myth of a polarized America. New York, NY: Pearson Longman.

Graber

D. A.

(2010). Mass media and American politics (8th ed.). Washington, DC: CQ Press.

10.

Green

M. N.

(2010). The speaker of the House: A study of leadership. New Haven, CT: Yale University Press.

11.

Groeling

(2010). When politicians attack: Party cohesion in the media. New York, NY: Cambridge University Press.

12.

Harris

D. B.

(1998). The rise of the public speakership. Political Science Quarterly, 113, 193-212.

13.

Hess

(1986). The ultimate insiders: U.S. senators in the national media. Washington, DC: The Brookings Institution.

14.

Hess

(1991). Live from Capitol Hill: Studies of Congress and the media. Washington, DC: The Brookings Institution.

15.

Hibbing

J. R.

Theiss-Morse

(1998). The media’s role in public negativity toward Congress: Distinguishing emotional reactions and cognitive evaluations. American Journal of Political Science, 43, 475-498.

16.

Highton

(2002). Bill Clinton, Newt Gingrich, and the 1998 House elections. Public Opinion Quarterly, 66, 1-17.

17.

Jamieson

K. H.

(1988). Packaging the presidency: A history and criticism of presidential campaign advertising. London, UK: Oxford University Press.

18.

Kedrowski

K. M.

(1996). Media entrepreneurs and the media enterprise in the U.S. Congress. Creskill, NJ: Hampton.

19.

Kedrowski

K. M.

Gower

R. E.

(2009, January). Gender and the public speakership: News media coverage of Speaker Nancy Pelosi. Paper presented at the Annual Meeting of the Southern Political Science Association, New Orleans, LA.

20.

Kelly

N. J.

Enns

P. E.

(2010). Inequality and the dynamics of public opinion: The self-reinforcing link between economic inequality and mass preferences. American Journal of Political Science, 54, 855-870.

21.

Kelly

N. J.

(2005). Political choice, public policy, and distributional outcomes. American Journal of Political Science, 49, 865-880.

22.

Kuklinski

J. H.

Sigelman

(1992). When objectivity is not objective: Network television news coverage of U.S. senators and the “paradox of objectivity.” Journal of Politics, 54, 810-833.

23.

Lichter

S. R.

Amundson

D. R.

(1994). Less news is worse news: Television coverage of Congress, 1972-92. In Mann

T. E.

Ornstein

N. J.

(Eds.), Congress, the press, and the public (pp. 131-140). Washington, DC: Brookings Institution.

24.

Malecha

G. L.

Reagan

D. J.

(2004). Coverage of the postreform House majority party leadership: An expanding or a shrinking image. Congress & the Presidency, 31, 53-76.

25.

Morris

J. S.

Clawson

R. A.

(2005). Media coverage of Congress in the 1990s: Scandals, personalities, and the prevalence of policy and process. Political Communication, 22, 297-313.

26.

Patterson

S. C.

Caldeira

G. A.

(1990). Standing up for Congress: Variations in public esteem since the 1960s. Legislative Studies Quarterly, 15, 25-47.

27.

Patterson

T. E.

(1994). Out of order. New York, NY: Knopf.

28.

Patterson

T. E.

(2000). Doing well and doing good: How soft news and critical journalism are shrinking the news audience and weakening democracy and what news outlets can do about it. Cambridge, MA: Joan Shorenstein Center on the Press, Politics, and Public Policy.

29.

Peters

R. M.

(1990). The American speakership: The office in historical perspective. Baltimore, MD: The Johns Hopkins University Press.

30.

Peters

R. M.

(1999). Institutional context and leadership style: The case of Newt Gingrich. In Rae

N. C.

Campbell

C. C.

(Eds.), New majority or old minority? The impact of Republicans on Congress (pp. 43-68). Lanham, MD: Rowman & Littlefield.

31.

Peters

R. M.

Rosenthal

C. S.

(2010). Speaker Nancy Pelosi and the new American politics. New York, NY: Oxford University Press.

32.

Robinson

M. J.

Appel

K. R.

(1979). Network news coverage of Congress. Political Science Quarterly, 94, 407-418.

33.

Rohde

D. W.

(1991). Parties and leaders in the postreform House. Chicago, IL: University of Chicago Press.

34.

Rozell

M. J.

(1994). Press coverage of Congress, 1946-92. In Mann

T. E.

Ornstein

N. J.

(Eds.), Congress, the press, and the public (pp. 59-129). Washington, DC: The Brookings Institution.

35.

Rozell

M. J.

(1996). In contempt of Congress: Postwar press coverage on Capitol Hill. Westport, CT: Praeger.

36.

Sabato

L. J.

(2000). Peepshow: Media and politics in an age of scandal. New York, NY: Rowman & Littlefield.

37.

Sellers

P. J.

(2002). Winning media coverage in the U.S. congress. In Oppenheimer

B. I.

(Ed.), U.S. Senate exceptionalism (pp. 132-154). Columbus, OH: The Ohio State University Press.

38.

Sellers

P. J.

(2010). Cycles of spin: Strategic communication in the U.S. Congress. New York, NY: Cambridge University Press.

39.

Sigal

L. V.

(1986). Sources make the news. In Manoff

R.K.

Schudson

(Eds.), Reading the news: A pantheon guide to popular culture (pp. 9-37). New York, NY: Pantheon Books.

40.

Sinclair

(1998). Legislators, leaders, and lawmaking: The U.S. House of Representatives in the postreform era. Baltimore, MD: Johns Hopkins University Press.

41.

Sinclair

(2000). Republican House majority party leadership in the 104th and 105th Congresses: Innovation and continuity. In Bianco

W. T.

(Ed.), Congress on display, Congress at work (pp. 98-112). Ann Arbor, MI: University of Michigan Press.

42.

Squire

(1988). Who gets national news coverage in the U.S. Senate? American Politics Research, 16, 139-156.

43.

Tidmarch

C. M.

Pitney

J. J.

(1985). Covering Congress. Polity, 17, 463-483.

44.

Weaver

(1972). Is television news biased? Public Interest, 27, 69.