Pulling the Plug on Grandma: Obama’s Health Care Pitch, Media Coverage & Public Opinion

Shahira Fahmy, Christopher J. McKinley, Christine R. Filer, Paul J. Wright School of Journalism, College of Social and Behavioral Sciences, University of Arizona, Tucson, USA Department of Communication Studies, Montclair State University, Montclair, USA Department of Communication, College of Social and Behavioral Sciences, University of Arizona, Tucson, USA Department of Telecommunications, Indiana University, Bloomington, USA Email: sfahmy@email.arizona.edu


Pulling the Plug on Grandma: Obama's Health Care Pitch, Media Coverage & Public Opinion
With growing unease over the trillions of dollars being spent on bailouts and economic stimulus on top of an already unprecedented level of debt, President Obama signed into law a historic health care overhaul that according to him "won't pull the plug on grandma."In March 2010, the President signed The Patient Protection and Affordable Care Act (PPACA).Controversies over the Obama's health care reform emerged in public discourse however.The reform faced near unanimous opposetion from Republicans and a serious divide among Democrats and the American public.The failure among Democrats to support Obamacare, intense Republican attacks against the issue, and sagging poll numbers pushed the President to address Congress specifically to fix the nation's ailing health care system.
As the President pitched to sell the country on the need to reform health care, coverage of the debate became widespread.Media outlets often had a tendency to frame the issue in terms of economic controversy, such as framing the issue as an economic necessity or as a bad economic policy.As a result, the way the President and the media framed the debate may have influenced the way the public understood, evaluated, and supported Obamacare.
This study uses the signing of the PPACA law to explore how interpretive frames activate and spread from public officials through the news media to the public (Entman, 2003).While numerous studies have examined who sets the media agenda (Gandy, 1982;Wanta, Stephenson, Turk, & McCombs, 1989) less attention has been paid to investigating the influence of important news sources on health care reform.In fact, regarding issues where health care and politics intersect, research examining agenda-building of health care reform appears to be scarce.Guided by agenda-building research, the current research focuses on the health care debate by examining the relationship among media coverage, presidential rhetoric, and public opinion starting with President Obama's inauguration in January 2009 to the date the PPACA was signed into law in March 2010.

The Health Care Debate
Obama placed health care reform on his agenda long before entering the White House.One month before Election Day, he chose to give a speech in Virginia focusing on health care reform. 1 He stated, "The real solution is to take on drug and insurance companies; modernize our health care system for the twenty-first century; reduce costs for families and businesses; and finally provide affordable, accessible health care for every American.And that's what I intend to do as President of the United States" (Obama, 2008).It should be noted that after winning the 2008 presidential election, Obama continued to cite health care reform as a priority (Obama, 2009).
Over a year of debate followed Obama's inauguration, as reforming health care proved to be an arduous process.Government officials discussed not only what should be included in the health care plan, but also whether the federal government had the power to require that each citizen purchase health care.
Extensive media coverage followed the debate.Rather than attempt to explain what the proposed health care reform plans might mean for the average citizen, most news coverage focused on controversial issues such as death panels and the economic implications of overhauling health care.
With proposed plans costing anywhere from $750,000 to $1 trillion, much of the debate concerned the national debt and the financial effects of an expensive health care bill.At the same time, others argued for the economic benefits of extending health insurance coverage to millions of people.Some of the narrower financial issues discussed included cutting medicare, creating a public option plan, and making deals with pharmaceutical companies (Holan, 2009;Klein, 2009, Health insurance reform and medicare: Making medicare stronger for America's seniors, 2012).
Coverage of the debate was also fueled by bipartisan divide.The media publicized intergovernmental disagreement over health care reform legislation along party lines, as well as rare instances of Republicans supporting the legislation and Democrats opposing it.Experts on health care, insurance industry executives, and task groups assigned to drafting legislation were cited in the news as they provided their opinions on not only the proposed plans but also the process of passing health care reform.Obama was often referenced throughout the debate, both championing support for legislation and recognizing obstacles faced in passing it.
On November 7, 2009, the House of Representatives passed H.R. 3962, colloquially referred to as the Affordable Health Care for America Act.

Theoretical Framework: Agenda Building
This study examines the agenda-building framework from the perspective of the reciprocal relationship among presidential rhetoric, media coverage, and public opinion (Lang & Lang, 1983).It approaches the Obamacare debate as a form of political discourse controlled by competing perspectives toward the issue by examining the interaction among these three different variables through which different agendas are discerned (Fahmy, Wanta, Johnson & Zhang, 2011;Johnson & Wanta, 1996;Lang & Lang, 1983;Wanta & Kalyango, 2007).
A review of the literature indicates other agenda-building studies have focused on this cyclical process involving a similar reciprocal agenda-building relationship.For example, Lang and Lang (1983) examined the relationship between the press and public opinion during the Watergate era.They found a cyclical three-way relationship among the press, the public, and the presidency, suggesting that more complicated issues go through the process of agenda-building.Johnson and Wanta (1996) examined the relationship among the public, the media, and the Nixon administration regarding the war on drugs.They found that real-world events set into motion the agenda-building process.This drove news media to increase their coverage of the issue that, in turn, led to the public learning of the importance of drugs as a major issue.Finally, the President reacted to public's concern.
Recently, Fahmy and colleagues (2011) examined the interaction among the President, the media, and the public for an event that was not considered an existing "real-world" condition.They found evidence that President Bush influenced media coverage of the Iraq War, supporting the notion that news values and journalistic norms have traditionally placed a priority on gathering information from authoritative and official sources (Ragas, 2012).
These studies and others adopting the cyclical process approach of agenda-building have traditionally been guided by the use of public opinion polls and content analyses of media frames as well as frames used by the presidency (i.e.analyzing weekly compilation of presidential documents) (Fahmy, Wanta, Johnson, & Zhang, 2011;Wanta & Kalyango, 2007).
Framing is the process of selecting, emphasizing, and interpreting of a situation to promote a particular interpretation of an issue or event (Entman, 1993).Using content analysis to look at framing in media coverage and presidential rhetoric has been validated across studies.In the case of presidential rhetoric, numerous studies have indicated that the U.S. President is the nation's first news source that gets cited regularly by the news media (Adams & Cozma, 2012;Fahmy, Relly, & Wanta, 2010;Fahmy, Wanta, Johnson, & Zhang, 2011;Wanta & Foote, 1994).Regarding content analysis, Borah (2011) found that the majority of framing research in the past decade consisted of content analyzing media coverage.
Therefore, understanding the use of framing by the news media and the president, offers us an understanding of how competing views of Obamacare might have used the media to build a frame of coverage in ways that would create support for their cause.For example, Conway (2012) examined secondlevel agenda setting effects of six news outlets on public opinion about the health care reform bill proposed by President Obama.Her analysis revealed that cumulative affective attribute salience in the media was a significant predictor of support in public opinion polls (Adams & Cozma, 2012).
Within the context of this study, government sources (including President Obama) could have carefully chosen strategies to build the agenda for public discourse regarding Obamacare over time.This would have allowed the diffusion of preferred frames and policies to dominate the U.S. media, and the mobilization of the public toward support for health care reform advocated by the White House.The impact would have been substantial simply because how an issue is framed by public officials and the media comprise the principle arena within which controversial issues come to the attention of policy makers and the public.For example, Adams and Cozma (2012) found that Obama was the top news source in newspaper stories covering the health care reform.While this study offered an in-depth investigation of sources and specific types of media frames addressing health care reform, it did not examine tone of coverage, presidential documents and/or compare them to public opinion.Thus, it remains unclear whether this particular issue supports the traditional perspective on agenda-building.
Furthermore, historically within the processes of framebuilding and agenda-building, competing sources have operated as news sources to provide strategic information to the media (Lambert & Wu, 2011;Wu & Lambert, 2010).For example, Wu and Lambert (2010) found that government sources were relied on the most in news coverage of the health care debate, and that public opinion toward Republicans and Democrats trended downward, a finding traced to these sources being used more frequently in media.While that study in many ways mirrors this investigation, the current study applies a more com-S.FAHMY ET AL.
prehensive assessment of the links among presidential rhetoric, media coverage, and public opinion.It adapts from Robert Entman's (2003) cascading activation model that explains how interpretive frames activate and spread from the top level of a stratified system (the White House) to the network of non administration elites, and on to news organizations, their texts, and the public.Thus, by focusing on tracing the diffusion of frames by the White House (President Obama) and media coverage, the current study hopes to add to agenda-building literature by exploring how interpretive frames might have activated and spread from the top level in the context of one of the most controversial national health care policies.

Method
The news sources sampled in this study included news articles and opinion pieces published between January 20, 2009 and March 23, 2010 in The New York Times, LA Times, Chicago Tribune, and The Washington Post.These newspapers were selected based on their influence and prestige.The literature suggests national coverage of news tends to follow the elite media (Bennett, 1990;Fahmy, Wanta, Johnson, & Zhang, 2011;Nisbet & Lewenstein, 2002).Therefore, as newspapers of record (that have knowledgeable news professionals) these four elite major metropolitan publications tend to set the news agenda for regional news organizations.
The time frame analyzed represents Obama's inauguration (January 20, 2009) and the date the PPACA was signed into law.Coverage containing both "Health Care" and "Obama" anywhere in the text was retained (N = 2856: The Chicago Tribune = 1146; The Washington Post = 746; The New York Times = 743; The LA Times = 221).Fifteen percent of the articles including these terms and relating to health care reform were then randomly selected from each paper's pool, resulting in a final sample size of 428 articles.Of the 428 articles sampled, 40.2 percent (n = 172) were selected from the Chicago Tribune, 26.2 percent (n = 112) from The Washington Post, 25.9% (n = 111) from The New York Times, and 7.7 percent (n = 33) from the LA Times.
Weekly presidential compilation documents were also examined.The documents were downloaded from the U.S. Government Printing Office's Federal Digital System, which has electronic versions of the White House Press Secretary's official publications available at http://www.gpo.gov/fdsys/browse/collection.action?collectionC ode=CPD.To be included in our analysis, each document had to be published between inauguration and bill passage and contain both "Health Care" and "Obama".These initial requirements yielded 26 documents.Of these documents the ones that did not mention Obamacare or health care reform were excluded from the sample, leaving 17 documents to be analyzed.2

Coding Categories
Each article was coded for the content categories described below, while each presidential document was only coded for focus and tone.The weekly presidential compilation documents were not coded for source frequency or dominant source, as they were comprised of statements by Obama and transcripts from Obama's speeches and town halls.
Focus.Each article's/document's primary focus was assessed by its lead.To ensure that the dominant focus was captured, the expanded definition of a news lead that includes up to three paragraphs was used (Hillback, Dudo, Wijaya, Dunwood, & Brossard, 2008).Possible foci included the ethics of health care reform (i.e., emphasis on religious perspectives, moral views, the obligations of a civilized society), political/policy implications (i.e., laws, regulations, partisanship), economic implications (i.e., cost to private companies, government, individuals), the public view of the reform(i.e., polling data, public support, public concerns), and humanistic concern about the reform (i.e., an individual's narrative; the fate of a particular group in need).Themes that fell outside of these parameters were given an alternative synopsizing moniker.
Tone.The overall valence of each article and document was determined by coding each paragraph as "positive," "negative", or "neutral" towards Obama's health care reform plan and then assigning a single summary valence code based on the most frequently occurring tone.Articles/opinion pieces and documents with an equal number of positive and negative paragraphs were coded "neutral".To provide a few examples, a paragraph would have been coded positive if it emphasized "good news" for Obama's plan (i.e., a report indicating that the plan would alleviate participants' health costs), reported that particular noteworthy individuals favored the plan, or in the case of an editorial, if the writer openly advocated for the plan.A paragraph would have been coded negative if it exhibited themes converse to these.Paragraphs coded neutral either lacked valence or featured an equal number of positive and negative statements.
Source frequency.Sources were coded each time they were referenced in the articles.Citations could have been cumulative both within and between sources.For example, an article sourcing President Obama four times would be allotted a "4" code for the category President Obama.An article citing three difference private physicians a single time each would be allotted a "3" code for the category private physicians.Possible sources included physicians in private practice, government or academic physicians, economists, academics, political figures (i.e., secretary of state, speaker of the house, senate majority leader), government bureaucrats (i.e., behind the scenes government workers), tea party members, President Obama, insurance company officials, pharmaceutical company officials, nonprofits (i.e., advocacy groups, lobbyists, religious figures), citizens, international sources (including politicians or scientists abroad), and anonymous.Sources that fell outside of these parameters were given an alternative synopsizing moniker.
Source categories were then combined for each article into "elite" and "non-elite" based on previous research (Fahmy, Wanta, Johnson, & Zhang, 2011): (1) physicians in private practice; (2) government or academic physicians; (3) economists; (4) political figures; (5) academics; and (6) President Obama were categorized as "elite".All other categories were coded as "non-elite".Dominant source.Each source category was tabulated every time it was referenced in an article.A "dominant source" code was awarded to the category mentioned most frequently for each article or opinion piece.

Coding Reliability
keen degree of synchronicity was achieved.Ten percent of articles were ultimately selected at random for a formal reliability assessment.Reliability was calculated using the reformulated Pi equation (Potter & Levine-Donnerstein, 1999).Results indicated favorable reliability for all variables: source frequency = 0.92; tone = 0.87, dominant source = 0.86, focus = 0.76.Reliability was then calculated for the presidential documents.
Coder training and practice coding were carried out before 11.5 percent of documents were randomly selected for a formal reliability assessment.Reliability measures were strong for both variables: focus = 1.00, tone = 1.00.

Public Opinion
Public opinion data gathered between inauguration and bill passage regarding health care reform were obtained.Because different polls were taken on the same date and time periods for different polls overlapped (i.e., Gallup poll from 2/6 -2/8, ABC news from 2/5 -2/8), Rasmussen, which conducted the most polls during this time period, was the only public opinion data examined in this analysis.Rasmussen provided data on the percentage of Americans who, at the time the poll was conducted, either favored or opposed Obamacare.The earliest poll employed was conducted between 10 and 11 July, 2009.The latest poll employed was conducted between 5 and 6 March, 2010.

Proposed Analyses
To explore the agenda-building process, this study used frequency tests to assess percentage differences across categories of dominant frame, valence (tone), source categories, over-time changes in valence and over-time changes in public opinion.In addition, inferential analyses using independent samples t-tests, Pearson correlation tests, and simple regression models were used as post-hoc assessments of source categories.It is important to note that the post-hoc simple regression tests were solely employed for exploratory purposes.Specifically, by using the Rasmussen polling data (as will be subsequently discussed), the researchers could not control for key outside factors, including demographics and media use, that may impact the fluctuations in the main outcome variable (favorability ratings).Thus, simple regression analysis was only run to address the possibility that a relationship may exist between these variables.

Results
Roughly two-thirds (64.5%, n = 276) were coded as news, while roughly 35 percent (35.5%, n = 152) were coded as opinion.Regarding dominant news frame, politics/policy implications of health care reform was by far the most common (69.9%, n = 299).The second most frequent dominant news frame fell under the 'economics/financial' category (15.4%, n = 66).The remaining portion of the sample-roughly 16 percent of the total-represented mostly the public view concerning health care reform (11.4%, n = 49).Less than 4 percent of the articles on health care reform focused on humanistic concerns (n = 10) or ethics (n = 4).It is important to note that when examining every sample month individually, politics/policy was found to be the dominant theme in at least 50% of all news articles discussing the health care reform bill.In fact, in 7 of the 15 months sampled, the politics/policy theme was the dominant focus of 75% or more of all articles.However, while there did not appear to be substantial variability in the focus of articles across the sample period, from April to June 2009, the economics tied to health care did become a substantially more prevalent theme in news articles, reflecting 42.9% of dominant focus in April and June, respectively.
Of the 17 presidential documents dealing the health care reform act, the dominant frame used by the President also centered on the politics/policy implications of health care reform (n = 9).No other dominant frame was used more than two times in the president's speeches on health care reform during the time period analyzed.
The valence of the articles was fairly evenly split between positive and negatively-toned articles, although a slighter greater number of articles addressing Obama's health care plan had a positive tone (46.0%, n = 197) compared to those that were negative (41.1%, n = 176).Less than 13 percent of articles were neutral in tone (12.9%, n = 55).While there appeared to be little difference in the overall percentage of positively versus negatively toned articles, tone varied more substantially based on article type.In particular, results showed that 56.5 percent of news articles had a positive tone compared to only 27 percent of opinion articles, whereas 58.6 percent of opinion articles had a negative tone compared to only 31.5 percent of news articles.In addition, cross-tabulations were run to assess whether the percentage of negatively vs. positively-toned articles varied based on focus of article.Given that the dominant focus of the majority of articles (414 out of 428) comprised the three categories representing politics/policy, economics, and public view, these were the only categories examined for analysis.Results showed that there was an association between dominant focus and tone, χ 2 (4, N = 414) = 17.33, p < 0.01).Interestingly, among articles with the dominant focus being politics/policy, fully 48.2%(n = 144) had a positive tone, whereas 36.5% (n = 109) were negative in tone.Conversely, among articles with the dominant focus being the public view of health care reform, only 32.7% (n = 16) were positive in tone, while 63.3% (n = 31) had a negative tone.Roughly equivalent percentages of positive (45.5%; n = 30) and negative (48.5%, n = 32) toned articles were present when the dominant focus was economics of health care.
The next analyses involved the dominant sources for news stories on health care reform.Initial tests showed that nearly 19 percent of articles (18.9%, n = 81) had no clear dominant source.Furthermore, an additional 7.2 percent of articles (n = 31) had multiple dominant sources.Thus, articles that had neither dominant source nor multiple dominant sources were removed prior to subsequent analyses involving dominant source.Results showed that more than two-thirds of the dominant sources for the remaining articles (n = 316) were elites (69.3%).Subsequent analyses comparing frequencies and percentages of dominant sources across article tone showed that elite source categories were slightly more positive in tone (51.1%) than non-elite source categories (45.4%).
In addition, it is important to note that in less than 12 percent of articles was President Obama either the only dominant source or one of the major sources (11.5%).However, references to Obama varied based on article type.In particular, the President was cited significantly more frequently [t (368.08)= 4.56, p < 0.01] in news articles (M = 1.37) than opinion articles (M = 0.43).In the 41 articles where President Obama was the sole dominant source, 24 of these had a positive tone whereas only 13 had a negative tone.Furthermore, results showed there was a significant positive correlation between the frequency that Obama was cited in an article and the tone of the article (measured on a 1 -3 scale, with 1 = negative, 2 = neutral, 3 = positive): r (428) = 0.18, p < 0.01.Overall, the more frequently President Obama was cited in an article, on average, the more positive the tone of the article.

Over-Time Changes in Article Tone, Public Opinion
Tests were done to examine whether any changes occurred in the tone of articles discussing Obama's health care plan between his inauguration and the date the PPACA was signed into law.Tables 1 and 2 break down percentage of positive and negative tone articles in one and two month time frames.The percentage of positively toned articles appears to have peaked between March and May 2009, with roughly 70 percent of all articles discussing Obama's health care plan reflecting a more positive tone (see Table 2).In particular, nearly 80 percent of articles from late April through late May had a positive tone.However, following this time period, the percentage of positively toned articles diminished substantially.When examined on a one-by-month basis, in only 2 of the final 10 months of the time frame did the percentage of positively toned articles exceed 50 percent (see Table 1).Conversely, the percentage of negatively toned articles was equivalent or higher than the percentage of positively toned articles for 7 of the final 10 months of this time frame.
Interestingly, the percentage of positively toned media articles across the sample time period was found to be significantly correlated with the number of presidential speeches given on this topic (r = 0.76, p < 0.05).This indicates that the more speeches President Obama gave on the health care act in any given month corresponded with an increase in the positive tone of articles discussing this bill.
When examining public opinion, average favorability ratings for Obama's health care plan across all polls during this time  frame was slightly greater than 40 percent (M = 41.93).The percentage favorability for the health care reform bill ranged from 25 percent (8/11-13/2009) to 53 percent (3/1/2010).However, when examining month-by-month averages across polls, there was little variability.In particular, from the first month where multiple polls were taken (July 2009) until the final month of the sample (March 2010), average percentage favorability ranged from a high of roughly 44 percent (M = 44.4) to a low of roughly 40 percent (M = 40.73).Additional analyses were performed comparing article tone across the study time period with public support for Obama's health care plan.Results from these analyses showed that media tone at time 1 significantly predicted favorability ratings at time 2 (β = 0.76, p < 0.05).Table 3 provides the comparison of percent articles with positive tone during a specific time period with the earliest favorability rating taken in the following month.Although there was little overall variability in favorability ratings across the time period, it is interesting that in the two months where the least percentage of articles were positive in tone (November, 2009, January, 2010), subsequent favorability ratings in the next month's poll were also the lowest for the entire sample period (see Table 3).However, while the percentage of positively-toned news articles in November, 2009 (29%) and January 2010 (24%) were substantially less positive than the tone of the respective preceding months [October, 2009 (54%); December, 2010 (40%)], this did not correspond with a similar dramatic drop in favorability across the same time period.Specifically, favorability only dropped four percent between the first November poll and the first December poll, and only three percent between the first January poll and the first February poll.
The final analyses examined the same time period listed in Table 3 to assess whether frequency of presidential statements predicted subsequent favorability ratings.Results of simple regretssion analysis showed that frequency of presidential statements in time 1 (i.e., August) was a significant predictor of favorability ratings in time 2 (i.e., September), β = 0.72, p < 0.05.Furthermore, it is important to note that the poll taken in February 2010, which showed the lowest favorability ratings across the entire sample period, was conducted following the only month where President Obama did not give a single speech discussing the health care act.

Discussion
This study makes several meaningful contributions.While two previous studies have employed the agenda-building framework to examine how presidential rhetoric and elite sourcing may contribute to the news media and publics' perception toward controversial topics such as stem cell research (Fahmy, Relly, & Wanta, 2010) and war (Fahmy, Wanta, Johnson, & Zhang, 2011), this is one of the first studies to apply a comprehensive approach of agenda-building in the context of Obama's health care reform (Lambert & Wu, 2011;Wu & Lambert, 2010).Adapting from Robert Entman's (2003) cascading activation model, this study traced the diffusion of frames by the President and the network of elites through the media to the public, and thus added to the body of agenda-building literature by examining how interpretive frames regarding a health care issue activated and spread in the context of a domestic and controversial public health debate.
This study also aids in explication of agenda-building in the health care sphere.In exploring the connections among elite sourcing (i.e.President Obama), tone of coverage, and public opinion, results of this study indicate that there was, at best, only a modest link between President Obama's stance on health care reform and the news coverage of this issue-contradicting findings that suggest the President was the top source of news in the coverage of the health care reform debate (Adams & Cozma, 2012).Consistent with other recent studies involving this issue (Wu & Lambert, 2010), our findings showed more than two-thirds of dominant sources cited were elites; however, in less than 12 percent of the articles analyzed in this study was President Obama either the only dominant source or one of the major sources.Particularly, while the President continued to campaign for his health care plan in 2009-2010, nearly half of all articles sampled emphasized the negative aspects of the plan.In addition, the percentage of negatively-toned articles continued to increase leading up to passage of the bill.That said, the percentage of positively toned articles was significantly correlated with the number of presidential speeches discussing the bill.In other words, an increase in presidential speeches on the topic was associated with a similar increase in positive articles during this time period.However, this only indicates that there is a relationship between these factors-it does not show that frequency of presidential speeches led to an increase in the percentage of positively-toned articles.In fact, results of post-hoc analyses showed that there was no significant association between the frequency of presidential speeches given in one month (i.e., August), and the percentage of positively-toned articles in the following month (i.e., September).
Although frequency of positively-toned articles did significantly predict public opinion in the following month, a closer inspection of the data showed that, surprisingly, public opinion toward health care reform appeared to be mostly unaffected by media coverage.Specifically, average support for the health care bill from June, 2009 through March, 2010 ranged between 40-44 percent.Even following months where articles' tone became substantially less positive, public support did not seem to change significantly, only dropping a few percentage points.The lack of variability in public support across the sample period may be linked to how the news media chose to frame this issue.Recall that the majority of articles framed this issue as politics or policy.This precise frame may be of less interest to the public because it may not specifically address how the reform would affect them.In particular, within stories dealing with the politics surrounding this issue, many articles likely focused on health care reform as a 'contest/game' or stressed elements of conflict for dramatic effect (Adams & Cozma, 2012).Public opinion data taken around this time frame showed that a large percentage of Americans perceived the debate over health care to be negative in tone and disrespectful (Pew Research Center for the People & the Press, 2009).Therefore, by focusing attention on this debate, news media outlets may have turned off readers.Furthermore, those articles that paid more attention to the policy implications of health care reform may have failed to engage readers (McManus, 1992).
Conversely, if more articles focused on the economics of health care, humanistic concerns, or general support from U.S. citizens, this may have resonated more with the public, and therefore, led to greater fluctuations in public favorability.Future experimental research should examine the ways in which different news frames concerning health care influence public support.Furthermore, given how media are increasingly becoming more fragmented, it is possible that the public is turning to more diverse outlets to gain information on important social topics.Thus, the influence of any one media outlet on public opinion may be diluted by access to more information sources.
The modest percentage of positively-toned articles can at least partially be linked to results indicating that news stories covering the health care bill rarely relied on President Obama as a main information source.Moreover, the less frequently Obama was cited in articles, the more negative the tone of the coverage.These findings contradict assumptions made by prior researchers (Wanta & Foote, 1994), which posit that journalistic values allow presidents to dominate news frames.Overall, this suggests that within the context of health care reform, the media were modestly successful at building the media agenda.
However, results also indicated that presidential rhetoric might have influenced public opinion.Specifically, findings showed the frequency of Obama's speeches regarding the bill in time 1 significantly predicted subsequent favorability ratings in time 2. Given that there were only 17 presidential speeches dedicated to this topic from June, 2009 through March 2010, these results suggest that there may have been greater support for the bill during this timeframe if a larger number of the president's speeches were focused on this issue.

Limitations and Areas for Future Research
As noted in the methods section, a key limitation of the analyses was the inability to control for key factors that may explain fluctuations in public support for health care reform.More specifically, to examine over-time support and its relationship with presidential speeches and tone of news stories, this study used data from Rasmussen, one of the leading polling companies as well as the source that conducted the most polls during this time period.Essentially, the simple regression analyses were performed at an aggregated, macro-level to examine associations between percentages.Unfortunately, by using Rasmussen data, it was not possible to assess demographics, political beliefs, media use habits, or any other attributes that may factor into individual attitudes toward this issue.Future researchers should conduct more individual-level analyses of personal characteristics to control for these factors and ultimately provide a clearer assessment of how presidential communication and news coverage influence public opinion of health care.
Further research should continue to explore the role of agenda building regarding controversial issues and how this role might differ between domestic versus foreign policy debates.Another fruitful area of research would be to explore the controversy after passing the bill.Since so much of the subsequent coverage appeared to be negative, perhaps as a result, public opinion turned against it.Furthermore, because the Republican Party strongly opposed the bill from its inception it would be interesting to examine the effects the Republican Party had on framing the debate in the public sphere.Finally, the authors acknowledge that this study did not include the analysis of online media.Specifically the role of social media-sparked by Facebook, Twitter, YouTube and other online outlets-in agendabuilding appears to be a further area of research.It would be interesting for example to look at how competing sources operate online to provide strategic information and influence the way interpretive frames activate and spread to the public.
The Senate then passed H.R. 3590, or the PPACA, on December 24, 2009.Building upon the PPACA, Obama released his proposal on February 22, 2010.The PPACA (H.R. 3590) passed in the House on March 21 and was signed into law by Obama on March 23, 2010.

Table 1 .
One month time frame for percentage of positive and negative tone articles.

Table 2 .
Two month time frame for percentage of positive and negative tone articles.